2. SOILS AND TOPOGRAPHY

Topography and geology
Soil types

3. CLIMATE AND AGRO-ECOLOGICAL ZONES

Climate
Wind
Air temperature
Rainfall
Snow
Sunshine
Geographical regions
Agro-ecological regions
Forests

4. RUMINANT LIVESTOCK PRODUCTION SYSTEMS

Recent history and context
The livestock sector
Dairy production
Beef production
Sheep and other ruminants
Organic farming and livestock
Veterinary and animal health problems
Animal diseases and nutritional disorders

5. THE PASTURE RESOURCE

Crop and grassland production
Forage production systems
Pastures and forage crops
Description of the main permanent grassland and rangeland types
Temporary pastures
Legumes in pasture swards
Pasture management and forage conservation
Temporary or permanent pastures
Fertilization
Production
Grazing systems
Forage conservation
Herbage seed production
Integration of forage resource utilization with environmental objectives and food quality issues
Environmental objectives
Food quality

6. OPPORTUNITIES FOR IMPROVEMENT OF PASTURE RESOURCES

7. RESEARCH AND DEVELOPMENT ORGANIZATIONS

8. REFERENCES

9. CONTACTS AND ACKNOWLEDGEMENTS

1. INTRODUCTION

Country location
Belgium is in the northwest of Europe, between 51° 30' and 49° 30' N, and 2° 33' and 6° 24' E. It borders the Netherlands to the north, Germany and Luxemburg to the east, France to the south and the North Sea to the west (Figure 1). It has a land area of 30 528 km² and is traversed by two main rivers, flowing north-south: the Scheldt and the Meuse. Both rise in France and flow into the North Sea in the Netherlands. For administrative purposes, the country is divided into three regions, ten provinces and 589 municipalities (Figure 2).

Figure 1. Map of Belgium showing the main towns and borders with surrounding countries and the North Sea coast Source: World factbook, 2009.

Figure 2. Administrative map Source: IGN, 2007. [Click to view full map]

Language, communities, government, administration and economy. Belgium is a constitutional monarchy, established in 1830. Its structure and functions have been changed by a series of constitutional reforms (in 1970, 1980, 1988-­89, 1993 and 2001) into a federal State of three regions, the Flemish, the Walloon and Brussels-Capital which have authority over socio-economic matters such as economy, environment and agriculture. The Federal Government is responsible for topics such as justice, social security, fiscal affairs and national defence.

Dutch, French and German are the three official languages. Dutch is spoken by about 6 000 000 Flemish people, mainly in Flanders. French is spoken by about 4 000 000 mainly in Wallonie and in Brussels. German is used by a community of about 74 000 (in 2007) (FPS Economy, 2008), in the east of the Walloon Region. The Flemish and the Walloon Regions are officially unilingual and the Brussels-Capital Region is officially bilingual. The vast majority of the inhabitants of Brussels speak French (Belgian Federal Government, 2008).

The Belgian State is a pioneer member of the European Union. With the Netherlands and Luxemburg, it created Benelux in 1944, a core number of States that coordinated several aspects of their socio-economic policies before the creation of the European Economic Community (EEC) in 1957. Its currency is the euro.

Belgium had a gross domestic product (GDP) in 2007 of €319 billion (FPS Economy, 2008), which is supported by small and medium-sized enterprises (SMEs). Approximately 83 percent of Belgian companies have less than ten employees and 97 percent of the companies employ less than 50 people. SMEs account for over 70 percent of the GDP (FPS Foreign Affairs, Foreign Trade and Development Cooperation, 2006).

Population. In January 2008, the population comprised 10 666 866 people (Flanders 6 161 600 , Wallonie 3 456 775 and Brussels 1 048 491) The population increased at a rate of 0.34 percent between 1960 and 2008, and was higher after 1989. Three-quarters of this increase is explained by the migration balance and one-quarter by natural growth (FPS Economy – Directorate-General Statistics and Economic Information [DGSEI] Press Release, 28 August 2008). Belgium is one of the most densely populated countries (about 350 inhabitants/km² in 2007) in the world (FPS Economy, 2008). However, the Flemish Region is twice as populous as the Walloon Region. The lowest density is in the Province of Luxemburg (about 60 inhabitants/km²) (Table 1). [Editors note : according to the World Factbook the population estimate for July 2009 was 10,414,336 with an estimated growth rate of 0.094%].

Belgium is very urbanized, with 15 urban areas of more than 80 000 inhabitants, comprising about 53 percent of the total population (FPS Economy – DGSEI, 2009).

Land area and land structure. Agriculture occupies almost 60 percent of the land, woodland about 20 percent and urban areas, including residential, industrial and green areas as well as roads and railways, about 20 percent (Table 2). There has been a continuous decrease in the agricultural area (AA) and increase of urban areas since the nineteenth century, when many semi-natural areas were planted with exotic trees (Picea spp., Pinus spp., Populus spp.). The permanent grassland area increased considerably in the second half of the nineteenth century, but decreased slightly after the 1970s. In 2007, it represented 37 percent of the AA while arable land area accounted for about 61 percent. Temporary and permanent grasslands occupied 43 percent of the AA and annual crops 55 percent. The area and the proportion of grasslands (permanent + temporary) are higher in Wallonie (371 000 ha and 50 percent of the AA) compared with Flanders (221 000 ha and 36 percent of the AA) (Table 3).

Socio-economic and structural aspects of agriculture. In 2007, Belgium had about 48 000 farms, two-thirds in the Flemish Region and one-third in Wallonie (Table 4). Farm size in Wallonie is more than double (47 ha) compared with Flanders (20 ha) (Table 4 and Figure 3). In Wallonie, the industrial revolution began early in the nineteenth century, creating jobs outside the agriculture sector and promoting a labour transfer from agriculture to industry (coal mines and steel factories). This permitted a faster farm size increase compared to Flanders, where the industrial revolution only started in the mid-twentieth century. Flemish farmers therefore had to intensify to survive; they specialized in horticulture, pig, poultry or dairy production, which provide a significant income on small land areas. In Wallonie farms specialized in cereals, sugar beet, beef and dairy production. Most Belgian farms (73 percent) have permanent grasslands, and half have forage crops and/or cattle (Table 4).

Figure 3. Farm size in the Flemish and Walloon Regions in 2007 Source: FPS Economy, 2008.

In 2007, approximately 89 000 people worked in agriculture, about 2 percent of all employment (Organisation for Economic Co-operation and Development [OECD], 2008). The agricultural population is declining fast (by about 4 percent in the period 2006/2007) (FPS Economy, 2008); the number of farms fell by 50 percent between 1980 and 2005 (Goor, 2006). Most farms (72 percent) are managed by full-time farmers, although part-time farming is significant (28 percent). The farm population is ageing; in 2007 only 7 percent were under 35 and 41 percent were over 55 (FPS Economy, 2009). This is mainly due to the low profitability of small farms, the lack of suitable land especially in Flanders and around Brussels, the high level of investment needed by young farmers to enter the sector, uncertainties about the sector’s future (price and income variability) and the lower income from agriculture compared with other economic sectors.

The sector accounts for less than 1 percent (0.78 percent in 2007) of the Belgian GDP (OECD, 2008; FPS Economy, 2008) but this increases considerably when the upstream (e.g. machinery, fertilizer and pesticide production) and downstream (e.g. mill and sugar industry, abattoirs, animal feeding and dairy industries) sectors are considered. The economic importance of the total chain is often estimated at about 10 percent of the GDP (FPS Economy, 2007). When considering exports, the whole agriculture sector (animal and plant products, food products, drinks and tobacco) represented 9.9 percent of the Belgian GDP in 2007. However, this proportion is declining. In 2007, the agriculture sector GDP amounted to €23 323 million, including €5 483 million for animal products and live animals (€2 454 million for meat and €2 324 million for dairy products and eggs), €5 188 million for plant products and €11 549 million for food products, beverages and tobacco (FPS Economy, 2008).

2. SOILS AND TOPOGRAPHY

Topography and geology
Belgium’s geology is complex and variation within relatively short distances is far greater than in many parts of North Europe. With regard to surface layers, the geological history covers all periods, from the Pre-Cambrian to the Quaternary, but primary rocks and tertiary material cover most of the country. Most underlying rocks are sedimentary. Quaternary glaciations have exerted a profound impact on the landscape, especially in the north and centre. Major geological outcrops are presented in Figure 4. Figure 7 shows the different regions mentioned in the text.

Along the coast and the Scheldt estuary, sedimentary clayey terrains were gained from the sea (polders) in the recent Quaternary. Other Quaternary terrains include Pleistocene sands. To the north and east of Ghent and in the east of the Kempen, these deposits are from Higher and Medium Pleistocene. In the rest of the Kempen, they date back to the Lower Pleistocene. These sandy layers were formed during ice ages when the sea level was several metres lower than the present. During inter-glaciation episodes and at the end of the ice age wind erosion removed different types of materials, from relatively coarse to very fine textures. The heaviest particles (sand) were deposited first in the north of Belgium, while finer particles were deposited further south, in central and upper Belgium. In the Quaternary, this loess covered almost all of the country but that layer was later largely eroded. The deepest layers remain the Loamy Region (in Hainaut, Brabant and Hesbaye). To the east of the Sambre-and-Meuse furrow, this loess is mixed with the degradation products of the underlying rock. The loess deposit can still have a significant depth in Condroz, although less than in the Loamy Region. In the Ardennes, some deposits remain, mainly on the plateau.

Brussels is almost in the centre of a sedimentary Tertiary basin, which includes a large part of Flanders and Central Belgium. The sedimentary material was sand, mainly from the Eocene, deposited on the bottom of Tertiary seas. In the north of the basin, some terrains date back to the Pliocene, Miocene and Oligocene.

Spots of Higher Cretaceous chalk can be observed to the north of the Meuse, in Hainaut, close to Mons, and in Liège Province, close to Liège, on both sides of the river Meuse, including the Herve country.

To the south east of the Sambre-and-Meuse furrow, geological history gave rise to an intricate pattern of rock type distributions and marked differences in landscape and relief, often within relatively small areas. Periods of mountain building followed by periods of erosion have also had a profound influence on the landscape. Along the river Meuse, limestone from the Dinantian (Carboniferous) alternates with schists and sandstones from the Westphalian (Carboniferous). The Condroz landscape is characterized by an alternation of crests (called locally ‘tiges’ or ‘tiennes’) and depressions (called locally ‘chavées’) which run southwest to northeast. On the anticlines of the crests, the micaceous sandstones from the Famennian (Higher Devonian) are relatively resistant to erosion. The synclines of the depressions are made of calcareous stones from the Dinantian (Carboniferous) that are much softer and susceptible to erosion.  Schists of the Famennian are characteristics of the Fagnes to the west of the river Meuse and of the Famenne to the east of the river. Since these schists are relatively soft (frost sensitivity), erosion created a depression in these two regions. The soils of the Famenne depression are made of silt-clay formations deriving from schists from the Frasnian and mostly from the Famennian (both from Higher Devonian). These formations produced wet and nutrient-poor soils. The southern strip of the Famenne, the ‘Calestienne’, is made of Devonian limestone. To the southeast, the Lower Devonian schists and sandstones outcrop in the Ardenne anticlinorium. There is no limestone in Ardenne.

Figure 4. Geological map of Belgium Source: Service Geologique de Belgique [Click to view full map]

The Cambrian is represented in Southern Ardenne by phyllads, quartzites and quartzophyllads. The sedimentary rocks of the Ordovician are mainly schists, phyllads and quartzophyllads. In the Ardenne, schists and sandstones are nutrient-poor, particularly sandstone and generate very poor and acid soils. These soils are especially poor in phosphorus. However, the decomposition of schists produces soils with good potash availability. The Ardennes are part of the Ardenno-rhenan schistose massif (with the Eifel in Germany and the Oesling in Luxemburg). This massif has been formed in two phases of orogeny during the Caledonian (–444 to –416 million years) and Hercynian periods (–400 to –245 million years), and was severely eroded afterwards. At the extreme southeast of the country, the sedimentary rocks of the Lower Jurassic are dominant in the Belgian Lorraine or Jurassic Region. Some Higher Triassic terrains are present in the northeast of this region. These Mesozoic formations belong to the geological entity of the ‘Gulf of Luxemburg’ that constitutes a junction between the Paris Basin and the Germanic Basin. The Trias period is represented by a clay-sand complex of fluvial origin, surmounted by red and green marls including gypsum nodules and whitish dolomite layers. In other places, soft sandstones alternate with black marl and clay. This lagoon formation is overtopped by a red clay layer: Levallois Clay. The Lias (Jurassic) deposits are characterized by alternating sand and marl-clay mixtures. Erosion has resulted in ridge formation that give a characteristic structure to the landscape of this region: the cuestas.

Soil types

The soil map of Belgium (Figure 5) was prepared between 1947 and 1990 at 1/5 000 and published at 1/20 000. Its pragmatic legend is based on objective criteria such as texture (relative proportion of sand, loam and clay), natural drainage, profile development, stone proportion and soil depth. The dominant soil types of the main regions are described below.

In the maritime plain, except for the sandy soils of the dunes (yellow, 1 and 2 = colour and numerical codes on the map), soils are mainly clayey in the polders (green, 3 to 10).

In Lower Belgium:

• dry (very pale blue, 14) and humid (pale blue, 15) sandy and sandy-loamy soils with a humic and/or ferric B horizon (in the Kempen and around Bruges and Diksmuide);
• humid sandy and sandy-loamy soils with a humic and/or ferric B horizon (pale blue, 15) in combination with sandy and sandy-loamy soils with a coloured or textural B horizon (pink grey, 19); both are humid soils;
• light sandy-loamy soils and sandy-loamy soils with a textural broken down B horizon (dark pink, 27);
• sandy-loamy soils or loamy with an undifferentiated substratum on a clay-sand complex (pale brown, 38) (good arable soils);
• sandy-loamy soils or loamy with an undifferentiated substratum on clay (pale brown, 39) (good arable soils).

In Central Belgium:

• loamy soils with a textural B horizon (orange brown, 30 to 36). Drainage can be variable, soils can be dry to wet. These soils are the most representative of the soils of the Loamy Region and the best soils of the Sandy Loamy region. They are also partly present in Condroz. They are good arable soils;
• humid light sandy-loamy and humid sandy-loamy soils with a broken down B horizon (dark pink, 27);
• sandy-loamy or loamy soils with an undifferentiated substratum on a clay-sand complex (pale brown, 38).

In Condroz:

• stony-loamy soils with a textural or structural B horizon with a psammite stony content (yellow brown, 43) (on the crests);
• stony-loamy soils with a textural or structural B horizon with a limestone stony content (pale blue, 44) (in the depressions);
• moderately dry to moderately wet loamy soils with a textural B horizon (orange brown, 32 and 33) (loess deposits on the plateau and gentle slopes).

In Famenne, in the depression:

• stony-loamy soils with a textural or structural B horizon with a schist stony content (pale brown, 46);
• clayey and stony-loamy soils with a schist stony content (dark green, 49).

In Famenne, in Calestienne:

• stony-loamy soils with a textural or structural B horizon with a limestone stony content (pale blue, 44);
• stony-loamy soils with a textural or structural B horizon with a schist and limestone stony content (pale blue, 47).

In the Ardennes:

• stony-loamy soils with a structural B horizon with a schist and phyllad stony content (grey brown, 50) (the most common soil type in Ardenne);
• stony-loamy soils with a structural B horizon with a schist and sandstone stony content (pink brown, 51);
• dry (yellow brown, 52) (widespread soil type in Ardenne but less than 50) or wet (dark brown, 53) loamy soils with a structural B horizon and few stones.

In the Herve country:

• loamy soils with a textural B horizon (orange brown, 31 and 34) (in the west of the region; these soils are partly used as arable land and cropped with the same crops as in the Loamy region);
• stony-loamy soils with a textural or structural B horizon with a chalk or silexite stony content (pale green, 41) (covered by permanent grasslands);
• stony-loamy soils with a textural or structural B horizon with a limestone stony content (pale blue, 44) (covered by permanent grasslands);
• stony-loamy soils with a textural or structural B horizon with a schist stony content (pale brown, 46) (covered by permanent grasslands).

In the Jurassic Region:

• sandy to sandy-loamy soils with a textural B horizon (pale pink, 55);
• clayey to stony-loamy soils with a textural B horizon (red, 56);
• clayey soils with a structural B horizon (pale green, 57);
• clayey soils with a textural B horizon (dark green, 58).

According to the Soil Atlas of Europe (European Soil Bureau, 2005), the five major soil types in Belgium are: Albeluvisols in the north (e.g. part of the Sandy and Sandy-Loamy Regions), Luvisols covering a large part of the Sandy-Loamy and Loamy Regions, Podzols mainly in the Kempen and a part of the Sandy and the Sandy-Loamy Regions, Fluvisols in the valley bottoms especially in Lower and Central Belgium, and Cambisols in the southeast (e.g. Ardennes).

Figure 5. Soil map of Belgium Source: http://eusoils.jrc.ec.europa.eu/esdb_archive/EuDASM/lists/belgium.htm [Click to view full map]

3. CLIMATE AND AGRO-ECOLOGICAL ZONES

Climate
Belgium has an Atlantic temperate climate on a large part of its territory but a continental effect is recorded to the east of the Sambre and Meuse rivers. The Belgian climate belongs to the Cf category of the Köppen classification.

Rainfall and average temperatures of three representative sites are presented in Table 5 and Figure 6. Coxyde is on the coast; the climate is slightly cooler and the rainfall slightly lower compared to Brussels; rainfall is almost double and temperature significantly lower at the top of Higher Belgium (Mount Rigi). Climatograms reveal no periods of drought (Figure 6).

Table 5. Rainfall and average temperature per month of three representative sites
	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
Coxyde
Rainfall mm	53	42	44	44	50	53	60	66	69	74	69	63
Temperature 0 C	2.9	3.4	5.2	8.3	11.3	14.3	16.1	16.3	14.3	10.6	6.4	3.6
Brussels
Rainfall mm	67	54	73	57	70	78	75	63	59	71	78	76
Temperature 0 C	2.5	3.2	5.7	8.7	12.7	15.5	17.2	17.0	14.4	10.4	6.0	3.4
Mont Rigi
Rainfall mm	134	108	101	94	109	125	140	136	122	121	124	136
Temperature 0 C	-1.6	-0.7	1.5	5.2	9.2	12.3	13.9	13.8	11.4	7.4	2.5	-0.4
Source: Météo Nature, 2009

Coxyde (4 m asl)	Brussels (100 m asl)	Mount Rigi (680 m asl)
Lower Belgium	Central Belgium	Higher Belgium
P = 687	P = 821	P = 1 450
T = 9.4	T = 9.7	T = 6.2
Legend: Y1 axis and histograms: total monthly rainfall (mm); Y2 axis and red curve: average monthly temperature (°C); X axis: months of the year (from January to December); P: annual rainfall (mm); T: annual average temperature (°C).
Figure 6. Climatograms of three representative Belgian sites Source: Météo Nature, 2009.

Wind
The dominant winds come from the southwest and bring mild and wet air masses if they blow from south to west, unstable and cool air if they come from west to north. The north to east airstreams and those from east to south are mainly dry and cold in winter and dry and warm in summer.

Air temperature
Air temperature is mainly determined by distance from the sea and altitude. The sea has a thermal inertia that decreases and delays the seasonal variation of temperature along the coast: the winter is milder and the summer cooler compared with the interior of the country. This factor increases average temperatures by 1 °C in winter and decreases these temperatures by 1 °C in summer compared with Lower and Central Belgium.

Average annual temperature is about 10 °C in Central Belgium (Uccle) and about 6 °C at 650 m asl. Average monthly temperatures range from 3 °C in January to 18 °C in July and August in Central Belgium, and from –2 °C in January to 14 °C in July and August at 650 m asl (Table 5). Average temperatures decrease by 0.6 °C per 100 m of elevation. The average daily difference between the lowlands and the elevated plateau of the Ardennes thus reaches about 3.6 °C, although frequently this daily difference can reach about 6 °C.

Absolute extreme temperatures range between +40 °C and –30 °C. The annual absolute maxima reach, on average, 30 °C to 32 °C in Lower and Central Belgium. In Ardenne, these maxima reach only 28 °C. The annual absolute minima reach, on average, –10 °C at the coast, –11 °C to –14 °C in Lower and Central Belgium, –15 °C on the plateaus of Upper Belgium and -19 °C in the valley of the Ardennes. Temperatures fall below –10 °C, on average per year, for 2–3 days at the coast and for up to 12 days in the valleys of the Ardennes.

The number of frost days per year varies considerably from the coast to the Ardenne. At the coast (Oostende), they are below 45; in Central Belgium from 55 to 70. It increases very quickly from Namur on the river Meuse to the east, rising from 70–75 to more than 120 days. In Ardenne, frosts can occur as late as mid-June and as early as mid-September. This factor has a profound influence on the vegetation and agricultural production in this region.

Rainfall
Average annual rainfall ranges between 700 to 850 mm in Lower and Central Belgium (from 0 to 200 m asl). In Upper Belgium, rainfall increases with altitude up to 1 400 mm at 650 m. This doubling of annual rainfall occurs over less than 100 km. Rainfall is relatively well spread over the year (50 to 80 mm per month in Central Belgium); it can vary by 40 to 50 percent some years. On average there are 140 days of rain over most of the country (Table 6). The number of rainy days is higher in Upper Belgium and lower at the coast. The average number of rainfall days per month is higher (about 13 days) from November to January and lower (about 10 days) from June to October.

Snow
Snow is rare in Lower and Central Belgium: once every second year on average. The maximum snow depth does not exceed 6-13 cm in these regions. Snow usually increases with altitude. In Ardenne, the maximum snow depth is above 30 cm and can reach 70 - 80 cm. First snow appears on average at the end of November; 15 days earlier in the Ardennes. Last snows are observed at the beginning of April – end of April in the Ardenne. In Lower and Central Belgium, on average, 15 days of snowfall are recorded, 30 days in Upper Belgium and about 40 days at 650 m asl (Baraque Michel). The average period of snow cover is about three to five days in Lower and Central Belgium. In the Ardenne, it is longer and can even reach two months some years.

Sunshine
Total average sunshine is about 1 600 hours in Central Belgium (Table 6) and does not vary much throughout the country – extreme values are included within a range of 1 500 to 1 750 hours. As a result, total solar radiation also does not vary greatly. The higher values of sunshine are observed between April and September (about 140 - 200 hours per month).

Table 6. Average climate data (1971–2000) in Central Belgium (Uccle, Brussels)

Geographical regions
There are three geographical areas: Lower Belgium (up to 100 m asl, including the coast area), Central Belgium (between 100 and 200 m asl) and Upper Belgium (from 200 to 694 m asl) (Figure 7). The lowest point is slightly below sea level in the polders near the coast. The highest point reaches 694 m asl (Signal of Botrange), near the German border. Climate components are contrasted when comparing the coast and the highest altitudes. Geological layers and soils are extremely diverse when considering the land area of the country.

Lower Belgium comprises the dunes, the polders, the Flemish plain and the Kempen.

The altitude of Central Belgium increases slightly from west to east. Soils are deep, sandy or loamy. Loams (Quaternary loess) are among the best soils in the country for annual crops. In the main part of this region, the landscape is rolling, except when dissected by river valleys. The plateaus are loamy and erosion has exposed sand layers of the Tertiary period on the slopes of the valleys. In the southern (Hainaut) and eastern (Hesbaye) parts of this region, large farms grow mainly cereals, maize, sugar beet and potatoes.

Upper Belgium begins east and south of the Sambre and Meuse. Altitude increases quickly eastwards from 70–90 m asl (Meuse level) to 694 m asl (in the High Fens). Climate is more continental and soils are extremely diverse. They can be sandy, loamy or clayey and can derive from limestone, sandstones, schists or marine deposits of sand. Many soils are mixed with stones; some are shallow and cannot be ploughed.

Legend: Littoral = Coast area; Basse-Belgique = Lower Belgium; Moyenne-Belgique = Central Belgium; Haute-Belgique = Higher Belgium; Escaut = Scheldt; Hautes Fagnes = High Fens; Flandre sablonneuse = Sandy Flanders; Flandre Sablo-limoneuse = Sandy-loamy Flanders; Collines = Hills; Campine = Kempen; Région limoneuse hennuyère = Hainaut Loamy Region; Brabançonne = Brabant Region; Pays de Herve = Herve country; Ardenne du Nord-Est = Northeast Ardenne; Ardenne centrale = Central Ardenne; Lorraine belge = Belgian Lorraine; Région industrielle = Industrial Area. Figure 7. Geographical regions Source: Météo Nature, 2009. [Click to view full map]

Agro-ecological regions
Fourteen agro-ecological regions are identified (Table 7 and Figure 8). Four are exclusively in the Flemish Region (Dunes, Polders, Sandy Region, Kempen), three are partly in Flanders and partly in Wallonie (Sandy-Loamy Region, Loamy Region, Liège Grassland Region) and seven are exclusively in Wallonie (Hainaut Kempen, Condroz, Famenne, Grassland Region – Fagnes, Ardenne, Upper Ardenne, Jurassic Region). Lower Belgium includes the dunes, the polders, the Sandy Region, a part of the Kempen and the Sandy-Loamy Region. Central Belgium includes a part of the Kempen and the Sandy-Loamy Region, as well as the Hainaut Kempen and the Loamy Region. The other regions (Condroz, Famenne, Grassland Region – Fagnes, Liège Grassland Region, Ardenne, Upper Ardenne, Jurassic Region) are located in Upper Belgium. Figure 8 shows the proportion of arable land and grassland in the AA of each agro-ecological region.

Table 7. Proportion (percentage AA) of crops and grasslands in the 14 agro-ecological regions

			Dunes-Polders	Sandy Region	Kempen	Sandy-loamy Region	Loamy Region	Liège Grassland Region	Hainaut Kempen	Condroz	Upper Ardenne	Fagnes Grassland Region	Famenne	Ardenne	Jurassic Region
Arable land			75	64	75	71	79	15	66	64	5	28	33	24	28
	Grain cereals		33	18	16	27	37	4	25	34	2	13	14	5	10
		Wheat	27	5	2	16	28	2	23	22	1	6	6	1	4
		Barley	1	1	1	3	6	1	2	9	1	3	3	1	2
		Grain maize	4	10	12	7	2	0	0	0	-	1	0	0	0
	Industrial crops		12	2	1	9	19	1	8	13	-	2	3	0	1
		Sugar beet	8	2	1	8	13	0	5	7	-	1	1	0	0
		Chicory	0	0	0	1	2	0	2	1	-	-	0	-	-
		Flax	2	1	0	1	3	0	1	1	-	-	0	-	-
		Oilseed rape	0	0	0	0	1	0	-	4	-	1	2	0	1
	Potato		7	6	3	9	7	0	11	2	0	0	0	0	0
	Dry grain legumes		0	0	0	0	0	0	0	0	0	0	0	0	0
	Arable land forages		19	32	51	17	10	9	19	9	3	11	14	18	16
		Forage maize	11	23	28	13	8	7	16	7	2	8	8	4	8
		Temporary grassland	6	8	21	4	1	2	1	2	0	2	5	14	7
	Outdoor vegetable		2	4	2	6	3	0	2	1	0	0	0	0	0
	Outdoor ornamental crops		0	0	0	0	0	0	-	0	0	0	0	0	0
	Set-asides		2	1	1	2	3	0	2	4	0	1	2	0	1
Permanent crops			0.67	1.84	1.53	3.34	1.87	0.98	0.09	0.09	0.00	0.10	0.10	0.10	0.05
	Outdoor nursery		0.03	1.05	1.14	0.38	0.14	0.08	0.08	0.04	0.00	0.01	0.01	0.10	0.04
		Ornamental plants	0.03	0.69	1.06	0.23	0.04	0.04	0.08	0.03	0.00	0.00	0.01	0.01	0.02
		Forest plants	-	0.27	0.06	0.02	0.03	0.03	-	0.00	-	0.00	-	0.09	0.01
		Fruit trees	-	0.09	0.02	0.13	0.07	0.01	-	0.00	-	0.00	-	0.00	0.01
	Orchards		0.64	0.76	0.34	2.94	1.70	0.90	-	0.05	-	0.10	0.09	0.00	0.01
Permanent grassland			24	34	23	26	19	84	33	36	95	72	67	76	72
Glasshouse crops			0.06	0.69	0.36	0.19	0.02	0.01	0.02	0.02	0.00	0.00	0.00	0.00	0.00
Woodland			0.02	0.14	1.87	0.44	0.10	0.36	-	0.65	0.86	0.05	0.46	1.35	0.22
	Christmas trees		0.01	0.05	0.10	0.02	0.02	0.00	-	0.04	0.04	-	0.01	0.92	0.04
	Other woodland areas		0.02	0.09	1.77	0.42	0.08	0.36	-	0.61	0.82	0.05	0.45	0.43	0.18
	Source: FPS Economy - DGSEI, 2008; Statistics of 2007

Legend: Dunes = Dunes; Polders = Polders; Région Sablonneuse = Sandy Region, Campine = Kempen, Région Sablo-limoneuse = Sandy-Loamy Region; Région Limoneuse = Loamy Region; Région Herbagère Liégeoise = Liège Grassland Region; Campine Hennuyère = Hainaut Kempen; Condroz = Condroz; Famenne = Famenne; Région Herbagère (Fagnes) = Grassland Region – Fagnes; Ardenne = Ardenne; Haute Ardenne = Upper Ardenne; Région Jurassique = Jurassic Region. Figure 8. Agro-ecological regions and proportion of arable land and grassland (temporary + permanent) in the AA (percentage) Source: Map retrieved from: http://lv.vlaanderen.be/nlapps/docs/default.asp?id=102; Statistics from FPS Economy (2009); Drawing: Cooparch-RU (2009); Calculations: Peeters 2009. [Click to view full picture]

The dunes are a narrow, almost continuous strip, 65 km long between the French and Dutch borders. They have almost no agricultural importance but high ecological and landscape value.

The polders are lands gained from the sea, drained and protected from the tides by a system of dams and locks. Their altitude is lower than 5 m asl. Soils are deep, clayey and fertile. The land is divided by a network of canals that regulates the level of the water table and defines the fields. Winter wheat, sugar beet, potato and forage maize are the most important crops, and yields are among the highest of the country. Permanent grasslands (24 percent AA) are dominated by Lolium perenne owing to the ideal conditions for its development: nutrient-rich and deep soils, regular water availability (rainfall and water table) and regular defoliation by intensive grazing. Temporary grasslands occupy 6 percent of the AA (about half of the forage maize area). The Belgian Red from Flanders (now a threatened breed) and the Red and White were the two traditional, dual-purpose breeds of cattle; the latter still represents one-quarter of dairy cow herds but they are now much more dairy oriented; the Holstein Friesian represents three-quarters. Suckling cow herds (Belgian Blue) have increased greatly in the last 25 years, reaching the same number as dairy cows. As in the other Flemish regions, pigs and poultry are important though slightly less in this region. In all these regions, pig and poultry slurry is not easily spread under good agricultural and environmental conditions. A system of manure management has been implemented by the Flemish regional government. The polders are an important wintering site for large populations of wild geese (Anser spp., Branta spp.) that graze on permanent grasslands. The increase of nitrogen fertilizer use on grassland since the 1960s combined with the maritime climate of this region has induced an almost continuous grass growth in autumn and winter that provides nutritious food for these birds. Their populations exploded in winter during that period.

The Sandy Region (Flemish plain) is flat except for some small hills. The soils are sandy and mainly originate from Tertiary marine deposits. These soils were originally poor but, having been progressively fertilized and improved over centuries, they are now very fertile. Water availability is not as good as in the polders and yields are slightly lower. Forage and grain maize are the main crops. Outdoor and protected nurseries are highly developed, especially around Ghent. The production of ornamental plants like Azalea spp. and bulbs, and vegetable production in glasshouses (tomato and salad mainly) provide high income to the farmers of this region. Permanent grasslands (34 percent AA) are important compared to other Flemish regions. Their sward quality is very good but, in the lowest and the wettest parts of the river valleys, Agrostis stolonifera and Alopecurus pratensis can be locally important. In addition to vegetables and flowers, the region is also oriented to dairy, pig and poultry production. Dairy breeds are similar to those of the polders. The Belgian Blue cow is now the most important breed of the region, and suckling cow numbers are similar to the total number of dairy cows. The numbers of pigs and poultry are similar to those of the Sandy-Loamy Region. Both regions have the highest number of pigs in the country and the second most important numbers of poultry after the Kempen.

The Kempen have a slightly higher altitude (between 20 and 100 m) compared with the three previous regions and the climate is somewhat more continental. The sandy soils of the Kempen were mainly covered by dunes, heather (Calluna vulgaris) moors and peat bogs at the beginning of the twentieth century, when they were extensively used for sheep grazing. The majority of these semi-natural habitats have now been planted with pines (Pinus sylvestris) or converted into grasslands and forage maize fields. The remaining area of semi-natural habitats is protected in nature reserves. Agricultural soils are sandy or peaty but, since the 1950s, they have been improved on a large scale by drainage and fertilization. Green forage for cattle is provided by permanent grassland (23 percent AA), temporary grassland (21 percent AA) and forage maize (28 percent AA). Dairying is much more important than beef. The Holstein Friesian, including the Red Holstein, is the most important cattle breed. Cash crops are of minor significance; grain maize (12 percent AA) is the main cash crop. Outdoor ornamental plant production, especially trees and shrubs, is developing. Glasshouse production, mainly of tomatoes, peppers, salad and strawberries, is important – about one half of that of the Sandy Region. Asparagus is a speciality outdoor crop. The Kempen is one of the three important regions for pig and poultry production, together with the Sandy and Sandy-Loamy Regions.

The Sandy-Loamy Region lies mainly in the Flemish Region, with a small part in Wallonie. Plateaus are loamy, valley slopes are sandy and valley bottoms are covered by alluvial soils that are clayey and rich in organic matter. Annual crops are concentrated on plateaus, while the steepest slopes are often covered by forests (mainly Pinus spp.). Grasslands are located close to farms or in valley bottoms. Cereals, sugar beet and potato are the main cash crops. The total area of forage maize (13 percent AA), temporary grasslands (4 percent AA) and permanent grasslands (26 percent AA) is not as large as in the Sandy Region and the Kempen. The region is clearly more specialized in cash crops compared to the two others. It has the highest proportion of potatoes in its AA (9 percent) (the small Hainaut Kempen excepted) before the Sandy and the Loamy Regions. Most Belgian orchards (apple, pear, cherry and plum trees by decreasing order of importance) are in this region. The region also leads in outdoor vegetable production. Leeks, cabbages, Brussels sprouts and witloofs (chicory or endive) are the main species. These crops are highly developed around Mechelen. Witloof (white leaf in Flemish dialect) is a local specialty. It was domesticated from Cichorium intybus in Brussels by a gardener of the Botanical Garden at the beginning of the twentieth century and first developed at a large scale in the triangle Brussels-Mechelen-Leuven. The region is third after the Sandy Region and the Kempen for glasshouse production. Dairy and suckling cow populations are comparable but there are more Belgian Blue than Holstein Friesian cows, the two dominant breeds. Red and White cows are much less numerous compared with the Sandy Region and the Kempen. The number of pigs (about 2 million) is similar to the Sandy Region and twice the value of the Kempen (about 1 million). The number of poultry (7 million) is also similar to the Sandy Region but slightly lower than the Kempen (10 million).

In contrast with the Sandy-Loamy Region, the main part of the Loamy Region is in Wallonie. The loess deposit is almost continuous. The Quaternary loam layer is as deep as 20 metres in some parts of the region, constituting favourable conditions for annual crops. Cash crops cover almost 80 percent of the AA. The proportions of cereals, sugar beet and chicory in the AA are the highest of the country. The proportion of potato is one of the highest. In the 1960s and 1970s, the traditional crop rotation was ‘sugar beet–winter wheat–winter barley’, but this rotation diversified greatly thereafter. The sugar beet area decreased while potato and chicory areas increased. The winter barley area was heavily reduced while the winter wheat area increased greatly and the flax area developed slightly. Forage maize, as in the other regions, is not usually a part of the crop rotation; it is grown every year on the same fields and receives a large part of the animal manure. Its area has increased considerably since the 1960s. The region is the second most important after the Sandy-Loamy Region for outdoor vegetables (mainly peas and beans) and fruit (mainly apple and pear) production. Orchards are concentrated in the east of the region, around Tienen. Permanent grasslands are mainly on the wet soils of the valleys because they cannot easily be ploughed. They are mainly grazed by beef animals and dairy heifers, often at very high stocking rates. Some grassland fields are on well-drained soils, close to farms, especially for dairy cows. Suckling cows are clearly dominant compared with dairy cows and the Belgian Blue is by far the dominant breed. Animals are fed with forages from grasslands and maize but also with ensiled sugar beet pulp. The concentration of pigs and poultry is much lower compared with the four previous regions.

The Hainaut Kempen is a very small region. The proportion of annual crops and grasslands in the AA is close to that of the Sandy-Loamy Region.

In Condroz, on a north–west/south–east transect, the relief has the shape of corrugated iron, with crests and depressions alternating regularly. The sandstone crests are covered by forests of broad-leaf species where Fraxinus excelsior is dominant. Their average altitude is about 250 m asl. Permanent grasslands are found in the calcareous depressions. Arable farming is possible where loess deposits are adequate. As in all following regions, the Condroz is southeast of the river Meuse (except a very small part), in Upper Belgium. It is a transition region between the arable farming systems of the Loamy Region and the livestock-based systems of the other regions of the southeast of the country. Cash crops (55 percent AA) and forage crops (45 percent AA) have similar importance. Most farms have a mixture of livestock and annual crops. Winter wheat and winter barley occupy a similar proportion of the AA compared with the Loamy Region but sugar beet and chicory are less important and oilseed rape is much more important. A significant part of root crop production is replaced by oilseed rape on stony and shallow soils.

Permanent grasslands (36 percent AA) are mainly on the calcareous soils of the depressions. Before the widespread use of minerals in animal nutrition these grasslands were appreciated for their positive influence on animal skeletal development and bulls of this region were famous. Temporary grasslands (2 percent AA) and forage maize (7 percent AA) are grown on deeper soils. Lucerne was once widely grown but has almost been abandoned since the 1960s. However, there is a renewed interest for this persistent and high-yielding forage as in other regions, including the Loamy Region. Suckling cows (mainly Belgian Blue) are much more abundant than dairy cows (mainly Holstein Friesian). The breeding centre of the Belgian Blue is in Ciney in the middle of this region. Every week Ciney hosts one of the largest cattle markets of West Europe. The Condroz is the second most important region of Wallonie for pig production after the Liège Grassland Region and the most important for poultry (mainly broilers). The number of animals (pig and poultry) is, however, much smaller compared with the Flemish Regions. In Condroz, the manure of these monogastric animals can be used under good agricultural and environmental conditions on crop and grassland areas.

Famenne, Grassland Region – Fagnes, Ardenne and Jurassic Regions are meat-oriented. Liège Grassland Region and Upper Ardenne are specialized in dairy production.

The Famenne is a large depression dominated by silt-clay soils. It has an average altitude of about 100 m asl and is thus lower than neighbouring Condroz and the Ardenne. A calcareous strip, several hundred metres deep, follows this depression to the southeast of the region, the Calestienne (probably the contraction of ‘cales’ = limestone and ‘tienne’ = crest). Its altitude is about 100 m higher than the depression. Soils are very stony, dry and shallow. They warm up quickly in spring. Several sub-Mediterranean plant species reach one of their northern limits of distribution in Europe here (e.g. Buxus sempervirens, Dianthus carthusianorum, Helianthemum spp., Potentilla neumanniana, Quercus pubescens). Steppic or substeppic plant species can also be found (e.g. Acinos arvensis, Brachypodium pinnatum, Bupleurum falcatum, Koeleria macrantha, Seseli libanotis, Silene nutans, Vincetoxicum hirundinaria). It is one of the best regions for orchids in Belgium, including several Ophrys species. Most chalk grasslands that constitute the habitat of these species are now protected. The Calestienne is in contact with the Ardennes massif. Cold air masses from the higher plateau of the Ardennes can accumulate in the depression of this region, creating a temperature inversion and inducing a cool climate that is comparable to the Ardenne’s climate at a much lower altitude. The Famenne is a grassland-dominated region like all the following ones. Cattle provide the largest portion of farmers’ income. There are still some cereals and oilseed rape in the crop rotation but a very high proportion of the AA is devoted to forage crops, including to permanent grasslands (67 percent AA). The system of this region has remained relatively extensive for a long period because spreading fertilizers on the wet soils of the central depression and shallow calcareous soils of the Calestienne was not economic. Drainage of the wet soils has changed this situation and fertilizer use has increased since the 1970s. The structure of the cattle herd is similar to that of the Condroz. Belgian Blue animals are the most frequent.

Although the soils are different, the characteristics of the farming systems of the Grassland Region – Fagnes are similar to those of Famenne. The proportion of permanent grassland is slightly higher and dairy cows are more important in this small region; they have the same importance as suckling cows.

The Liège Grassland Region is a dairy region of two main parts separated by the river Vesdre: the Herve country in the north and the Liège Ardenne in the south (a very small part of the Herve country is in the Flemish Region, the Voeren). Soils are very different in these two subregions and altitude increases relatively quickly southwards after the river Vesdre. In the Herve country, the Cretaceous period gave rise to a silty marl or a clay rich in glauconite. Soils are wet and often shallow. In the Liège Ardenne, the underlying rock is mainly schist, soils are often shallow and drainage is usually good. The landscape of this region is very picturesque because of the steep slopes and a high proportion of forest. Both regions have a bocage landscape with hedges and trees marking field boundaries. In the Herve country, soils are heavy and wet. A grassland-based system using permanent pastures planted with apple and pear orchards and bounded by hedges, raising dairy cows and pigs (fed with the residues of fruit and butter processing) and selling milk, butter, cheese (Herve cheese), a local marmalade (Sirop de Liège) and pork-butchery has existed since the sixteenth century. The density of trees and hedges has decreased greatly since the 1960s, and traditional orchards have almost completely disappeared. The structure of farms is relatively similar in both subregions, all mainly based on permanent pasture (84 percent AA). Grazed forages from permanent pastures are complemented by forage maize (7 percent AA) and temporary grasslands (2 percent AA). Some cereals are also grown. Many farms have no crop at all. The Herve country is more intensive than the Liège Ardenne and can be compared with the systems of the south of the Netherlands or the Kempen, although the proportion of forage maize is smaller. Crop production is often difficult in the Herve country because soils are wet and shallow. In the Liège Ardenne, maize growing is limited by shallow soil, by the steepness of slopes and by a cooler climate because of the higher altitude. Holstein Friesian is the main breed. Belgian Blue suckling cows were introduced in the region after the establishment of milk quotas in 1984. While the number of dairy cows has decreased with the increase of their individual performances, the total number of cows has remained almost stable by the addition of suckling stock. Many farms have thus two specialized herds, one with dairy cows and another with sucklers. Since no cereal straw is produced locally, dairy sheds produce slurry. The sward quality of permanent grasslands is good because of satisfactory grazing management, high fertilization, dock (Rumex obtusifolius) control and perennial ryegrass (Lolium perenne) seed oversowing and sod-seeding. In contrast with the Ardenne, there are almost no temporary grasslands. Silage cuts are taken from permanent grasslands from early May until autumn. On average, each area is cut more than once a year (about 130 percent of cumulated cuts). Frequent defoliations associated with the alternation of grazing and cutting constitute an ideal system for producing high-quality grass and for controlling weeds. After a cut, the regrowth of grass is clean, homogeneous and very palatable. Slurry is applied in priority on plots reserved for silage. Total annual nitrogen fertilization often reaches about 300 kg N/ha or more. Pig concentration is one of the highest of Wallonie. Cow and pig slurry cannot easily be spread on the AA under good agricultural and environmental conditions.

Four types of Ardennes are recognized: the Lower and Central Ardennes of Luxemburg Province, the Liège Ardenne and the Upper Ardenne of Liège Province. In Central Ardenne, a rolling plateau is dissected by deep valleys. The average altitude is 500 m asl. The Liège Ardenne has a pronounced orography. The top of Upper Ardenne reaches 694 m asl and is covered by a large peat bog, the High Fens. In the Ardennes, the climate and the presence of some plant species are characteristic of a submontane region. Soils were originally nutrient-poor and acid. A large part of the region is covered by forests (52 percent). Broad-leaf forests (mainly Fagus sylvatica) cover 31.5 percent of the forest total, and plantations of exotic trees (mainly Picea excelsa) 56 percent. The remaining part of the land (48 percent) is mainly covered by grasslands. Human population density is low.

The Ardenne (of Luxemburg Province) is a meat-oriented region. Soils developed mainly on schist or sandstone and were originally acid and nutrient-poor. Drainage is usually good. Because of an average altitude of 500 m asl, the climate is cold and rainy (annual rainfall: 1 000 to 1 500 mm). For a long period this region was isolated from the rest of northwest Europe. Until 1840, moorland dominated the landscape. Mixed herds of cattle, sheep, horses and pigs grazed the moors and the forests, and shifting areas of rye, oats and potatoes were grown on the moor. Some areas of permanent cropping were concentrated close to villages. Hay meadows were located on the slopes of the valleys and were irrigated, mainly to induce faster grass growth in early spring and to provide nutrients that are dissolved in the river water. After the harvest of crops and hay, all animals could graze private fields. In winter they were fed with straw and low-quality hay and suffered from chronic undernourishment at that period of the year. The local red cow was small (about 300 kg), spirited and hardy. The pig breed was still of the Celtic type with a large head, narrow legs and a hairy skin. The local cow and pig breeds disappeared at the end of the nineteenth century. The local sheep, the ‘Ardennais Roux’, has a red face and was mainly bred for wool but its meat quality was famous. The breed was saved in extremis at the end of the twentieth century. A local draught horse, the ‘Ardennais’, is still used for working in the forest and for agri-tourism.

In the middle of the nineteenth century, the general economic context of the region changed with the development of the railway. The Belgian State organized lime depots along the railway to improve the moor and intensify agricultural production. The State also obliged the municipalities to transform the commons into productive lands and a large part of the land was sold to private owners. The privatization of the commons was achieved very late in Ardenne compared with the other Belgian regions. About half of the moor was planted with spruce (Picea excelsa), and the remainder was transformed into pastures and annual crops. Moors disappeared almost completely around 1900. Improved permanent pastures were sown with seeds collected in hay barns or with commercial seed mixtures. In addition to lime, phosphorus-rich slags from the developing steel industry of Wallonie were used as a source of phosphorus since the mid-nineteenth century. Higher soil pH and phosphorus availability were favourable conditions for the development of legumes: white clover (Trifolium repens) on permanent pastures and red clover (T. pratense) mixed with timothy (Phleum pratense) on temporary pastures. These modifications of soil fertility induced a better mineralization of soil organic matter that released mineral nitrogen. Temporary pastures were incorporated as leys in the crop rotation. Better organic nitrogen mineralization combined with nitrogen fixation by legumes dramatically changed the availability of this key nutrient and yields increased rapidly. The proportion of permanent pastures has increased continuously since the second half of the nineteenth century. Crops have progressively disappeared because yields are limited in Ardenne due to unfavourable climate and insufficient soil depth. The rotation ‘potato–spelt–mixture of spring oats and barley (‘méteil’)–temporary grassland’ was typical in the 1960s–70s. This crop rotation had an average duration of six years; temporary pastures were kept for three years. Grassland seeds are sown in spring cereal fields just after the sowing of cereals at the end of March or the beginning of April. They develop slowly under the cereal crop. After cereal harvest in August, grassland plants manage to establish fully just before winter. They can sometimes be harvested for a first silage cut in autumn. Grassland mixtures are based on Festuca pratensis, Lolium perenne, Phleum pratense and Trifolium pratense. In the mid-twentieth century, the simpler mixture of P. pratense and T. pratense was more frequent; it is well adapted to infrequent cuts and haymaking. The proportion of L. perenne in the mixtures has increased over time, especially after the adoption of silage making for the conservation of forages. Silage making induced earlier and more frequent cuts. Temporary pastures were subsequently cut twice and then grazed in summer and autumn, instead of being cut once and grazed as before. L. perenne is better adapted than P. pratense to this more intensive system. The mixture of two or three grasses is seen by farmers as an insurance against climatic or management accidents. Because of the continuous decrease of the crop area, the duration of temporary grassland plots increased and it is now not uncommon to find ten-year old or even older temporary grassland plots. This system, characterized by infrequent cuts, relatively high fertilization and long plot lifetime, favoured Elymus repens which develops in spring cereal fields that cannot be chemically weeded because of the presence of undersown pasture seedlings. Couch populations can then increase very rapidly from rhizomes in young grassland plots.

Temporary pastures are the main source of conserved forages while permanent pastures are exclusively grazed. Summer grass growth depression is compensated by the inclusion of temporary grassland plots in the grazing area. Nitrogen fertilization has been usual since 1960. It is now still ‘moderate’ in Ardenne compared with the other Belgian livestock regions. The average annual nitrogen fertilization is about 100–200 kg N/ha. The grazing period starts late in Ardenne, at the beginning of May, and ends in October or the beginning of November. In recent years, a higher proportion of heifers has been kept outdoors in winter because of an increase in herd size and a lack of available buildings. This system that is healthy for animals is damaging for grassland swards. It should be restricted to small, condemned grassland plots. The introduction of silage and haylage in place of hay in the 1970s and the development of the double-muscled Belgian Blue during the same period have boosted meat production. These evolutions have transformed a traditional system inherited from the Middle Ages into a modern dynamic system in about one century only. Farmers of this region have remained specialized in the production of young animals – bulls of about six months old – that are then sold and fattened in other regions, mainly in Lower and Central Belgium. Cereal straw is used for bedding. The choice of cereal species is determined by climate. Spelt is more frost-resistant than winter wheat and is also a good complement for animal feeding. Winter barley cannot be safely grown in Ardenne, and has been replaced by spring cereals that are not exposed to severe frosts. Some of the spelt is used for bread making; special cultivars are used for that purpose. Spelt bread is a specialty of this region. Cereal yields are relatively low and harvest can be complicated by rainfall frequency and intensity in summer. Some years (once every ten years or so), the cereal harvest has to be very delayed or even abandoned. This region suits potatoes, although late frosts in June can destroy fields, or parts of them. The local Corne de Gatte (Goat Horn) is a low-yielding high-quality cultivar. Farmyard manure (FYM) is traditionally applied to potatoes and temporary pastures. The development of modern composting techniques has encouraged farmers to spread composted FYM on permanent pastures since the 1980s.

The Upper Ardenne is a dairy-oriented region with an agricultural history comparable with that of the Ardenne. Soils are also similar while climate is slightly colder and rainier (about 1 500 mm of annual rainfall) because of a higher altitude (between 500 and 700 m asl). Production systems are completely different. Most farmers are German-speaking, except around Malmédy where they speak French. Small farms and a high division of plots are characteristics. In the 1960s, the cattle breed was a small cow, the Westphalian Red and White. It was a dual-purpose breed with an average annual dairy production of about 4 500 litres/cow, progressively crossed with (Red) Holstein and oriented to milk production. Apart from the breed, the management system is comparable to the dairy system of the Liège Grassland Region but somewhat less intensive.

To the south of the Ardenne massif, the Jurassic Region or Belgian Lorraine has a lower altitude, ranging from 250 to 400 m. The climate is milder than in the other regions of Upper Belgium and drier compared with the Ardenne. Soils are extremely diverse, from dry sandy to wet clayey soils, and marls (calcareous clay) and (calcareous) sands are dominant. This region is prolonged in Luxemburg by the Gutland and in France by the Lorraine. It has twice as many cereals (10 percent AA) and forage maize (8 percent AA) as the Ardenne, but the proportion of permanent grassland is very slightly lower (72 percent AA). Production systems are comparable even if the proportion of dairy cows is slightly higher. The Belgian Blue breed is still dominant. The proportion of French meat breeds in the total of Belgian Blue, Charolais, Limousin and Blonde d’Aquitaine cows is higher (18 percent) compared with the Ardenne (7 percent). [See Photos 1-5]

Photo 1. Flemish plain. Left: Grasslands, avenue of poplars (Populus) and some pollarded willows (Salix alba). Source: ILVO - Merelbeke. Right: Old pollarded willows (Salix alba). Source: ILVO - Merelbeke.

Photo 2. Flemish plain. Left: Canal delimiting a grassland plot. Source: ILVO - Merelbeke. Right: River and intensive grassland just before cutting for silage. Source: ILVO - Merelbeke.

Photo 3. Central Belgium. Grasslands grazed by Brabant horses in a rolling landscape with poplars (Populus) and a pollarded willow (Salix alba). Source: ILVO - Merelbeke.

Photo 4. Liege Grassland Region. Left: ‘Bocage’ landscape. Grasslands are delimited by hedgerow networks. Farms are isolated at distance from villages. Source: S. Rouxhet. Centre: Dairy cows close to farm buildings. Source: Wallonie Elevages. Right: Traditional orchard of pear trees (Pyrus communis) planted in a grassland plot. Source: ILVO - Merelbeke.

Photo 5. Ardenne. Left: Plateau with a village, grassland plots, arable land (in the background) and farm buildings (in the foreground). Source: S. Cremer. Middle: Temporary grassland (grass/clover mixture) managed by cutting for making haylage or silage. Source: S. Cremer. Right: Belgian Blue cows grazing. Spruce (Picea excelsa) forest in the background. Source: S. Cremer.

Forests
Forests are concentrated in the southeast of the country. The forest proportion in Wallonie (28 percent) is close to the European Union average, while Flanders (10 percent) is one of the European regions that has a small proportion of its territory covered by forest, like the United Kingdom, Ireland and the Netherlands. In Belgium and the two Regions, indigenous broad-leaf species and exotic coniferous species have a similar extent (Table 8). Compared with the situation in the mid-nineteenth century, the area of Belgian forest has almost doubled through planting of conifers mainly on semi-natural habitats (e.g. moors, dunes and chalk grasslands) and to a lesser extent by converting broad-leaf forests.

The proportion of forest in Luxemburg Province is the highest of the country (51.7 percent). The forest area of this province represents 42 percent of the Walloon and 37 percent of the Belgian forest areas. In the Ardennes (Ardenne of the Luxemburg Province, Upper Ardenne and Liège Ardenne) and the Kempen, conifers are dominant while in the Loamy Region, Famenne and Condroz, for instance, broad-leaf trees represent about 70 percent of the area (Table 9). Forest becomes the predominant land use above 300 metres and the proportion of conifers also increases with altitude (from 25 percent at 200 m asl to 85 percent at 600 m asl). The proportion of coppice increases from 59 percent to 99 percent within the same altitudinal range (Table 10).

Table 9. Proportion of forest area and percentage of broad leaf and coniferous species forests in selected agro-ecological regions in Wallonie

Table 10. Proportion (percentage) of forest types in the total forest area and proportion of forest according to altitude class in Wallonie

The continuous increase in the proportion of coniferous species since the mid-nineteenth century stopped after the 1980s (Table 11).

Table 11. Evolution of productive forest and proportion of broad-leaf and coniferous species in Wallonie since the nineteenth century

Eleven forest communities are identified in Wallonie: beech (Fagus sylvatica) forests (8.9 percent) (70 percent of which are in Ardenne), oak (Quercus pedunculata and Q. sessilis) forests (17.2 percent), mixed communities of noble broad-leaf species (mainly F. sylvatica, Quercus spp. and Fraxinus excelsior) (12 percent), other broad-leaf species communities (mainly Betula spp. and Quercus spp.) (9.1 percent), hybrid poplars (Populus spp.) (2.1 percent), coppices of Betula spp. and Quercus spp. (3.2 percent), spruce (Picea excelsa) (36.1 percent), Douglas fir (Pseudotsuga menziesii) (2.3 percent), larch (Larix spp.) (1.7 percent), pine (Pinus spp.) (3.2 percent) and other conifer plantations (4.3 percent) (Lecomte et al., 2000c).

4. RUMINANT LIVESTOCK PRODUCTION SYSTEMS

Recent history and context
Since the beginning of the Common Agricultural Policy (CAP) of the EU in the early 1960s, Belgian livestock and grassland systems have been highly intensified. Stocking rate, frequency of cutting for conservation, fertilizer use, drainage, irrigation, re-sowing, oversowing and overdrilling with improved grass and legume cultivars, and weed control with herbicides have become increasingly important. The number of plant species (and biodiversity in general) fell dramatically in grassland swards while forage yields increased and feeding quality improved. At the same time, farm and farmer numbers were reduced and farm size increased, thus modifying the traditional landscape by an enlargement of plot size and, as a direct consequence, a decrease in field margins and edges. In bocage regions, such as in the Liège Grassland Region, many hedges were cut after field enlargement. The pasture area was slightly reduced in favour of the production of fodder maize and cash crops. Maize developed from almost nothing in the 1950s up to 12 percent AA in 2007. In the last few years, the forage maize area has seemed to be stable or even slightly decreasing in Wallonie. Only a small proportion of grasslands have been abandoned, on marginal soils. Urbanization is responsible for a higher area loss.

Specialization of production has resulted in the progressive disappearance of mixed farming. Some regions specialized in arable crops (e.g. the Loamy Region), others in animal husbandry (e.g. the Ardenne). The area of temporary pastures and especially legume-based pastures (Medicago sativa, Trifolium spp.) declined in all regions. In the 1950s, this trend was reinforced by the decrease of agricultural labour that made traditional haymaking very difficult. The use and proportion of legume species (especially T. repens) in grazed swards were also reduced by the widespread use of nitrogen fertilizers. Animal breeds were specialized for milk or meat production, while dual-purpose breeds were reduced. Dairy systems were concentrated in low altitude regions. Beef systems occupied the more marginal soils and climates in the southeast of the country. After the implementation of the milk quotas, beef systems spread throughout the country. As a result of specialization, animal performances increased. In dairy systems, from an average annual production of 4 500 litres per cow in the 1970s, the production increased to 7 000 litres/cow in about 35 years, with some herds or cows now reaching an annual production of 10 000 to 12 000 litres/cow. This regular increase of dairy performances of about 1.3 percent per year has been possible because of an international effort to breed the Holstein. This has led to a decrease of the populations of many dual-purpose breeds (e.g. Belgian Red from Flanders, Red and White, dual-purpose Belgian Blue). Yield increases also induced changes in animal feeding; more concentrates were used at the expense of green forages.

The implementation of milk quotas slowed down this trend because control of the volume of production required a decrease in production costs. This was achieved by better utilization of green forage. Yields of 3 000 to 4 500 litres of milk per cow and per year are regularly obtained on the basis of green forage (forage maize and grassland). Lower production costs were also achieved by higher production per cow. The milk quota of each farm has been fulfilled with a decreasing number of cows, thus decreasing the proportion of maintenance feed and increasing the proportion of production needs in the total feeding costs. The CAP reforms of 1992 and 2000 induced a significant decrease of cereal grain price (about 50 percent) and that again encouraged dairy farmers to use these as livestock feed. Moreover, farmers tended to use more maize silage at the expense of grass grazing and grass silage when dairy cow production was above roughly 6 000 litres/cow. They did not trust the grass quality and grass intake potential of their high-yielding cows, especially in rainy weather and with unfavourable temperatures. They thus tended to keep cows partially indoors or systematically to complement grass grazing with maize silage, leading to a decrease of grassland proportion in the AA. This trend was particularly prominent in Flanders (Polders, Sandy Region, Sandy-Loamy Region, Kempen). In the Grassland Regions of the southeast the farmers’ strategy was somewhat different; they used less maize but increased concentrate use. They also tried to reduce production costs by better use of grazing and grass silage. In beef production systems, the Belgian Blue emerged as almost the only profitable breed in the 1970s. Grazing remained the basis of suckling cow systems and animals were fed in winter mainly with hay and haylage. Concentrates and maize silage were restricted mainly to bull fattening. Ox fattening almost disappeared in favour of young bulls.

These system changes had impacts on landscape and wildlife by reducing diversity and complexity. Farmers had to face criticisms for their negative effect on the quality of ground- and surface waters. Measures had to be taken to decrease nitrate and phosphate pollution.

The CAP reform of 2003 introduced four new principles: decoupling, cross-compliance, modulation and partial re-nationalization. With regard to grasslands, decoupling and cross-compliance had some important consequences. Decoupling ensured that premiums were not linked with crop or animal types but to the eligible area, thereby suppressing the ‘maize premium’ which encouraged farmers to use this forage at the expense of pastures. The use of pastures was also no longer indirectly supported through animal premiums but directly through area payments (the system was, however, applied with some flexibility; suckling cows and ewes are still supported). This measure reduced the distortions that were unfavourable to pastures. Cross-compliance required that farmers must comply with a set of good agricultural and environmental practices (GAEP), including the obligation to maintain the proportion of permanent grassland in the AA, in order to maintain their land in good agricultural and environmental condition (GAEC).

Farmers also have to respect the Habitat and Bird Directives, although these represent less than 5 percent of the AA in Belgium. The Nitrate Directive is also mandatory for farmers. This had a significant influence on farm structures and practices of intensive livestock, pig and poultry systems by regulating the stocking rate and the management of nitrogen.

The successive EU CAP reforms led to modernization of the sector, farm size increase, a dramatic decrease in the farming population, specialization of production, intensification of grassland and animal husbandry, increase in the volume of production, increase of grassland and animal yields, reduction in legume use, reduction in grassland area and its proportion in the AA, and reduction in the diversity of landscape, grassland species and communities, and domestic animal breeds. The implementation of milk quotas reduced the number of dairy cows, which led to a stocking rate decrease in some cases or the development of suckling cows systems in complement to dairy systems in most cases.

The livestock sector
Belgian agriculture is strongly oriented to meat and dairy production. The ranking of the country in the EU-27 is much higher for animal production (Table 12) than for area, illustrating the intensity of Belgian agricultural systems.

Table 12. Importance (ranking and size) of Belgian production in Europe for some selected parameters

The proportions of animals slaughtered were 58.9 percent pigs, 40.1 percent cattle, 0.3 percent sheep and goats, and 0.8 percent equines in 1980 (tonnes of live weight [LW]) and 74.8 percent pigs, 24.7 percent cattle, 0.2 percent sheep and goat, and 0.3 percent equines in 2007 (FPS Economy, 2008). The pig/cattle ratio thus increased greatly during this 27-year period, indicating that Belgian animal production evolved towards a system that is less ground-bounded. Between 1997 and 2007, there has been a decrease in beef production and an increase in pork production. Milk production has been fairly constant, owing to EU milk quotas restricting production (Table 13).

Table 13. Production of meat and milk in Belgium in 1997 and 2007

Since 1990, cattle and sheep numbers have decreased while pig and poultry numbers have increased. For all species, the numbers increased per farm. Between 1990 and 2004, the number of animals per farm grew from 56 to 86 head for cattle, from 332 to 785 head for pigs and from 1 851 to 6 575 head for poultry (Genot, 2005).

In 1999, the auto-sufficiency ratio of Belgian production was 148 percent for cattle, 234 percent for pigs, 157 percent for hens and 138 percent for eggs (Genot, 2005). Exports of meat and eggs are thus significant.

Table 14 presents the intensification levels in several animal production systems. These levels are considered as ‘medium’ for cattle and sheep and ‘high’ or ‘very high’ for pigs and poultry; they are defined in the legend of the table. Compared with the European average, even the cattle and sheep intensification levels are high or very high.

Table 14. Proportion (percentage) of farms and animal numbers per intensification level – total numbers of farms and animals

A high proportion of cereals and several agro-industrial by-products (sugar beet, chicory, oilseed rape, flax) are used as animal feed; forage crops and pastures are also used. The majority of Belgian plant production (72 percent in 1999) is transformed into animal products (Genot, 2005). Belgium is not self-sufficient in cereals, and massive cereal imports are necessary for animal feed. For potato, vegetable and sugar production, the rate of self-sufficiency is significantly higher than 100 percent (Genot, 2005) (Table 15).

Table 15. Self-sufficiency (percentage) in crop production in 1990 and 1999

Dairy and beef cattle production dominates the ruminant sector. The cattle herd total was approximately 2.6 million in 2007, almost evenly distributed between the two main regions (Table 16). The number of dairy cows reached half a million but has declined since the implementation of the milk quotas. As dairy performances per cow increased continuously for economic reasons and as a result of breeding efforts, the milk quota was fulfilled with a decreasing number of dairy cows. The number of suckling cows is similar to that of dairy cows and has increased significantly since the 1970s (from about 40 000 cows in 1970 to about 140 000 in 1980). Belgian stockbreeders have a strong passion for breeding based on meat and muscle genetic characteristics that can be observed in several species: poultry (e.g. Coucou of Malines), rabbit (e.g. Flemish Giant), sheep (e.g. Belgian Texel), pig (e.g. Piétrain), cattle (e.g. Belgian Blue) and even horses (the powerful Brabant (Brabançon) and Ardenne (Ardennais) draught horses). The number of suckling cows increased even faster after 1984. The decreasing number of dairy cows has been compensated by an increasing population of suckling cows. The number of farms raising suckling cows increased from 10 200 in 1982 to 18 600 in 1994. In the European CAP reform of 1992, it was decided to increase the premium for male bovines and suckling cows, thus further stimulating the production of beef. Between 1997 and 2007, dairy cow numbers still decreased by 19 percent (FPS Economy, 2009) while suckling cow numbers remained almost stable. Dairy cows are slightly more numerous in Flanders and suckling cows more numerous in Wallonie.

Table 16. Structure of the cattle population in Belgium and the two main regions in 2007

Dairy farming is traditionally concentrated in Lower and Central Belgium as well as in the Liège Grassland Region and Upper Ardenne. The Sandy Region, the Sandy-Loamy Region, the Loamy Region, the Kempen and the Liège Grassland Region include 76% of the national dairy cow population. Suckling cows are concentrated in Central and Higher Belgium (Namur and Luxemburg provinces). The Sandy-Loamy Region, the Loamy Region, the Condroz, the Famenne, the Ardenne and the Jurassic Region include 76% of the national suckling cow population. Dairy cows are thus more abundant in the North and the West of the country; suckling cows in the South and the East. Most dairy cows are bred at low altitude. In higher altitudes, suckling cows are dominant. The Atlantic climate of the Flemish plain and the Kempen provides good grass-growing conditions: a relatively regular growth throughout a prolonged growing season.

Despite an intensive breeding programme and the general intensification of agriculture, Belgium still has some hardy breeds which are supported by subsidies from the agri-environmental programme. They include three cattle breeds (Belgian Red, White and Red, dual-purpose Belgian Blue), nine sheep breeds (Entre-Sambre-et-Meuse, Spotted Ardennais, Kempenish, Laeken Sheep, Mergelland, Belgian dairy sheep, Flemish flock sheep, Flemish sheep, Red Ardennais), two goat breeds (White, Chamoisée) and two horse breeds (Belgian or Brabant draught horses and Ardenne draught horses). [See Photos 6-8]

Photo 6. Belgian Blue cattle. Left: Bull. Source: Wallonie Elevages. Middle: Young cow. Source: Wallonie Elevages. Right: Suckling cow and calf. Source: Wallonie Elevages.

Photo 7. Other beef breeds. Left: Limousin cows and bull. Source: Wallonie Elevages. Right: Charolais cow. Source: Wallonie Elevages.

Photo 8. Dairy breeds in the Flemish plain. Left: Red and White cows by a canal. Source: ILVO - Merelbeke. Right: Holstein cows in an urbanized landscape with poplar (Populus) plantations. Source: ILVO - Merelbeke.

Dairy production
The dairy sector has undergone a significant transformation since the 1960s. Both the number of dairy farms and the number of dairy cows have fallen, especially since the middle of the 1980s after the implementation of the milk quotas that aimed at reducing the European milk overproduction and stabilizing the production. This trend is still ongoing. The dairy cow population was, for instance, 594 000 in 2000 and 500 000 in 2007. These two trends have been accompanied by an increase in the average milk yield per cow, from about 5 000 litres per cow in the late 1980s to about 7 000 litres per cow in 2009. Despite these major structural changes, the total milk output has remained relatively stable: 2 907 700 litres in 1997 and 2 878 600 in 2007. Dairy cow herd size is typically 50 to 60 on specialized farms (Kempen: 56; Liège Grassland Region: 57; national average: 39) (DGSEI, 2007).

The national average production per cow hides a huge variability of dairy performances between dual-purpose Belgian Blue cows that produce about 3 000 litres/cow, dual-purpose Red and White cows that yield about 4 500 litres/cow and Holstein Friesian cows that produce about 7 000 litres/cow. Average values for milk fat and protein contents are 4.0 percent and 3.2 percent, respectively.

The dairy cow population is dominated by Black and White Holstein-Friesians (about two-thirds of the dairy cow population in 2007) (DGSEI, 2008). The local Red and White cows were originally a dual-purpose (meat and milk) breed. They belonged to the Meuse-Rhin-Yser breed (a Dutch, German and Belgian breed) and were mainly distributed over the Flemish plain and the Kempen. They were progressively crossed with Red Holstein bulls and are now mainly specialized in milk production. Artificial insemination is the main method of reproduction in dairying.

Heifers are inseminated at 18–22 months and are kept in production for four to seven lactations. Bull calves are sold in the week after calving while heifer calves are kept on the farm. Only the best heifers are kept for the replacement of old cows, others are sold in calf. The replacement ratio is normally below 25 percent.

A high-yielding dairy cow ingests about 1 000 to 1 500 kg concentrate per year. With regard to green forages, the proportion of green grass, grass silage and maize silage in the diet varies considerably according to farming system and region.

Winter housing is either a free-stall with a slatted floor and a milking parlour or a tie-stall where milking is carried out in the individual stall. The first type of building is slurry-based, the second one FYM-based.

Dairy farms are usually smaller than beef farms. Their income per hectare is higher.

Beef production
Suckler herd size varies considerably according to the region and the farming system. Beef farms have about 50 cows on average (Ardenne: 52; national average: 28) (DGSEI, 2007).

Among meat breeds, the Belgian Blue, a native breed of Upper and Central Belgium, is largely dominant. However, French beef breed numbers are increasing, especially the Limousin (Table 17).

Table 17. Distribution of cattle breeds in Belgium and the two main regions in 2007

Cattle registration and organized breeding started at the end of the nineteenth century from a heterogeneous population (Hanset, 1998). At that time, Black and White, Red and White and Shorthorn bloods largely influenced the Belgian cattle population. The Shorthorn carried the roan gene (R+ gene) and produced the blue phenotype when crossed with black cattle (EE gene). The White-Blue breed of Central and Upper Belgium emerged slowly from this blend while discarding the propensity inherited from its Shorthorn ancestor to lay down an excess of fat. The ‘breed of Central and Upper Belgium’ was first a dual-purpose breed (annual dairy production of about 4 000 litres per cow). At the end of the 1950s and in the 1960s, selection for meat was adopted by an increasing number of breeders. The Belgian Blue Herd-Book was created in 1973. The double-muscle character (mh gene located on chromosome 2) was progressively fixed in the vast majority of the population. Only a small proportion of the present population is still dual-purpose. The present day Belgian Blue has retained from its ancestors an early maturity, a quiet temperament and the colour polymorphism of the Shorthorn (white, roan, red) but with a black background instead of the red of the Shorthorn. As a consequence of an excellent conformation owing to the double-muscle character, the slaughter yield is very high (> 70 percent). Up to 80 percent of the carcass can be used as meat; bones are fine. The meat is extremely tender and lean. Age at first calving is 30 months and even 25–26 months in the most intensive systems. Almost all calvings require caesareans. The average calf weight at birth is 44 kg. Average growth index of bulls is 1.539 kg/day and average feeding index is 5.109. Fattened bulls can be slaughtered as young as 17 months. Belgian Blue bulls are increasingly popular worldwide for industrial crosses. Calving is usually natural when Belgian Blue bulls are used on dairy cows. Cross-bred animals have improved body composition: more muscle, less fat, better lean/fat and lean/bone ratios. The breed has shortcomings: little maternal instinct of cows, lack of hardiness, little flavour.

Mating is mostly natural in beef production. During the grazing period, each suckling cow herd on the farm includes a bull, including the herd of the oldest heifers. Calving is between December and March. Male and female calves suckle their mother who produces about 1 000 to 2 000 litres per year. Winter housing is either free-stall or a tie-stall. Both are bedded with straw and produce FYM. In tie-stalls, calves are kept in pens and have access to their mother for a limited time per day. There is an increasing trend to separate calves and cows after calving in Belgian Blue systems. Calves are then fed artificially with powdered milk and later with roughage and concentrates. Young bulls never graze in this system since they are kept indoors until the end of the fattening period. In traditional systems, cows and calves start to graze together in April–May. Calves receive milk, graze and have access to a feeding system for concentrates at pasture. Cows only graze. Calf daily gains on pasture are high, more than 1 000 g/day. In the southeast of the country, especially in Ardenne, Famenne and the Jurassic Region, most bull calves are sold between July and September. They are fattened in Lower and Central Belgium or in Condroz . Farms in other regions fatten their own bulls. Fattening is usually more profitable than breeding. Most young females are kept for reproduction since the replacement ratio is high in Belgian Blue cows. After three to four caesareans, cows have to be culled. They are slaughtered after a short fattening period.

Sheep and other ruminants

Sheep are raised in small numbers (about 151 000 in total) mainly for meat (Table 18). Sheep breeding is often a secondary or hobby activity. Sheep are housed in winter. Lambing is between February and April for the production of ‘herbage lambs’ or ‘grey lambs’. In this case, ewes and lambs graze together from April to August or September. Lambs can receive concentrates in grassland for a short period before slaughter at the end of summer. Dry ewes graze till the middle of November. Texel and Bleu du Maine are the main breeds used for this purpose. Suffolk and Hampshire and their crosses are regularly used for the production of ‘Easter lambs’ or ‘white lambs’. In this system, lambing occurs indoors in December–January. Lambs can suckle and also receive concentrates. They are slaughtered at three months old, in April, so never graze. Ewes graze from April to November. Other breeds are used for a hobby and for nature reserve management.

There are only about 29 000 goats. Among the curiosities in this part of Europe, American bison (579) and red deer (3 300) can be cited. Most bison are raised in Wallonie, in the Ardennes, and most deer in Flanders (Table 18).

Table 18. Structure of other animal populations in Belgium and the two main regions in 2007

Organic farming and livestock
In the Flemish Region the area under organic farming (OF) increased regularly up to 4 026 ha in 2001 (Table 19). Between 2002 and 2005, the total area and the number of organic farms decreased noticeably. In 2006, this trend was halted by a slight increase in the organically farmed area to 3 836 ha in 2007. This area corresponds to only about 0.62 percent of the total AA in Flanders. Several socio-economic factors underpin the stagnation since 2000. First, organic farmers have had difficulty in marketing their products. Second, the traditional farmers’ advisory circuits did not support OF. The stagnation of OF in Flanders since 2001 sharply contrasts with its strong increase in other regions of the EU-15.

In the Walloon Region, the number of organic farms and their total AA are much more on the rise, especially since 1996 (Table 19). At the end of 2007, OF covered 29 200 ha, or 3.9 percent of the Walloon AA. This proportion is close to the EU-15 average: 4.1 percent of the AA in 2005. The recent increase in the extent of OF in 2006 and 2007 can be explained by the implementation of a new regional support scheme that provides more generous subsidies and an increased awareness of producers and consumers. In 2007, 73 percent of the Belgian organic farms and 88 percent of the Belgian OF area were in the Walloon Region.

Table 19. Evolution of the number of organic farms and their area in Belgium, the Flemish and the Walloon Regions between 1987 and 2007

In the Walloon Region, in 2008, 86 percent of organic areas were covered by pastures, 12 percent by cereals, and about 2 percent by fruit and vegetables. Most organic farms are livestock farms and are located in the grassland-specialized regions of the southeast. The main organic production is beef and milk. The total number of organic cattle amounted to 1.3 percent of the total. The beef market was strongly influenced by the dioxin food crisis of 1999 and bovine spongiform encephalopathy (BSE) in 2000–2001 that boosted the market by a factor of ten. Organic dairy production represented 2.5 percent of the total in 2004. The majority of organic milk is sold to large dairy factories; only 3 percent of the total is marketed in short marketing chains.

In arable areas, the development of OF is limited by several factors. Sugar beet and sugar chicory are very profitable crops. Roots of these plants are sold to large sugar factories that do not pay an extra price for organic products. Their mechanical weeding and biological pest control are difficult. Wheat yields are about 30 percent lower in OF compared with conventional farming. Inexpensive organic wheat grain or flour is available on the international market. Nitrogen supply to arable crops is very difficult without integrating crops and livestock. Annual pulse and vegetable legumes (Pisum sativum, Phaseolus vulgaris) and annual green manures (ex.: Vicia faba, V. sativa, Trifolium spp.) may leave a residue of nitrogen in soils for the following crops but their input is not likely to be sufficient under most conditions. Only perennial forage legumes can fix enough nitrogen for the needs of arable crops. They should occupy about half of the arable area, which is rarely possible with annual legumes and green manures. Since selling hay or silage of red clover or lucerne-based mixtures is limited to niche markets (horse-riding or horse-breeding centres, for instance), growing perennial forage legumes is difficult in arable farms. In contrast, on livestock farms, there are few problems of weeding and pest and disease control. Rumex obtusifolius is the main problem but it can realistically be solved by a combination of several strategies. Nitrogen fertilizers can be replaced by the use of Trifolium repens in grazed pastures and T. pratense or Medicago sativa in mown pastures. These plants can achieve high levels of biological N fixation. This nitrogen is transferred to annual crops by plant residues in soil (senescent organs, stubble, roots), or by animal manure (slurry, FYM). Growing silage maize in complement with pastures is easy in OF since mechanical weeding is possible and this crop has almost no important pest or disease in Belgium. With sufficient organic fertilization, organic maize yields are comparable with conventional yields.

The yields of temporary cutting pastures based on organic legume mixtures are the same as from N fertilized grass mixtures. The stocking rate possible on organic grazed pastures is only about 20 percent lower compared with N fertilized swards. Animal disease control must be implemented through integrated strategies, for instance for intestine parasite control in young animals, but solutions exist even if they are more complex than in conventional farming. In cases of severe disease attack, organic farmers are allowed, under strict control, to use conventional medicine for curing their animals, including antibiotics for the treatment of mastitis in dairy cows, for example. The limited grassland yield differences and the other inconveniences associated with OF in livestock farms are compensated by subsidies and possibly by higher prices. Product transformation, ­ for instance the processing of fresh milk for the production of cheese, yoghurt and ice cream ­ is relatively frequent in organic farms. Direct sale of products is also much more widespread than in conventional farms. Both strategies are efficient for increasing farmers’ income. These elements confirm that conversion to OF is realistic and relatively easy in livestock farms. It must be noted that the majority of organic beef is sold on the conventional market. Consequently, organic breeders cannot get higher prices for their meat. This is due to a lack of organization of the meat marketing and processing chain, which should be improved in the future. Organic milk is usually sold at a higher price by dairy factories compared with conventional milk. In the long term, organic farmers who process their milk into higher added value products will probably be more secure and better rewarded for their work.

OF is protected in Belgium by the ‘Biogarantie’ label. Within the framework of European Directives, this label is based on strict specifications. Certification of products is provided by independent, approved organizations (Ecocert and Blik) on the basis of regular controls on farms.

The rise in organic food demand by Belgian consumers only partially benefits Belgian organic farmers because of large imports of inexpensive organic products.

Veterinary and animal health problems

Identification, recording, monitoring and control.
In the late 1980s, Belgium was the first European country to implement a computer system of identification, recording and monitoring of its herds ­ the ‘SANITEL’ system. This centralized system monitors each animal, its responsible owner or keeper, each movement outside the farm, and each inventory on the farm. The system focused initially on cattle but was progressively extended to pigs, sheep, goats, deer and poultry. Each animal is recorded under an identification number and supervised by modern technologies. The system allows the detection of diseases and the implementation of strategies for the control of epizootics. Each bovine is identified within seven days after calving by two earmarks (plastic loops). A ‘passport’ is then assigned to the animal. The information on this passport, harmonized within a European Union context, includes birth date, sex and breed of the animal, and data on its male and female genitors. In addition, since 1988 the Walloon Region has developed a system of biological filing of its cattle herds, which was later followed by the Flemish Region. Simultaneously with the earmarking of a young animal, hairs are sampled for genetic characterization and later identification of the animal by DNA analysis. These data are stored in a hair sample database (the pilothèque). This database is used as a reference for solving problems associated with the identification of the filed animals. It can be used in certification chains (control of correspondence between animals and the derivative products) as well as for breeding programmes (control and reconstitution of pedigrees). By the combination of these two systems, the movement of individual animals and animal food products can be traced along the production, processing and marketing chain (ARSIA, 2008). The SANITEL system for the registration and identification of cattle was completely revised in 2008 in collaboration with the regional associations of animal health: DZG (Flemish Region) and Association Régionale de Santé et d’Identification Animales (ARSIA) (Walloon Region) (Houins, 2009).

Controls are organized by the Federal Agency for the Safety of the Food Chain (FASFC or Food Agency) that was established as a reaction to a series of food crises in Europe and in Belgium, in particular after the dioxin crisis. It was founded in 2000 and merged former departments of the Ministry of Agriculture (inter alia plants and raw materials; veterinary inspection and animal products), the Ministry of Public Health (inter alia general food inspection) and the Institute for Veterinary Inspection. The integration into one single body of all control and inspection services dealing with the food chain greatly improved the efficiency of the system. The task of the FASFC is to guarantee the safety and quality of the food chain, and protect human, animal and plant health. Its four major areas are: food safety, animal diseases, plant diseases and animal welfare. The FASFC performs its controls in a completely independent manner. Concretely, the agency controls, analyses and inspects foodstuffs and their raw materials at all stages of the food chain. It controls operators in the production, storage, transport, trade, import and export of foodstuffs and the places where these activities are located (Houins, 2009).

At slaughter, all cattle, sheep, goats, pigs and horses must be subjected to ante mortem and post mortem inspections. The primary objective of these inspections is to isolate and exclude from human consumption those meats that show pathological and anatomical abnormalities, that are contaminated by pathogenic agents, or that contain residues of veterinary medicines, pesticides or contaminants. The inspection is carried out by an official veterinarian (Houins, 2009).

Animal diseases and nutritional disorders
Belgium has officially been declared free of several diseases affecting cattle (bovine leucosis, brucellosis, tuberculosis) (Houins, 2009). The last occurrence of brucellosis was recorded in 2000 for bovines (Brucella abortus). Brucella melitensis has never been recorded in sheep in Belgium and the eradication of porcine brucellosis with Brucella suis in domestic pig dates back several decades (CERVA, 2005). The pig herd has been declared free of Aujeszky’s disease and no cases of Trichinella in pigs and equidae have been found (Houins, 2009).

Belgium was relatively free from BSE. During the maximum prevalence of the epidemic, only 40–50 cases of BSE were recorded per year in Belgium and the last case was recorded in October 2006 (Houins, 2009).

The first cases of bluetongue disease (serotype 8) appeared in Belgium in August 2006. In 2007, 6 870 cases were detected in cattle, sheep and goats. Numerous clinical problems and high mortality rates were observed in 2007 in cattle and sheep. The problems were mainly abortions and the birth of deformed and unhealthy calves. In 2008, a vaccination campaign was started and almost all cattle and sheep were vaccinated. Relatively low numbers of new cases were diagnosed thereafter. The vaccination campaign was supported by the FASFC, the budgetary fund for animal health and animal products, as well as by the European Commission. Around €10 million were paid to stock breeders out of a total campaign cost of €14 million. Particular vigilance was paid to the introduction of new serotypes in Belgium. In 2008, France was affected with serotype 1 and the Netherlands and Germany with serotype 6, two serotypes not yet discovered in Belgium (Houins, 2009).

Oral dosing of antihelmintics for gastro-intestinal parasites is organized on a routine basis on farms.

Grass tetany occurs most frequently in spring, when grass is short and growing rapidly, often during a cold period. Suckling cows are less susceptible than dairy cows. Dairy cows are also affected by milk fever (hypocalcaemia) just before or just after calving. The occurrence of this metabolic disorder is higher in spring when cows are grazing short grass with a high potassium content. Nitrate toxicity is irregularly observed, mainly in dairy cows, in autumn on pasture fertilized late and during periods of low night temperatures. Bloat is very rare as the average proportion of white clover in grazed sward is relatively low.

5. THE PASTURE RESOURCE

Crop and grassland production
Most cereals, many by-products of industrial crops, forage crops and pastures are used for feeding animals. The extent of the main crop and pasture types is shown in Table 20. Cereals cover about 24 percent of the Belgian AA with few differences between the Flemish and the Walloon Regions. In Wallonie, winter cereals are more important compared with Flanders whereas Flanders has more maize for grain. Grain maize is better adapted to Flanders because of the milder climate. Industrial crops cover about 13–14 percent of the AA in both regions. The Flemish Region, which has a higher proportion of sandy soils, is more specialized in potato and the Walloon Region in all other industrial crops. Among forage crops, forage maize and temporary pastures are dominant. The forage maize area has increased greatly since the 1960s, and now covers 18 percent of the Flemish AA and 7 percent of the Walloon AA. It seems that the evolution curve of its area is now reaching a plateau, at least in Wallonie. Fodder beet and forage legumes are not very important (about 0.5 percent AA). Temporary pastures are also more important in Flanders (8.5 percent AA) than in Wallonie (3.7 percent AA). They can be sown with pure grass mixtures but some sowings include grass–legume mixtures. In contrast, permanent pastures are much more important in Wallonie (45.7 percent AA) than in Flanders (26.6 percent AA). The extent of the total area of forage crops and permanent pastures is, however, very comparable in both regions (54 percent of the Flemish AA and 57 percent of the Walloon AA). Both forage production models are very different. The Flemish model is based 50 percent on forage crops, the Walloon model relies 80 percent on permanent pastures, reflecting the availability of land in both regions. Annual forage crops, while more expensive per kg of digestible organic matter, are more productive per ha. When land is scarce, as in the Flemish Region, their use becomes more necessary compared with regions where land is more abundant. Vegetables, orchards and glasshouses are all more important in Flanders, which reflects the intensity of labour in the Flemish agricultural system.

Table 20. Extent (area and proportion) of the main crop and grassland types in Belgium and the Flemish and Walloon Regions in 2007

Agriculture is highly intensive compared with most European and OECD countries. However, since the 1990s, input use per unit volume of output diminished. Between 1990–92 and 2002–04, the volume of inorganic fertilizers declined by about –15 percent for nitrogen and over –30 percent for phosphorus, pesticides by –19 percent and direct on-farm energy consumption by –6 percent (OECD, 2008).

Yields and total production of the main crop and grassland types are shown in Table 21. Typical winter cereal yields are about 8 tonnes of grain per ha but yields higher than 10 tonnes are common. Sugar beet produces about 70 tonnes of roots with a sugar content of about 15–16 percent, which corresponds to a production of about 10–11 tonnes sugar per ha. Potato can yield 40 to 50 tonnes of tubers per ha. Fodder beet yields are high, about 100 tonnes of roots per ha. Forage maize usually produces 50 tonnes fresh matter (FM) (whole plant) per ha (about 15 tonnes dry matter (DM)/ha). Average yields of cut pastures range between 6 and 7 tonnes DM per ha, but these pastures are also partly grazed. Total annual yields are thus higher; they can be estimated at about 12–16 tonnes DM/ha for cut temporary pastures. Grazed permanent pasture yields are lower (8–12 tonnes DM/ha). These yields are among the highest of the European Union for non-irrigated crops and pastures (Peeters and Kopec, 1996).

Table 21. Yields and total production of the main crop and grassland types in Belgium in 2007

Crop/grassland	Nature of the harvest	Yield (100 kg/ha)	Production (tonnes)
Cereal for grain
Winter wheat	grain	79.2	1 565 671
Spring wheat	grain	54.1	11 671
Winter barley	grain	79.9	344 090
Spring barley	grain	51.6	17 767
Spelt	grain	67.0	67 909
Maize for grain	grain	120.0	698 899
Industrial crops
Sugar beet	root	693.3	5 730 543
Potato	tuber	469.4	3 189 817
Oilseed rape	grain	37.4	40 495
Chicory	root	424.5	379 217
Flax	straw	55.1	78 712
Forage crops
Fodder beet	root	993.7	327 513
Forage maize – whole plant	green mass	490.4	7 784 981
Protein pea	grain	31.0	3 142
Pastures
Cut temporary grassland (all cuts)	dry matter	73.5	577 176
Cut permanent grassland (all cuts)	dry matter	60.9	1 173 964
Source: DGSEI, 2008

Forage production systems
There are two main forage production systems in Belgium. The Flemish system is based on regularly resown pastures and on annual forage crops (temporary pastures and maize). The system of Wallonie is mainly based on permanent pastures. The proportion of permanent pastures in Wallonie is 46 percent AA, in Flanders only 28 percent AA. The highest proportions of pastures in the AA are mainly observed in low human population density areas (Provinces of Luxemburg, Namur and Liège) (Table 22).

Table 22. Population density and extent of grasslands (% AA) in January 2005

	Population density (inhabitants/km²)	Permanent grassland [% PG]	Permanent and temporary grasslands [% PG + TG]
Belgium	342	38	44
Flemish Region	447	28	37
Walloon Region	202	46	50
Brussels-Capital Region	6238	32	33
Province of Luxemburg	58	74	87
Province of Namur	124	43	46
Province of Liège	268	58	59
Province of Walloon Brabant	334	16	17
Province of Limburg	334	24	34
Province of Hainaut	340	29	31
Province of West Flanders	362	27	34
Province of East Flanders	463	33	40
Province of Flemish Brabant	493	25	28
Province of Antwerp	585	26	49
% PG = Permanent grassland/(Permanent grassland + Arable land) (%); % PG+TG = Permanent + Temporary grasslands/(Permanent grassland + Arable land) (%). Source: FPS Economy – DGSEI, 2009

Table 23 compares three contrasting regions. The Kempen has a high proportion (49.5 percent) of forage crops (maize and temporary pastures) and a small proportion of permanent pastures (23 percent) in the AA. Forage maize (28 percent AA) is slightly more important than temporary pastures (21 percent AA). It is mainly a dairy region (Red Holstein). The total stocking rate (dairy and suckling cows) calculated on the total of forage crops and permanent pastures is just slightly higher than those of the Liège Grassland Region or the Ardenne but the total stocking rate calculated on the permanent pasture area is significantly higher. This reflects the fact that temporary pastures are grazed to a large extent and that most conserved forages are produced from forage crops. Permanent pastures are almost exclusively grazed. The Liège Grassland Region is also a dairy region (Holstein Friesian) but is characterized by a strong dominance of permanent pastures in the AA (84 percent); forage crops do not reach 10 percent. Most conserved forages are produced on permanent pastures that are also grazed. Grazing and cutting are thus closely integrated. There is almost no temporary pasture (2 percent AA). Maize is cropped on suitable plots. The Ardenne is a beef production region (Belgian Blue); suckling cow stocking rate is much higher and dairy cow stocking rate much lower compared with the two previous regions. The total stocking rate is similar to that of the Liège Grassland Region. Permanent pastures are important and are mainly grazed. Conserved forages are mainly produced on temporary pastures (14 percent AA) that are only grazed at the end of the growing season. Grazing and cutting are thus largely separated. Maize is not important (4 percent AA) for climatic reasons.

Table 23. Comparison of three agro-ecological regions for structure of forage area and stocking rates

	Kempen	Liège Grassland Region	Ardenne
Forage area structure (% AA)
Forage maize	28	7	4
Temporary grassland	21	2	14
Forage crops (Forage maize + Temporary grassland)	50	9	17
Permanent pasture	23	84	76
Stocking rate (cow/ha)
Dairy cows/(Forage crops + Permanent grassland area)	1.14	0.96	0.22
Suckling cows/(Forage crops + Permanent grassland area)	0.31	0.29	0.99
Dairy + suckling cows/(Forage crops + Permanent grassland area)	1.45	1.24	1.21
Dairy + suckling cows /Permanent grassland area	4.52	1.38	1.49
Source: DGSEI, 2007

Pastures and forage crops
Table 24 summarizes typical Belgian yields of forage maize and fodder beet, compared with intensive cutting temporary pastures and grazed permanent pastures. Table 25 compares typical feeding values and intake characteristics of these forage crops in Belgium.

Table 24. Comparison of typical yields of the main forage crops in Belgium

	Forage maize	Fodder beet	Cut temporary grassland	Grazed permanent grassland
tonnes FM/ha	50	120	50	57
% DM	30	13-19	20-30	15-20
tonnes DM/ha	13-18	14-21	12-16	8-12
Source: Deprez et al., 2007.

Table 25. Comparison of typical feeding value and intake characteristics of the main forage crops in Belgium

	Forage maize	Fodder beet	Cut temporary grassland	Grazed permanent grassland
Energy/kg DM
UFL1	0.90	1.15	0.90	0.80–0.98
UFV	0.80	1.16	0.84	0.83–0.94
Protein (g/kg DM)
CP	70–85	50–100	120–170	120–240
DP	45–55	65	70–100	80–120
PDIA	20	10	25	27
PDIN	50	60	85	75–115
PDIE	65	85	75	80–110
Minerals (g/kg DM)
P	1.6–2.8	1.5–2.0	2.8–4.0	3.0–4.8
K	12–16	22	22–30	25–40
Na	0.05–1.0	1.0	3.0–4.0	1.0–2.0
Ca	2.5–4.4	2.0–2.5	3.0–6.0	5.5–9.4
Mg	1.5–1.8	1.5	1.5–3.0	1.3–2.5
Digestibility (%)	68–73	90–94	68–72	70–80
Intake (g DM/kg LW)	23–30	Excellent but must be limited (acetonemia)	16–20	20–33
1 UFL = Unité fourragère lait (Fodder unit milk) = Net energy content of 1 kg of barley for milk production (1700 kcal NEl); UFV = Unité fourragère viande (Fodder unit meat) = Net energy content of 1 kg of barley for meat production (1 820 kcal NEg); CP = Crude protein; DP = Digestible protein; PDIA = Protéines alimentaires digestibles dans l’intestin grêle = Feed protein ruminally undegraded and truly digested in the small intestine; PDIN = Protéines digestibles dans l’intestin grêle avec azote limitant = Protein truly digested and absorbed in the small intestine when a degradable N deficient diet is fed; PDIE = Protéines digestibles dans l’intestin grêle avec énergie limitante = Protein truly digested and absorbed in the small intestine when a ruminal fermentable energy deficient diet is fed. Source: Deprez et al., 2007.

Table 26 compares the profitability of the same forage crops and additionally of species-rich meadow hay. Annual crops and cutting pastures have similar costs per ha (slightly higher for grass silage) and much higher than grazed pastures and hay meadows. Temporary pastures and hay meadows have the highest costs per kg DM and per energy content and grazed pastures the lowest. All pastures and especially grazed pastures have the lowest costs per kg of crude protein.

Table 26. Parameters of forage crop profitability in Belgium in 2005

	Forage maize	Fodder beet	Cut temporary grassland	Grazed permanent grassland	Species-rich hay meadow
Farm rent (€/ha)	150	150	150	135	135
Installation costs (€/ha)
Seed bed preparation	125	125	120	-	-
Sowing	50	50	100	-	-
Seeds	165	125	75	-	-
Maintenance	-	-	-	150	75
Total	340	300	295	150	75
Inputs (€/ha)
Fertilizers	270	230	275	90	-
Herbicides	100	210	30	-	-
Insecticides		20
Total	370	460	305	90	0
Harvest costs (€/ha)
Cutting-ensiling	265	340	650	-	-
Plastics + silo	150	75	130	-	-
Total	415	415	780	0	135
Annual yields
tonnes DM/ha	17	18	14	10	4
UFL/tonne DM	940	1100	830	900	750
kg CP/tonne DM	70	75	130	140	100
Cost without subsidy (€)
per ha	1275	1325	1309	375	345
per 100 kg DM	7.50	7.36	9.35	3.75	8.63
per UFL	0.08	0.07	0.11	0.04	0.12
per kg CP	1.07	0.98	0.72	0.27	0.86
Source: Deprez et al., 2007

 

Description of the main permanent grassland and rangeland types

Belgian rangelands derive from the destruction of forests, particularly on nutrient-poor soils ­ either very acidic or calcareous. These soils are sometimes dry, sometimes very wet. They have thus intrinsic limitations for plant growth. After forest clearing, the ground was colonized by herbaceous plants or dwarf shrubs and grazed by hardy breeds of sheep, cattle and horses. Some were occasionally and temporarily cropped in the past in a shifting agriculture system. Rangeland grazing disappeared almost completely in Belgium at the beginning of the twentieth century. Many rangeland areas were planted with trees (e.g. Pinus sylvestris in the Kempen and in the Sandy-Loamy Region, Picea excelsa in the Ardennes, Pinus nigra in Famenne). Some small areas of these communities have persisted. They belong to the following communities: dry moors and acid grasslands of the Calluno-Genistion pilosae, Nardus stricta communities (Violo-Nardion and Violion caninae mainly), wet moors of Erica tetralix, calcareous grasslands of the Festuco-Brometalia (Box 1). The largest site of rangeland communities remains in the Natural Park and the Nature Reserve of the High Fens – a vast territory of 4 500 ha of moorland and peatland along the German border.

Permanent grasslands can have a very ancient origin. They usually arise from the spontaneous evolution of a grassland, sowing or, anciently, from the colonization of the ground after land clearing or ceasing cultivation. Numerous plant species appear in the sward and its final botanical composition depends on the original environmental characteristics and the management practices.

Extensive grasslands can include up to 50 species on 100 m². They are mostly grasslands that are cut with little or no fertilizer applied. Among their species, grasses are usually dominant but spectacular flowering dicotyledon species can be abundant. These last belong mainly to the Apiaceae, Asteraceae, Fabaceae, Polygonaceae and Ranunculaceae. The area of extensive grasslands decreased rapidly until the beginning of the 1960s; they are now rare in Belgium. Remaining species-rich grasslands are usually in a bad state of conservation. They are mainly classified in the Arrhenatherion, Calthion (Bromion racemosi), Cynosurion, Molinion and Polygono-Trisetion phytosociological alliances (Box 1). The Arrhenatherion, Molinion and Polygono-Trisetion are considered as NATURA 2000 habitats.

Intensive pastures represent almost all permanent pastures in Belgium. Their swards can include 15 - 20 species of higher plants on 1 ha, i.e. about ten species on 100 m². Among these, grasses dominate. The commonest are Agrostis stolonifera (on cool soils), Alopecurus pratensis (on cool soils), Dactylis glomerata, Holcus lanatus, Lolium perenne, Poa pratensis (especially on shallow or sandy soils that are dry during part of the growing season) and Poa trivialis. Some nitrophilous dicotyledons can be locally abundant: Cirsium spp., Ranuculus repens, Rumex crispus, R. obtusifolius, Stellaria media, Taraxacum spp. (especially in the Ardennes) and Urtica spp., R. obtusifolius is the main grassland weed. Cirsium arvense and Urtica dioica can also be problematic locally. The proportion of Trifolium repens in swards is usually low but it can be important when the defoliation frequency is high and usually when nitrogen fertilization is relatively low. Other legume species are rare in intensive permanent pastures. [See Photos 9-12] 

Photo 9. Grasses and legumes. Left: Perennial ryegrass (Lolium perenne) - white clover (Trifolium repens) mixture. Source: ILVO - Merelbeke. Right: Perennial ryegrass (Lolium perenne) spikes. Source: ILVO - Merelbeke.

Photo 10. Legumes and grasses Left: Red clover (Trifolium pratense). Source: ILVO - Merelbeke. Right: Grass - lucerne (Medicago sativa) mixture. Source: ILVO - Merelbeke.

Photo 11. Perennial ryegrass (Lolium perenne) - white clover (Trifolium repens) mixture grazed by dairy cows. Source: ILVO - Merelbeke.

Photo 12. Species-rich cutting meadow in a nature reserve. Source: S. Rouxhet

 

Box 1 describes the botanical composition of the main grassland and rangeland vegetation types. The Poo-Lolietum association is by far the most common type; it covers almost all the permanent grassland area. All other permanent grassland types, including the Lolio-Cynosuretum, are now rare. A significant proportion of the remaining agricultural surfaces of species-rich grasslands attracts subsidies from the agri-environmental programme. The majority of rangeland vegetation is protected in nature reserves where it can possibly be extensively grazed by re-introduced hardy breeds.

BOX 1 Grassland vegetation types 1. Arrhenatherion elatioris Cut grasslands of low an medium altitude. 1.1. Arrhenatheretum elatioris Arrhenatherum elatius cut grasslands of low altitude. Five communities: – Arrhenatheretum elatioris (medioeuropaeum), mesophilous, mesotrophic to meso-eutrophic, Crepis biennis and Arrhenatherum elatius grassland; – Alchemillo-Arrhenatheretum, mesophilous, mesotrophic, Alchemilla xanthochlora and Arrhenatherum elatius grassland; – Arrhenatheretum subatlanticum, mesophilous, meso-eutrophic to eutrophic, Bromus hordeaceus and Arrhenatherum elatius grassland; – Heracleo-Arrhenatheretum, mesophilous, meso-eutrophic to eutrophic, Heracleum sphondylium and Arrhenatherum elatius grassland; – Alopecuro-Arrhenatheretum, meso-hygrophilous, meso-eutrophic to eutrophic, Alopecurus pratensis and Arrhenatherum elatius grassland. Characteristic species are: Anthriscus sylvestris, Arrhenatherum elatius (dominant), Centaurea gr. jacea, Crepis biennis, Daucus carota, Galium mollugo, Helictotrichon pubescens, Heracleum sphondylium, Knautia arvensis, Leucanthemum vulgare, Pastinaca sativa, Pimpinella major, Rhinanthus angustifolius, R. minor, Tragopogon pratensis. Other representative species: Bromus hordeaceus, Campanula rapunculus, Geranium pratense, Lathyrus pratensis, Phleum pratense, Ranunculus acris, Rumex acetosa, Trifolium dubium. 1.2. Alchemillo-Trisetetum Festuca rubra and Alchemilla xanthochlora submountain (medium altitude) cutting grasslands. Mesophilous, meso-oligotrophic to mesotrophic grasslands. Characteristic species are: Alchemilla filicaulis, A. vulgaris (Syn.: A. xanthochlora), Festuca rubra (dominant), Geranium sylvaticum, Hypericum maculatum, Knautia arvensis, Lathyrus linifolius, Phyteuma nigrum, Polygonum bistorta, Sanguisorba officinalis, Trisetum flavescens. Other representative species: Anthriscus sylvestris, Arrhenatherum elatius, Bromus hordeaceus, Centaurea gr. jacea, Crepis biennis, Daucus carota, Galium mollugo, Heracleum sphondylium, Leucanthemum vulgare, Pastinaca sativa, Pimpinella major, Rhinantus angustifolius, R. minor, Tragopogon pratensis, Trifolium dubium. 2. Polygono-Trisetion Mountain moderately fertilized cut grasslands. Two communities: 2.1. Meo-Festucetum Meum athamanticum mountain cut grasslands. Mesophilous, meso-oligotrophic to mesotrophic grasslands. Characteristic species are: Centaurea nigra (Syn.: Centaurea gr. jacea), Knautia dispacifolia, Meum athamanticum (sometimes abundant), Poa chaixii. Other representative species: Alchemilla vulgaris (Syn.: A. xanthochlora), Hypericum maculatum, Knautia arvensis, Lathyrus linifolius, Phyteuma nigrum, Polygonum bistorta. 2.2. Geranio-Trisetetum Geranium sylvaticum and Trisetum flavescens mountain grasslands. Mesophilous, mesotrophic to meso-eutrophic grasslands. Characteristic species are: Centaurea nigra (Syn.: Centaurea gr. jacea), Geranium sylvaticum (can be dominant), Knautia dispacifolia, Poa chaixii, Trisetum flavescens (can be dominant). Other representative species: Alchemilla vulgaris (Syn.: A. xanthochlora), Festuca rubra, Hypericum maculatum, Knautia arvensis, Lathyrus linifolius, Phyteuma nigrum, Polygonum bistorta. 3. Molinion Wet cut grasslands. Characteristic species are: Agrostis canina, Carex panicea, C. pulicaris, Cirsium oleraceum (in the Jurassic Region), Juncus acutiflorus, J. conglomeratus, Molinia caerulea, Ophioglossum vulgatum, Parnassia palustris, Scorzonera humilis, Silaum silaus, Selinum carvifolia, Stachys officinalis, Succisa pratensis, Valeriana dioica. Other representative species: Carex echinata, C. flava s.l., C. hostiana, Luzula multiflora, Potentilla erecta, Serratula tinctoria and Sieglingia decumbens. Two community types: 3.1. Acidophilous type dominated by Molinia c. caerulea and Juncus acutiflorus. Other characteristic species are: Carex echinata, Gentiana pneumonanthe, Juncus acutiflorus, J. squarrosus, Nardus stricta, Pedicularis sylvatica. 3.2. Neutral to baciphilous type dominated by Molinia c. arundinacea and Silaum silaus. Other characteristic species are: Colchicum autumnale, Selinum carvifolia, Crepis paludosa, Sanguisorba officinalis, Silaum silaus, Juncus inflexus. 4. Calthion palustris (or Bromion racemosi) Wet, mesotrophic to meso-eutrophic cut grasslands. Six communities: 4.1. Colchico-Brometum: Bromus racemosus and Colchicum autumnale wet cut grassland 4.2. Deschampsieto-Polygonetum: Deschampsia cespitosa and Polygonum bistorta wet cut grassland 4.3. Juncetum filiformis: Juncus filiformis and possibly Polygonum bistorta wet cut grassland 4.4. Cirsieto-Angelicetum: Cirsium oleraceum and Angelica sylvestris wet cut grassland 4.5. Senecio-Brometum: Senecio aquaticus and Bromus racemosus wet cut grassland 4.6. Epilobio-Juncetum: Juncus effusus wet cut grassland. Characteristic species are: Agrostis canina, Angelica sylvestris, Bromus racemosus, Caltha palustris, Carex disticha, Cirsium oleraceum (Jurassic Region), Cirsium palustre, Dactylorhiza majalis, Epilobium parviflorum, Juncus acutiflorus, J. conglomeratus, J. effusus, Lotus uliginosus, Lychnis flos-cuculi, Mentha aquatica, Myosotis nemorosa, Polygonum bistorta, Scirpus sylvaticus, Senecio aquaticus, Stachys palustris, Valeriana dioica. 5. Cynosurion Grazed grasslands. 5.1. Meso-oligotrophic to eutrophic grazed grasslands 5.1.1. Lolio-Cynosuretum Intensively used Lolium perenne and Cynosurus cristatus grazed grasslands. Dominant species: Agrostis stolonifera (on cool soils), Cynosurus cristatus (if not too heavily fertilized), Dactylis glomerata, Holcus lanatus, Lolium perenne (dominant), Phleum pratense, Poa pratensis (on shallow or dry soils mainly), P. trivialis, Ranunculus repens, Trifolium repens (Rumex obtusifolius and Rumex crispus). Few dicotyledons except Bellis perennis, Ranuculus repens, Rumex crispus, R. obtusifolius, Taraxacum spp. and Trifolium repens. 5.2. Extensive, not or little fertilized grazed grasslands Three communities: 5.2.1. Festuco-Cynosuretum Festuca rubra and Cynosurus cristatus extensive grazed grasslands. Characteristic species are: Agrostis capillaris, Anthoxanthum odoratum, Briza media, Campanula rotundifolia, Cynosurus cristatus, Festuca rubra (dominant), Holcus lanatus, Hypochoeris radicata, Leontodon autumnalis, Luzula campestris, Ranunculus bulbosus, Trifolium repens. On the nutrient-poorest soils: Briza media, Danthonia decumbens, Hypochoeris radicata, Luzula campestris, L. multiflora, Pimpinella saxifraga, Potentilla erecta. Possible and occasional presence of Nardus stricta. 5.2.2. Galio-Trifolietum Galium verum and Trifolium repens extensive grazed grasslands. Dry soils. Characteristic species are: Dactylis glomerata, Galium verum, Medicago lupulina, Plantago media, Poa pratensis, Primula veris, Sanguisorba minor, Trifolium repens, Trisetum flavescens. 5.2.3. Junco-Cynosuretum Juncus spp. and Cynosurus cristatus extensive grazed grasslands. Wet soils. Characteristic species are: Achillea ptarmica, Carex ovalis, C. panicea, Cirsium palustre (abundant), Cynosurus cristatus, Equisetum palustre, Galium uliginosum, Holcus lanatus, Juncus acutiflorus, J. conglomeratus, J. effusus, Lotus uliginosus, Lychnis flos-cuculi, Molinia caerulea, Myosotis scorpioides, Succisa pratensis, Trifolium repens. 6. Poo-Lolietum Intensive, heavily fertilized, grazed Lolium perenne grasslands. Dominant species are: Agrostis stolonifera (on cool soils), Alopecurus pratensis (on cool soils), Dactylis glomerata, Holcus lanatus, Lolium perenne (dominant), Poa pratensis (on shallow or dry soils mainly), P. trivialis. Few dicotyledon species except Ranuculus repens, Rumex crispus, R. obtusifolius, Taraxacum spp. and Trifolium repens. Rangeland vegetation types 7. Calluno-Genistion pilosae Dry moors. Characteristic species are: Calluna vulgaris, Carex pilulifera, Danthonia decumbens, Genista pilosa, Lycopodium clavatum, Polygala serpyllifolia, Vaccinium vitis-idaea. 7.1. Calluno-Vaccinietum vitis-idaeae Vaccinium and Calluna submountain dry moors. Characteristic species are: Calluna vulgaris (dominant), Vaccinium myrtillus, V. uliginosum, V. vitis idaea (Vaccinium spp. co-dominant with Calluna vulgaris). Other representative species: Antennaria dioica, Anthoxatum odoratum, Arnica montana, Cytisus scoparius, Deschampsia flexuosa, Festuca filiformis, F. rubra ssp., Galium saxatile, Genista anglica, Hieracium umbellatum, Holcus mollis, Luzula multiflora, Meum athamanticum, Molinia careulea, Nardus stricta, Potentilla erecta, Rumex acetosella. 7.2. Calluno-Genistetum Calluna and Genista subatlantic dry moors. Other representative species: Agrostis capillaris, Aira praecox, Antennaria dioica, Carex arenaria, Cladonia spp., Festuca filiformis, Galium saxatile, Hieracium pilosella, Luzula campestris, L. multiflora, Nardus stricta, Potentilla erecta, Rumex acetosella. 7.3. Calluno-Antennarietum Calluna mesotrophic moors. Characteristic species: Carex flacca. Other representative species: Carlina vulgaris, Euphorbia cyparissias, Fragaria vesca, Genista tinctoria, Genistella sagittalis, Hieracium pilosella, H. umbellatum, Hypericum pulchrum, Orchis mascula, Platanthera bifolia, Prunus spinosa, Rosa canina, Silaus pratensis, Succisa pratensis, Viola canina. 7.4. Deschampsia flexuosa acid grasslands 8. Nardus stricta communities pecies-rich on siliceous substrate of mountain and submountain areas. Characteristic species are: Galium saxatile, Lathyrus linifolius, Nardus stricta (dominant). Four community groups: 8.1. Nardetalia: Atlantic Nardus and related communities. 8.2. Violion caninae, Polygalo serpyllifoliae-Nardetum, Lathyro montani-Nardetum, Centaureo nigrae-Meetum athamantici: meso-hygrophilous Nardus communities. 8.3. Violo-Nardion, Polygalo serpyllifoliae-Nardetum and Lathyro montani - Nardetum: lowland and submountain Nardus communities with Polygala serpyllifolia and/or Lathyrus linifolius. 8.4. Violo-Nardion: Centaureo nigrae - Meetum athamantici: mountain Nardus communities.with Meum athamanticum and Centaurea nigra. Characteristic species are: Centaurea nigra, Meum athamanticum, Phyteuma nigrum, Sanguisorba officinalis. 9. Erica tetralix northern atlantic wet moors Characteristic species are: Calluna vulgaris, Drosera rotundifolia, Erica tetralix (dominant), Eriophorum angustifolium, Gentiana pneumonanthe, Lycopodiella inundata (rare), Polytrichum longisetum, Scirpus cespitosus, Vaccinium uliginosum. 9.1. Erica tetralix wet moors Wet to peat moors dominated by Erica tetralix or Calluna vulgaris, and including species such as Drosera rotundifolia, Juncus squarrosus, Pedicularis sylvatica and peat mosses like Sphagnum compactum, S. molle and S. tenellum. Molinia caerulea, Scirpus cespitosus subsp. germanicus, Vaccinium uliginosum can also be present. 9.2. Erica tetralix moors on wet sand Other representative species: Carex pilulifera, Drosera intermedia, Juncus squarrosus, Rynchospora alba, R. fusca, Sphagnum compactum. 9.3. Vaccinium and Erica tetralix peat moors Usually dominated by Vaccinium spp., Erica tetralix and sometimes by Calluna vulgaris, often in combination with Molinia caerulea, and peat mosses: Sphagnum compactum, S. molle, S. tenellum. Other species can include: Carex panicea, Dicranum scoparium, Drosera rotundifolia, Gentiana pneumonanthe, Juncus squarrosus, Leucobryum glaucum, Rhacomitrium lanuginosum, Scirpus cespitosus subsp. germanicus. 9.4. Calluna vulgaris and Scirpus cespitosus wet moors Other representative species: Carex pilulifera, Eriophorum vaginatum, Galium saxatile, Nardus stricta, Potentilla erecta, Scirpus cespitosus subsp. germanicus, Sphagnum compactum, S. molle, S. tenellum. 10. Dry, calcareous grasslands (Festuco-Brometalia) 10.1. Mesobromion erecti Mesophilous to meso-xerophilous calcareous grasslands. Characteristic species are: Ajuga genevensis, Allium oleraceum, Anacamptis pyramidalis, Anthyllis vulneraria, Asperula cynanchica, Avenula pratensis, Brachypodium pinnatum (can be dominant), Bromus erectus (dominant), Campanula glomerata, Carex caryophyllea, C. flacca, C. tomentosa, Carlina vulgaris, Centaurea scabiosa, Cirsium acaule, Eryngium campestre, Festuca lemanii, Galium pumilum, Gentiana cruciata, Gentianella ciliata, G. germanica, Gymnadenia conopsea, G. odoratissima, Helianthemum nummularium, Himantoglossum hircinum, Hippocrepis comosa, Koeleria macrantha, K. pyramidata, Linum catharticum, Medicago lupulina, Onobrychis viciifolia, Ononis repens, O. spinosa, Ophrys apifera, O. fuciflora, O. insectifera, Orchis anthropophorum, O. militaris, O. simia, O. ustulata, Plantago media, Polygala comosa, Potentilla neumanniana, Primula veris, Prunella laciniata, Salvia pratensis, Sanguisorba minor, Scabiosa columbaria, Sesleria caerulea, Teucrium chamaedrys, Thymus praecox subsp. praecox, Trifolium montanum. 10.2. Xerobromion erecti Xerophilous calcareous grasslands. Characteristic species are: Allium sphaerocephalon, Arabis hirsuta, Artemisia alba, Aster linosyris, Carex humilis, Dianthus carthusianorum, Fumana procumbens, Globularia bisnagarica, Helianthemum apenninum, H. nummularium, Hippocrepis comosa, Linum leonii, L. tenuifolium, Melica ciliata, Orobanche teucrii, Phleum phleoides, Potentilla neumanniana, Pulsatilla vulgaris, Sesleria caerulea, Stachys recta, Teucrium chamaedrys, T. montanum, Thlaspi montanum, Veronica prostrata subsp. scheereri. 10.3. Koelerion-Phleion phleoidis Xerophilous grasslands on calcareo-siliceous rocks. Artemisia campestris, Aster linosyris, Campanula patula, Festuca heteropachys (dominant), Hieracium peleterianum, Lychnis viscaria, Phleum phleoides, Potentilla rupestris, Silene armeria. 10.4. Festucion pallentis Festuca pallens dry grasslands on calcareous rocks. Characteristic species are: Artemisia alba, Biscutella laevigata subsp. varia, Dianthus gratianopolitanus, Draba aizoides var. montana, Festuca pallens (dominant), Hieracium glaucinum, H. vogesiacum, Sesleria caerulea.

Many species of bird nest in grasslands or on their margins or feed therein (Peeters et al., 2009).

Temporary pastures

Temporary pastures are managed either mainly by cutting (a hay cut and aftermath grazing or 2–3 silage cuts + grazing or 3–4 silage cuts), or mainly by grazing (grazing only or grazing + 1–2 silage cuts). Typical temporary pastures for cutting are sown for one to five years. In recent years, a trend of keeping these pastures for a longer period has been observed in grassland specialized regions of the southeast. In the Flemish Region, between 1960 and 1990, mainly grazed temporary pastures were established for about five years and re-sown. After 1990, there has also been a trend to convert them into permanent pastures with or without regular oversowing.

Mixtures for mainly cut pastures include one to five species. The most widespread species are Lolium perenne and L. multiflorum (about 90 percent of the seed market). Other sown species are grasses: Dactylis glomerata, Festuca pratensis, F. arundinacea, Phleum pratense, Poa pratensis, hybrids of Lolium spp. and sometimes Lolium x Festuca; and legumes: Medicago sativa, Trifolium pratense, T. repens. In the Ardennes, a traditional mixture includes F. pratensis, L. perenne, P. pratense and T. pratense. The frost resistance of F. pratensis and P. pratense is appreciated above 500 m asl. These species are added into this mixture as an insurance against climatic variations. Simple mixtures of L. perenne, usually pure or mixed with T. pratense and/or T. repens, and simple mixtures of L. multiflorum, mixed or not with T. pratense, are the most frequent for the establishment of pastures harvested for silage making. The mixture of D. glomerata and Medicago sativa was almost abandoned in the second half of the twentieth century but recently its potential has been recognized again by farmers from Low and Central Belgium, as well as those from Condroz and the Jurassic Region.

Mixtures for mainly grazed pastures are even simpler. The main species are Lolium perenne and Trifolium repens. Poa pratensis and Phleum pratense are sometimes added into the mixtures. P. pratensis is mainly used on sandy soils and for horse pastures. P. pratense is mainly appreciated in the Ardennes; however, its persistence in grazing is low.

Single species mixtures of Lolium perenne, sold by commercial companies, are made up of three to six cultivars. Mixtures can include up to eight cultivars for the grass/clover associations.

Some species appear spontaneously in sown swards of temporary pastures, such as: Bromus hordeaceus, Elymus repens, Poa trivialis, Rumex spp., Stellaria media and Taraxacum spp. The total number of species is, however, usually lower compared with intensive permanent pastures.

The main mixtures used in Belgium are summarized in Box 2.

BOX 2 Examples of the most common mixtures for temporary and permanent pastures in Belgium Short-lived temporary pastures, mainly cut 0–1 year, for cutting for sowing in intercropping M1: 100% Lolium westerwoldicum. 2 years, for cutting M2: Lolium multiflorum, 100% pure-sown or 70% mixed with 30% Trifolium pratense (moderately acidic to neutral soil) 3 years, mainly for cutting M3: 100% Lolium perenne, early cultivars in the lowlands or intermediate cultivars at altitudes over 500 m; suitable for dairy cows. M4: 70% L. perenne, intermediate cultivars + 30% Phleum pratense for the Ardennes. M5: 45% L. perenne, early or intermediate cultivars + 25% Festuca pratensis + 30% P. pratense; all-purpose mixture in the Ardennes; used for animals, e.g. suckler cows, with moderate nutritional requirements. T. pratense (6 kg/ha) and possibly Trifolium repens (Hollandicum or Ladino cultivars) (3–4 kg/ha) are regularly added in mixtures M3 to M5. The addition of T. repens is mainly advantageous if the mixture is occasionally grazed. It can also make grass/red clover mixtures more persistent. M6: 40-60% Dactylis glomerata + 60-40% Medicago sativa for deep, neutral or basic soils Short-lived temporary (five years) or permanent pastures, grazed or mixed use (grazing and cutting) Everywhere on fertile soil, normally drained M7: 85% L. perenne (late cultivars or late + intermediate cultivars) +/- 15% T. repens On sandy and shallow soils mainly M8: 60% L. perenne (late cultivars or late + intermediate cultivars) + 30% Poa pratensis +/- 10% T. repens In Ardenne M9: 50% L. perenne (late cultivars or late + intermediate cultivars) + 20% P. pratense + 20% P. pratensis +/- 10% T. repens

Legumes in pasture swards
Trifolium pratense, T. repens and Medicago sativa are the three main forage legume species in Belgium. T. pratense and M. sativa are almost exclusively used in cut temporary pastures. T. repens is mainly used for sowing long-lived grazed swards but its larger cultivars are sometimes associated with short-term cutting mixtures. The capacity of N fixation and the high nutritive value and intake potential of these three legume species are appreciated by farmers. Both characteristics can reduce production costs and thus increase farmers’ income. However, areas under clovers and lucerne have decreased dramatically since the 1960s in Belgium; between 1990 and 2000 they decreased by 69 percent and 26 percent, respectively (DGSEI, 1990 and 2000). Belgian farmers acknowledge the theoretical interest of legumes but, in their intensive pasture production systems, they tend to prefer the use of nitrogen fertilization that provides important and regular yields, even if it leads to the sacrifice of legumes. This practice has been maintained by the relatively low prices of nitrogen fertilizers and by some drawbacks of legume species. Despite undeniable breeding progress, persistence remains a problem, especially for T. pratense. Slow growth and low nitrogen fixation in spring make grass/clover mixtures, especially grass/T. repens, less attractive than N-fertilized pure grass swards. Despite a low mortality rate, bloat risk induced by T. repens in grazed swards is often overestimated by farmers.

Forage legumes are one of the pillars of organic systems, not only for producing energy and protein-rich forages but also for replacing nitrogen fertilizers by biological N fixation. It is usually recognized that forage grass/legume mixtures must occupy about half of the arable area. T. repens must also be sufficiently abundant in grazed swards. The average target proportion is about 30 percent in the sward DM (40–50 percent in summer).

In Belgium, no significant annual yield differences were noted between Trifolium pratense and Medicago sativa sown in mixture with grasses on the same site. It can be concluded that the present cultivars of red clover have similar yields to lucerne cultivars. This was probably not the case in the past – lucerne had the reputation of yielding more than red clover. Very high annual yields up to 18 t/ha for lucerne and 21 t/ha for red clover can be recorded.

Medicago sativa is more persistent than Trifolium pratense but, in most cases, its period of high production does not exceed three to four years. In experiments in Central Belgium, average annual yields of lucerne ranged over 16.2, 15.6 and 17.2 tonnes DM/ha for the three production years in pure stands (Deprez et al., 2004a). Yields of up to 17.9 t DM/ha were harvested in the third year of production. In a mixture with grass, lucerne yielded on average 15.4 tonnes DM/ha in the first year and 15.2 in the second year of ley.

The major disadvantage of Trifolium pratense is its lack of persistence. It does not usually persist for more than two years, whereas temporary cut pastures are established for increasingly longer periods (three years and more). However, oversowing has proved to be a means to maintain a red clover population in a mixture, but this technique is not yet widely adopted by farmers. Moreover, new cultivars, in particular Swiss Mattenklee type cultivars of red clover, seem to be more persistent than older ones. In Belgium, very high annual yields of T. pratense were recorded in experiments (Deprez et al., 2004a). The maximum yields (tonnes DM/ha) for pure swards were recorded in Lower Belgium and were as high as 21.1 in A1 and 18.0 in A2; average yields (tonnes DM/ha) were 18.6 in A1 and 16.4 in A2 under the same conditions. For mixtures in less favourable conditions, maximum yields reached 17.3 in A1, 17.3 in A2 and 17.9 in A3. These high yields, persistent over three years of production, were achieved with Mattenklee cultivars. Average annual yields (tonnes DM/ha) of mixtures reached 16.4 in A1, 16.4 in A2 and 16.3 in A3 in Central Belgium and 14.1 in A1 and 9.0 in A2 in Ardenne. On many sites, a yield decrease is recorded in the second year of production and very low or nil yields are observed in the third year. Yields of red clover were significantly lower in Ardenne than in Lower and Central Belgium; this may be due to the climate conditions (severe frost in winter) of the Ardennes and to the presence of a fungus or a complex of fungi. Fungal rots are a major cause of deterioration of red clover tap roots (Peeters, Parente and Le Gall, 2006).

New aggressive cultivars of Trifolium repens are now available on the market. They are much more persistent than conventional old cultivars in grazed mixtures with Lolium perenne as the companion grass. The conditions for balanced and persistent legume/grass mixtures are also much better understood. An adapted management based on current knowledge can maintain white clover for long periods in the sward. White clover is almost always used in a mixture with grasses and the contribution of white clover to total yield is highly variable. Annual production of grazed white clover/grass mixtures under good environmental conditions is probably about 8–12 tonnes DM/ha.

In trials, red clover and lucerne-based swards were compared with fertilized, pure perennial ryegrass (Lolium perenne) in the loamy region. Annual dry matter yields up to 15.1 tonnes/ha for red clover and 15.4 tonnes/ha for lucerne were obtained without nitrogen fertilization. Those yields are only achievable with high annual fertilization of 300–400 kg N/ha in the case of a pure ryegrass sward (Deprez et al., 2004b).

Pasture management and forage conservation

Pasture management systems are quite diverse in Belgium though intensive almost everywhere.

Temporary or permanent pastures
In the Flemish Region, grazed and multiple-use pastures are traditionally resown, mostly in the same place, every 4–5 years. One-fourth or one-fifth of the pasture area of a farm is renovated at the end of summer by this technique. In the 1960–70s, sowing was preceded by ploughing. With the introduction of glyphosate, ploughing is no longer used. About three weeks after the chemical destruction of the sward, seeds are sown by direct drilling. However, this technique has led to regular failures of germination that are not totally understood. A toxicity induced by soil organic matter decomposition is believed to be responsible. The application of slurry before sowing greatly reduces the occurrence of poor seedling establishment. This system is still important in this region but there is a trend to increase the lifetime of the sward and to use more permanent pastures.

In the Walloon Region, grassland production is mainly based on permanent swards. The most intensive dairy producers of the Herve country and the Upper Ardennes prefer also to keep permanent grasslands but they try to improve their botanical composition by the introduction of Lolium perenne seeds. This is achieved by two techniques: either by regular over-sowing of seeds mixed with slurry or nitrogen fertilizers, or by over-drilling into the existing sward. In the first case, low amounts of seeds are spread on the soil surface, for instance after a silage cut. The germination success is low but regular seed applications can be successful in the long term. In the second case, seeds are injected into the soil by seeders designed for grassland renovation. This technique developed very much in the 1980s. In the 1990s, it appeared that simple harrows can be very efficient and at a lower cost than specialized over-drilling seeders. They can break the sod of the sward and create bare ground on top of which seeds are sprayed. A roller is then used to ensure a close contact between the seeds and the soil. The success of this last technique and that of over-drilling is variable. It depends on many factors. A probability of success of 40% is often cited. The techniques described above can be very useful but it is recognized that the best tool for improving sward quality is the grazing animal. Optimal grazing techniques are thus essential particularly after sward renovations.

Temporary pastures that are used for the production of conserved forages are normally within crop rotations. In Lower and Central Belgium, these short-term swards are sown either with Lolium perenne for three to four years, or with L. multiflorum for one (ssp. Westerwoldicum) or two years (ssp. multiflorum). Mixtures are usually composed of several cultivars. Early cultivars of L. perenne are liked for silage making; they are mixed with intermediate cultivars if the sward is sometimes grazed. Hybrid ryegrasses (L. perenne x L. multiflorum) are sometimes used for trying to combine the persistence and the excellent feeding quality of L. perenne with the high-yielding potential of L. multiflorum. In the last 40 years of the twentieth century, the use of grass/legume mixtures declined considerably but, at the beginning of the twenty-first century, there is a renewed interest in these types of mixtures, especially for Medicago sativa and its mixture with Dactylis glomerata. Some farmers have understood that these forages are inexpensive sources of quality protein which can reduce fertilizer costs by biological nitrogen fixation. Some are even proud to reduce their soybean purchases to reduce their impact on the deforestation of the Amazonian rainforest by producing their own source of protein.

In Upper Belgium, cutting mixtures are usually more complex. They include intermediate and late varieties of L. perenne, and also regularly Phleum pratense, Festuca pratensis and Trifolium pratense. Early cultivars of L. perenne are not recommended in the Ardennes because they can suffer from late frosts in spring and are thus not very persistent. M. sativa and its mixtures are cropped in Condroz and in the Jurassic Region.

In the Ardennes, temporary pastures are still regularly sown in spring under a cover crop, e.g. a cereal. The forage seed mixture is sown just after the sowing of the spring cereal. The germination of the forage mixture occurs in April and its growth is very slow but forage plants survive in the understorey of the cereal. The cereal crop is normally harvested in August–September and the sward then grows rapidly and can be harvested for conservation in the autumn.

Fertilization
Nitrogen fertilization is high in all regions though limited by law. Throughout the Flemish Region, organic N fertilization from animal manure is limited to 170 kg/ha on pasture and most crops. The total mineral and organic nitrogen application must always be lower than 350 kg/ha per year on grassland but no N fertilization is allowed on forage legumes (125 to 275 kg/ha on arable land). For the Walloon Region, organic N fertilization is limited on grassland to 230 kg/ha per year (115 kg/ha on arable land). The total mineral and organic nitrogen application must always be lower than 350 kg/ha per year on grassland (250 kg/ha on arable land).

Phosphorus fertilization was also overestimated during the 30–40 years of the ‘blind intensification period’. Many Flemish soils are saturated by phosphorus and P leaching is even observed on sandy, easily leachable soils. On the loamy soils of Wallonie, P leaching does not occur but many soils are so rich in P that the cessation of P fertilization during 20 years of experiments did not lead to yield reductions. However, surface waters are polluted by soil phosphate run-off of mineral and organic P in both regions, which induces eutrophication of rivers and the reduction of aquatic biodiversity.

It has been calculated that P and K fertilization is no longer necessary on pastures in dairy farms because inputs of these nutrients by concentrates are sufficient to compensate for the outputs by meat and milk if the nutrient cycle of these chemical elements through spreading of slurry, urine and dung deposits during the grazing period is well managed. In specialized cutting pastures, exports of P and K by forage harvesting for conservation are important. They are increasingly compensated for by a concentration of the organic manure application on this type of sward. N fertilizers are still necessary on pastures (except on well managed grass/legume swards) but even the use of this nutrient has been strongly reduced by better information for farmers on the fertilization value of organic manures. These manures were too often considered as wastes during the 1960–90 period, and their fertilizer value was not taken into account. This led to an overuse of chemical N fertilization and pollution of the water tables and surface waters by nitrate. The European Nitrate Directive has helped greatly in stimulating farmers to calculate precisely the N balance in their farms and the N requirements of grassland plants. As a result, N and P fertilizations have been significantly reduced over the last ten years.

A dense network of soil analysis laboratories provides fertilization advice to farmers who call upon their services. In Wallonie, these laboratories are heavily subsidized by public authorities.

Production

Belgium is one of the areas of Europe where pasture production is the highest without irrigation (Peeters and Kopec, 1996). In cutting experiments (3–4 cuts/year), annual yields of Lolium perenne are about 12–16 tonnes DM/ha in Lower and Central Belgium (10–14 tonnes DM/ha in Ardenne). Annual yields of L. multiflorum are about 15–20 tonnes DM/ha in Lower and Central Belgium (12–16 tonnes DM/ha in Ardenne). In a frequent defoliation regime (four-week interval between cuts, average annual yields of about 9 tonnes DM/ha were recorded in the Ardennes (Peeters and Kopec, 1996). Under farm conditions, annual yields of grazed swards typically range between 6 and 12 tonnes DM/ha. In Lower and Central Belgium, annual yields of 10–12 tonnes DM/ha are common in grazed swards.

The average stocking rate is about 2.4 livestock units (LU)/ha of the forage area (temporary and permanent pastures + forage maize) in the Walloon Region and about 3.2 LU/ha in the Flemish Region.

In the 1970s, nitrogen fertilizers and concentrates were inexpensive and farmers tended to use them in increasing amounts. In 1984, milk quotas were implemented in the European Union. Farmers could no longer increase their income by a continuous increase of milk production per farm and therefore they tried to decrease their production costs. This has been achieved by better grazing management and a better use of concentrates. Annual milk production produced on the basis of green forage (grazed swards, grass and maize silages) increased from very low levels to 3 500–4 000 litres per cow.

Grazing systems
Grazing systems have evolved greatly over time. They could sometimes be quite complex in the 1960s; they are now much simpler. Strip grazing was used in permanent pastures for grazing forage crops established for a short period of some weeks between two main crops and for grazing longer-term temporary pastures. This system, though effective, is now rare because it requires a high labour input. Rotational grazing was the reference system in the 1960s and the 1970s. Mineral nitrogen fertilization has increased progressively since the 1960s. Silage cuts were generalized in this system in the 1970s. Over the same period, the intensive set stocking system was promoted, which was associated with higher nitrogen fertilization levels compared with rotational grazing. At the beginning, many farmers were afraid to spread mineral N fertilizers in the presence of cows in the paddock but they quickly understood that this technique was safe. Most systems are now intermediate between pure rotational systems including about ten paddocks for the dairy cow herd and pure set stocking systems with a single paddock. These intermediate systems are a way for farmers to combine the advantages of both systems. They try to minimize labour such as in set stocking systems and to optimize the flexibility of management that is characteristic of rotational grazing systems.

Dung spreading and refuse cuts were frequently adopted by farmers in rotational systems in the 1960s and the 1970s. Dung spreading was aimed at distributing nutrients on the whole paddock surface, thus optimizing the use of nutrients by grassland plants. It has been progressively abandoned since the mid-1970s when N fertilization increased because optimum use of nutrients became less crucial. Moreover, grazing rotation speed increased with N fertilization and, when dung was spread on the whole paddock surface at the end of the previous grazing period, a bad odour remained on the fresh grass when ready to graze at the next grazing period. This decreased dairy cow intake and farmers abandoned this technique. This effect was even stronger in set stocking systems. A refuse cut of grass stems was increasingly seen as a waste and a result of bad management of an inadequate stocking rate in spring. Farmers tried to reduce it to a minimum.

Herbicide treatments are restricted to pasture renovation and to the control of invading species such as Rumex spp. (mainly R. obtusifolius), Urtica spp. (mainly U. dioica) and Cirsium spp. (mainly C. arvense). The establishment of temporary pastures sometimes also requires the control of annual weeds by herbicides (mainly for the control of aggressive annual weeds such as Chenopodium album, Raphanus raphanistrum, Sinapis arvensis and Stellaria media). However, topping the young sward is usually sufficiently efficient to control annual weeds. R. obtusifolius must sometimes also be controlled just after sowing.

In dairy systems, the number of paddocks for dairy cows typically ranges from one to five in Lower and Central Belgium. In the dairy specialized regions of Upper Belgium, the Herve country and Upper Ardenne, rotational grazing is more strictly applied and the number of paddocks is higher, from 10 to 15. Three-day grazing periods per paddock and per grazing cycle are typical in this case. Two or three paddocks are usually devoted to heifers.

In beef systems, the organization of grazing is more complex because of the existence of several herds of suckling cows and their calves grazing in separate groups of paddocks. It is common to observe three to five suckling cow groups grazing in separated grazing circuits. Each group, including the group of tall heifers, is accompanied by a bull. In the Ardenne, rotational grazing is usually adopted. Each group grazes in a small number of paddocks (four to six) during 10 to 20 days in each grazing period. Cows can be moved from one group to another for several reasons, ­ for instance when their calves are weaned.

In beef systems, the calving period is concentrated between December and March. The cow–calf couples are formed in the shed. They start grazing in April or at the beginning of May in the Ardennes. Calves suckle the milk of their mother, start progressively to graze and also have access to concentrate distributors. Their daily growth is fast, about 1 000 g of LW or more. In the most intensive beef systems, calves are weaned at calving and are fed artificially with powdered milk. Young bulls never graze in this system. After the raising period, they are fattened in lots in sheds with quality silage and concentrates.

In the Loamy and the Sandy-Loamy Regions, most pastures are concentrated in the bottom of the valleys where arable farming is not possible. The stocking rate can then be very high in these paddocks that can be considered as ‘parking pastures’. The animals are, however, partly fed by grazing but also, to a large extent, by complementary feeding in the paddock: grass and/or maize silage and concentrates. This system is frequent with suckling cows.

In dairy systems, there is also an evolution towards continuous (ad libitum) access of dairy cows to maize and/or grass silage during the grazing season. As the dairy performance of cows is continuously increasing over time, farmers start to lose confidence in the potential of pastures to feed high-yielding dairy cows properly. During unfavourable weather periods, rainy and cold or sunny and warm, farmers do not want to take the risk of an intake decrease that could led to a decline of dairy production. They start to provide silage to this kind of animal. Moreover, farmers tend to use more maize silage at the expense of grass grazing and grass silage when dairy cow production is above a certain threshold (roughly above 6 000 litres/cow). Progressively, the access to silage is no longer restricted in many cases. This trend is reinforced by the adoption of a milking robot. Since cows return several times a day to the robot for milking, the grazing area that can be really grazed is greatly reduced. It is limited by the distance to the robot. Cows thus need to be supplemented continuously by green forages.

Forage conservation
Hay was by far the dominant type of conserved forage in the 1960s. Since the 1970s, grass silage has developed quickly as it produces higher quality forage and reduces the need for energy and protein rich concentrates. Silage is almost always pre-wilted in Belgium, which reduces the need for preservatives and prevents leaking. In dairy systems, grass is cut at an early stage, at about 3 to 5 tonnes DM/ha, and is conserved at 30–40 percent DM. In beef systems, grass is collected mainly as haylage at a later physiological stage, at a higher production level and at a higher DM content (about 50–60 percent).

Silage clamps quickly became the main conservation system in the 1970s. Tower silage developed, however, in the east of the country.

Since the 1980s, round bale silage has been very successful in grassland specialized regions. Notwithstanding its higher cost compared with clamp silage, this system was largely adopted because it is very flexible, especially in regions where plots are small, the pasture area of the farm is broken up and the climate is rainy. Big bale silage is typically produced from drier and slightly more mature grass than clamp silage. It is thus well adapted for conserving haylage for suckling cows.

Hay continues to be made in relatively small amounts (< 25 percent of harvested grassland forage) for providing fibrous feed to particular classes of stock, e.g. young stock, horses and sheep. It is also produced in the framework of agri-environmental measures where species-rich pastures and field margins must be cut late to improve vegetation diversity, allowing reseeding of many flowering plants and wildlife breeding. [See Photos 13-17] 

Photo 13. Silage making. Left: Cutting. Source: ILVO - Merelbeke. Right: Wilting in windrows. Source: ILVO - Merelbeke.

Photo 14. Silage making. Grass conditioning for speeding of wilting. Source: Wallonie Elevages.

Photo 15. Grass harvesting for silage. Left: Demonstration of machinery types. Source: ILVO - Merelbeke. Right: Self-propelled forage harvester filling a wagon pulled by tractor. Source: ILVO - Merelbeke.

Photo 16. Silage making. Left: Unloading of chopped grass onto a heap. Source: ILVO - Merelbeke. Right: Rolling and compressing grass in a pit. Source: ILVO - Merelbeke.

Photo 17. Haylage. Left: Round bale protected by a plastic film. Source: ILVO - Merelbeke. Right: Belgian Blue cows eating haylage in a barn in winter. Source: Wallonie Elevages.

Herbage seed production
Belgium has been steadily increasing its production area of herbage seed since the 1980s. From just over 1 000 ha of forage and turf grass/legume seed production in 1980, this increased to about 3 700 ha in 2004 (Table 27) – only 1.6 percent of the EU-25 production in 2004. Belgian production areas are small compared with those of some neighbouring countries: Denmark: 85 129 ha; Germany: 33 380 ha; Netherlands: 25 946 ha; France: 23 150 ha; United Kingdom: 6 287 ha (Bondesen, 2006). Production is carried out primarily on arable farms that have harvesting equipment. It is located in Lower and Central Belgium, particularly in the polders, the Flemish Plain and the Loamy Region. Soils and climate are more suitable in these regions than in grassland specialized regions; soils are deeper and easily ploughed and the climate is drier, which is important for seed maturation. The isolation of the seed production plots from pollen contamination (cross-pollination) from surrounding grasses of the same species is also more easily achieved in arable areas than in grassland areas.

Table 27. Evolution of forage and turf grass and legume seed production area (ha) in Belgium between 1982 and 2004

1982	1985	1988	1991	1994	1997	2000	2003	2004
1 182	1 029	1 775	1 581	1 644	3 641	3 610	3 293	3 687
Source: Bondesen, 2006

There is no legume seed production in Belgium (in Europe, Trifolium repens seeds are mainly produced in Denmark, T. pratense seeds mainly in France, Medicago sativa seeds mainly in France and Italy). In Belgium, the majority of grass seed production is from Lolium perenne (35 percent) and L. multiflorum (62 percent) (2000–04 average) (Wong, 2005). Annual seed yields are about 1 000–1 500 kg/ha (Bondesen, 2006). In the 2006–07 production year, the total Belgian grass seed production reached 2 400 tonnes for L. multiflorum (71 percent) and 1 000 tonnes of seeds for L. perenne (29 percent) (ISF, 2007).

Integration of forage resource utilization with environmental objectives and food quality issues

Environmental objectives
Over the last 50 years, between 1960 and 2010, livestock systems have changed dramatically through specialization of production (crop or livestock specialization on the one hand, dairy or meat production in livestock systems on the other), intensification (higher fertilization, more frequent cuts for silage making, higher stocking rates, higher use of concentrates, higher production per animal), decrease of forage legume use, farm and plot enlargement, destruction of hedges and other ecological elements. All these system changes have had severe impacts on the landscape and wildlife by reducing diversity and complexity. Farmers have also had to face criticisms for their negative effect on the quality of ground- and surface waters and measures have had to be taken to decrease nitrate and phosphate pollution.

Political response was initiated at European level by the adoption in 1992 of the regulations on the agri-environmental scheme (2078/92 and CEE 1257/99 regulations) that include support to OF, of the Nitrate Directive (Directive 91/676/CEE) in 1991 and of Natura 2000 (Bird (1979) and Habitat (1992) Directives). The agri-environmental scheme was transposed into Belgian regional laws relatively quickly in the 1990s. The Nitrate, Bird and Habitat Directives were only fully transposed at the beginning of the year 2000.

Agri-environmental programmes are designed at regional level in Belgium. The adoption of the programme and the associated payments are optional for farmers. They have two main objectives: reducing environmental risks associated with intensive farming on the one hand, and preserving biodiversity and landscapes on the other. Agri-environment payments may only be made for actions above the reference level of mandatory requirements defined by codes of good farming practices (GFP). Agri-environmental measures (AEM) include support for conversion to OF and its maintenance.

In the Walloon Region, 44 percent of farmers were involved in AEM in 2007 (Le Roi and Walot, 2009). The adoption rate of the measures is particularly high in grassland specialized regions. The budget (OF excluded) was about €16 million in 2005 (Goor, 2006). About one-fifth of the budget is devoted to nature and landscape. The programme includes the protection of elements of the ecological network (hedges, woodland strips, isolated trees and shrubs, traditional orchards, ponds), extensification of pastures (late cuts and fertilization cessation or reduction), sowing of extensive grassy field margins unfertilized and cut late, sowing of nitrate catch-crops between two main crops, support to extensive cereal growing, support to local threatened animal breeds (horses, cattle and sheep), support to low stocking rate and the implementation of agri-environmental action plans on the whole farm area. These measures have probably decreased the reduction rate of biodiversity in grassland and the simplification of landscapes. They have certainly had some effect on the reduction of nitrate and phosphate pollution.

In the Flemish Region in 2005, the area under one or more AEMs was estimated to be about 60 000 ha, or 10 percent of the Flemish AA (Carels et al., 2005). The annual budget was €16 million (OF included). The two main management agreements provide support for soil cover on arable land (43 percent of the total surface of AEM) and for water-related issues as part of the manure policy in the ‘Vulnerable Areas – Water’ (34 percent). There are five types of agricultural management agreements between farmers and the Flemish Ministry that are explicitly aimed at enhancing nature: management of grassland birds (0.55 percent), parcel edges (0.34 percent), small-scale landscape elements (6.12 percent), nature and vegetation management (2.11 percent). Other measures are: mechanical weeding, reduction of fertilizers and pesticides in ornamental plant cultivation, conservation of genetic diversity of local species threatened by extinction, conversion of traditional to organic pig farms, OF and integrated fruit production.

The Nitrate Directive is mandatory for farmers. It has been transposed at regional level in Belgium. The regions have identified vulnerable zones where groundwater is affected or could be affected by pollution. They have defined a code of GFP to be implemented by farmers, and have designed and implemented action programmes in respect of each vulnerable area. Just under half (42 percent) of the territory of the Walloon Region and all of the Flemish Region are designated as vulnerable areas. These action programmes include the measures prescribed in the codes of GFP. They also include measures to limit the spreading on arable land and pastures of any fertilizer containing nitrogen and set limits for the spreading of livestock manure. These limits imply a control of the stocking rate on the farm area. Farmers are also required to have a storage capacity for their animal manures in order to be able to spread them under optimal conditions. For slurry storage, this capacity reaches six months, which represents a significant financial investment. The regions monitor water quality, applying standardized reference methods to measure the nitrogen compound content. The transposition of this Directive has had a significant influence on farm structures and practices of intensive livestock systems by regulating the stocking rate and the management of nitrogen. [See Photo 18] 

Photo 18. Slurry injection into a grass sward. Source: ILVO - Merelbeke.

The Bird and Habitat Directives that are the legal basis of the Natura 2000 network have an impact on the AA even if their application concerns the whole area of the country and not only the AA, such as woodlands, wetlands and coastal areas. They focus on biodiversity conservation. The Natura 2000 network covers 12 percent of the Flemish Region and 13 percent of the Walloon Region (CBD-Belgium, 2009). Socio-economic activities including farming are maintained in this network. In the Walloon Region, it is estimated that approximately 16 percent of the habitats in Natura 2000 areas are extensive pastures. In this region, grassland and arable land in Natura 2000 represent less than 5 percent of the AA. Measures are taken for maintaining or restoring, at favourable conservation status, natural habitats and species of wild fauna and flora of community interest. Financing of the network management in grassland is based on AEM.

Food quality
Food quality includes several parameters such as nutritional quality, hygienic quality (food security), aspect and technological characteristics that are the visual part of food quality (e.g. the SEUROP classification system of animal carcasses), organoleptic or gastronomic quality (e.g. flavour, taste, juiciness, tenderness, colour of meat) and ethical components (e.g. animal welfare, biodiversity conservation).

For consumers, nutritional quality is a concern mainly for fatty acids. Hygienic quality is taken as granted. Consumers have an increasing demand for organoleptic quality and the ethical components of food. In complement to the price criteria, they choose their products on the basis of the food aspect and the information provided on the product such as trademark, official label, explanatory text and symbols.

As in surrounding European countries, there is an increasing demand for organic products in Belgium. Belgian stockbreeders can provide organic milk, cheese, yoghurt, ice cream and other dairy products, as well as red meat. Consumers buying organic products also expect a better taste and thus the choice of animal breed, age at slaughter and feeding system are important. Production of the tasty meat of Limousin is developing and can be considered as an alternative to the tender but less tasty Belgian Blue meat. Aberdeen Angus is still very rare, though this breed could contribute considerably to the production of tasty organic meat. Ox production is still very limited but could be reintroduced as this meat is tastier than young bull meat. Several cheese and other dairy specialities are proposed by organic and non-organic producers. ‘Herve Cheese’ and ‘Ardenne Butter’ are promoted by a Protected Designation of Origin (PDO) label. Protected Geographical Indication (PGI) labels have been attributed to ‘Ardenne Ham’ and ‘Gaume Paté’. Projects of PGI have been introduced by the Walloon Region to the European Commission for quality pig and poultry meat (see below).

The ‘Walloon Farm Broiler’ and the ‘Organic Broiler’ labels promote systems that use slow-growing strains of broilers. The animals must have access to a grassland paddock (4 m² of grassland range per animal). They are raised for 81 and 70 days minimum respectively and receive vegetal feed only, GMO free. Private companies such as ‘CoqArd’ (‘Coq de Pêche’ = ‘Peach Broiler’) and COPROSAIN propose trademarks of free-range broilers (2 m² of grassland range per animal). Quality broiler production represents about 15 percent of the production capacity in Wallonie. The production volume increased by 600 percent between 2001 and 2009.

There has been a recent development of free-range pigs protected by the general ‘Ardenne Quality Pig’ label in Upper Belgium. In the ‘Outdoor Pig’ label, each sow has a separate paddock with a shed for shelter and farrowing. Weaning of piglets occurs after six weeks. The piglets are then led to fattening paddocks. The paddock area per mature pig is 400 m² of grassland. A shed is present in these fattening paddocks, of sufficient size for sheltering all animals. Concentrate feeding is provided ad libitum. For the ‘Organic Pig’ label a minimum outdoor paddock is compulsory for exercise. The private company ‘Magerotte Enterprises’ proposes its own trademark called ‘Ardenne Grassland Pig’. This system is close to the ‘Outdoor Pig’ label.

Belgian production of ‘foie gras’ started in 1985. It developed very fast, mainly in Wallonie, with Mulard ducks (cross between a male Muscovy Duck and a female Pekin). It increased from about 2 t in 1991 to about 90 t in 2003 (Belgian consumption is estimated at about 200 t). There are about 40 breeders and 200,000 ducks in the country. Ducklings normally have access to a grassland plot for 3 months before a cramming period of 15 days.

Direct sale and short marketing chains are attracting an increasing number of farmers. Initiatives are supported by the Walloon Region.

---

6. OPPORTUNITIES FOR IMPROVEMENT OF PASTURE RESOURCES

Major technical innovations have been introduced in the Belgian agriculture sector over the last 50 years (1960–2010). Among these, the following techniques were particularly important. The spread of the use of ammonium nitrate in the early 1960s greatly increased pasture yields and animal stocking rates on pasture. This was particularly important in Belgium where land price is very high and AA is lacking. The introduction of grass silage in the 1970s significantly improved animal nutrition in winter. This was especially important for dairy cows. After the introduction of the milk quotas in 1984, high-quality grazed grass and grass silage became the cornerstones of dairy production systems. The improvement in machinery for cutting, harvesting, oversowing and overdrilling, silage making and distribution had a major impact on labour organization as it allowed higher frequency of cuttings and improved forage quality. Breeding of grass and legume cultivars improved forage yield (quantitative trait) but even more persistence and disease resistance (qualitative traits). The use of efficient herbicides made aggressive weed control possible (especially the control of Rumex obtusifolius) and facilitated pasture sward renovation. The adoption of better grazing management techniques decreased labour input while improving the quality of pasture sward, and the quality of grazed and conserved forages. In particular, the management requirements of Trifolium repens in grazed swards were much better understood. These grazing techniques together with silage making techniques played a crucial role in the decrease of production costs.

In the 1950s and the 1960s, grassland research in Belgium focused mainly on grassland typology, on the ecology of grassland species and communities. More applied research on, for instance, the effect of nitrogen fertilization, grassland renovation and forage species growing on grassland yields and quality, was initiated quickly thereafter. Later, a better understanding of soil processes, plant nutrition, forage plant physiology and quality, microbiological processes in silage, genetics in general, feed evaluation, rumen functioning and ecology, animal feeding and grazing behaviour contributed to the improvement of grassland management techniques and increased plant and animal yields. The number and the research topics of grassland experiments have increased greatly since the 1960s. The breeding research Centre of Merelbeke (now the Institute for Agricultural and Fisheries Research [ILVO]), near Ghent, has released many cultivars of forage grass and legume species including, for instance, one of the most famous cultivars of Lolium perenne, ‘Vigor’ also known as ‘Melle’, in 1963. This late cultivar persisted for a long time in commercial seed mixtures. [See Photos 19 and 20] 

Photo 19. Forage trials. Left: Grass and legume variety trial for recording yield, quality and persistence. Source: ILVO - Merelbeke. Middle: Grass plants in tufts for qualitative character recording. Source: ILVO - Merelbeke. Right: White clover (Trifolium repens) variety trial. Source: ILVO - Merelbeke.

Photo 20. Forage trials. Left: Plot harvesting and sampling. Source: ILVO - Merelbeke. Right: Trial visit and explanations for farmers and farmer’s advisers. Source: ILVO - Merelbeke.

Since the 1990s, the negative environmental impacts of intensive pasture management have become increasingly recognized. Nitrate leaching and water pollution by nitrate and phosphate were the first topics to emerge in grassland research on the environment. The best techniques for slurry application were studied in detail. The effects of the total amount and the date of application of mineral and organic fertilizers, as well as the effect of stocking rate and supplementary feeding in grassland paddocks on the nitrate residue in soil profiles in autumn, were investigated. Nutrient balances were calculated at field and farm-gate levels. Research on biodiversity conservation and restoration, and the integration of species-rich pastures in efficient and profitable livestock systems, followed quickly after these studies on water pollution. New research was initiated on legumes that had been neglected by pasture research over the previous 30 years. The inclusion of legumes in pasture swards was also seen as a way of increasing the diversity of swards. ‘Secondary’ or wild grass species were studied in a context of biodiversity restoration, of extensification and for better understanding of the competition between grasses. At the beginning of the twenty-first century, greenhouse gas (GHG) emissions from grassland soils, from the rumen of herbivores and from animal manure during storage and spreading, became important research topics. The positive influence of green forage on the conjugated linoleic acid (CLA) and polyunsaturated fatty acid (PUFA) contents of milk and meat have received much attention for their positive effect on human health. Holistic approaches have developed slowly compared with the scientific effort in reductionist approaches.

Future research must still address these issues by integrating agricultural management with environmental protection. Research priorities for the future must notably be related to the four priorities defined in the ‘Health Check’ of the CAP in 2008: climate change, renewable energy, water management and biodiversity. Among this research, energy consumption of agricultural production, GHG emissions and biodiversity conservation and restoration should be particularly developed.

The issues of climate change on the one hand and energy savings and production on the other are intrinsically linked. The reduction in energy use by agriculture will result in lower CO₂ emissions into the atmosphere. Future increases of fossil fuel price and the consequent increase of the nitrogen fertilizer price will be strong driving forces of change for Belgian farming systems in the next decades. Fundamental reforms of the systems are inevitable. In the future, systems will have to be less energy demanding. Fortunately, there are important possibilities to save energy per ha or per tonne of milk, for instance (Haas et al., 2001). Future research should focus systematically on energy costs and GHG emissions per production system and per product for developing such energy-efficient systems. A new integration of pasture and arable land at the farm and/or the regional levels will probably be necessary, for instance for reducing transportation costs. Biological nitrogen fixation by legumes will be one of the pillars of these future systems for saving the huge amounts of fossil energy that the synthesis of nitrogen fertilizers requires. A special effort will have to be made for livestock systems because they are less energy efficient than arable systems per kg of food produced. Future systems will also have to release less GHG into the atmosphere, not only CO₂ but also CH₄ and N₂O.

Climate change predictions of the Intergovernmental Panel on Climate Change (IPCC) suggest higher mean temperatures and changes in rainfall distribution (more summer droughts and heavier winter rain) in Belgium. Adaptation measures should be adopted. Dactylis glomerata could, for instance, replace Lolium spp. in cutting and mixed-use (cutting and grazing) systems. New management systems will have to be designed for defining the best techniques for using D. glomerata in mixed-use swards since this grass cannot resist intensive grazing. Legumes and particularly Medicago sativa will certainly have an advantage over grasses in summer drought periods. Systems will have to change by starting grazing earlier in spring, by later grazing in autumn and by a higher silage use in summer. Nitrogen fertilization patterns will also have to be adapted. Since climate change models are predicting more brutal rainy precipitations in the future, pastures could also be used for flood buffering and for enhancing water infiltration. This objective could be combined with wetland and biodiversity restoration. Research could examine the optimization of these changes over time when they will progressively occur.

Climate changes could also be mitigated in grassland by biofuel production and carbon storage. This can be achieved by converting arable land to pasture and by producing second-generation agro-fuels such as very short rotation coppice of Salix spp. or Miscanthus x giganteus on grassland margins. Biomass produced by ‘energy lucerne’ could possibly contribute to biofuel production on arable land. Hedge planting and woodland strip plantation on grassland margins could also be carried out. Research should study the potential benefits of these techniques while innovative bio-energy production techniques should be developed and improved.

There are possibilities to increase the role of many of the lesser-used pasture species (grasses and legumes) for particular environments and for species-rich sward restoration. This particularly applies to Upper Belgium. Grassland species can be used for several environmental enhancement purposes such as grassy field margins for erosion control in arable land, surface water protection (buffer zone) and water infiltration in grassland and arable land, and species-rich field margins for restoring inter alia farmland bird and pollinator populations.

Further breeding improvement of cultivars of traditional legume and grass species as well as breeding of secondary grasses could certainly induce progress in forage quality and cost reductions in more environment-friendly systems.

The use of legume forages should be increased on most grassland farms not only for saving energy because they have the potential to replace many N-fertilized grass swards but also because they can reduce the need for purchased protein-rich feed. They can also help to achieve goals of environmental policy and to meet consumer expectations (Peeters, Parente and Le Gall, 2006).

Grasslands should produce high amounts of quality water by favouring clean water infiltration. Nitrate residues in soil water should thus be reduced to a minimum but at the same time water infiltration should be enhanced. Trampling of grassland soils by high animal stocking rates and heavy grass harvesting machines for silage making has often degraded soil structure. This has led to a reduction of water infiltration in grassland soils. Future research should examine ways of improving this situation.

A continuous improvement of nutrient management should be a target of future research for reducing production costs and pollution.

Grassland forages are now recognized as having a beneficial effect on animal product composition and human health. CLA contents in meat and dairy products can have a positive anti-carcinogenic effect while a low ratio of n-6 to n-3 PUFA has been associated with a low susceptibility to coronary heart disease. These findings create possibilities for further research on grassland-based feeding systems. Grass-based bull and ox fattening systems should be studied for increasing efficiency, profitability and quality of the products. Similar research should be conducted in dairy systems for the quality of dairy products. Pasture-based systems of tasty products should be developed in dairy, beef, pig and poultry production. New animal breeds could be introduced (e.g. Aberdeen Angus and hardy pig breeds) or created by selection in the perspective of quality product diversification. New quality production systems should be developed or improved while minimizing their possible negative impacts on the environment, or for combining biodiversity restoration on the one hand and the production of quality animal products on the other.

Future pasture and grassland-based system research will have to focus on the multiple functions that grasslands offer to society. Forage and livestock systems should be supported by society for the services they provide, including biodiversity conservation and landscape protection. They also offer recreational opportunities and contribute to the quality of life by producing healthy and tasty products. Future research should take all these aspects into account.

Policy research should examine the efficiency and reform of existing programmes: reinforcement of the cross-compliance principle, agri-environmental schemes, support to OF, the implementation and development of Natura 2000, the implementation of the High Nature Value (HNV) farmland programme, quality product policy, and the integration of ecosystem services provided by grasslands in the price of products or in financial support. This research should also look at the design of new policy programmes for reaching the multiple objectives of future Belgian agriculture.

In the longer term, the effects of the increase in food demand from Asia and the competition for other land uses (urbanization) could induce a new phase of intensification of grasslands. This should be achieved at an acceptable environmental cost. Research is thus needed for designing systems that could be productive, energy efficient, low GHG emitters and producers of biodiversity. This will not easily be achieved.

7. RESEARCH AND DEVELOPMENT ORGANIZATIONS

8. REFERENCES

References

Association Régionale de Santé et d’Identification Animales (ARSIA). 2008. Identification animale. Available at: http://www.arsia.be/identification

Bondesen O.B. 2006. The global seed production now and in the future, where and who? Danish Seed Council. Available at:

http://www.dansklandbrug.dk/NR/rdonlyres/F07E001C-702E-45B1-A66E-32BC6CF338E0/0/Global_seed_now_and_future.pdf

Carels K., De Clercq P., Van Gijseghem D. 2005. Impacts of Agricultural Policy on Rural Development in Belgium: case study of the Flemish Region. OECD workshop on Evaluating Agri-environmental Policies, Bratislava, 24-26 October 2005: 13 pp.

CBD-Belgium. 2009. Fourth National Report of Belgium to the Convention on Biological Diversity. Royal Belgian Institute of Natural Sciences: 96 pp.

CERVA. 2005. Brucella. Newsletter: 1 pp.

Deprez, B., Lambert, R., Decamps, C. & Peeters, A. 2004a. Production and quality of red clover (Trifolium pratense) and lucerne (Medicago sativa) in pure stand or in grass mixture in Belgium. Grassland Science in Europe, 9: 498-500.

Deprez, B., Lambert, R., Decamps, C. & Peeters, A. 2004b. Nitrogen fixation by red clover (Trifolium pratense) and lucerne (Medicago sativa) in Belgium leys. Grassland Science in Europe, 9: 469-471.

Deprez, B., Parmentier, R., Lambert, R. & Peeters, A. 2007. Les prairies temporaires : une culture durable pour les exploitations mixtes de la Moyenne-Belgique. Les Dossiers de la Recherche agricole 2. Namur, Belgium, Ministère de la Région Wallonne, Direction Générale de l’Agriculture, Direction de la Recherche. 84 p.

DGSEI. 1990. L’agriculture. Available at: http://statbel.fgov.be/fr/statistiques/chiffres/economie/agriculture/index.jsp

DGSEI. 2000. L’agriculture. Available at: http://statbel.fgov.be/fr/statistiques/chiffres/economie/agriculture/index.jsp

DGSEI. 2007. L’agriculture. Available at: http://statbel.fgov.be/fr/statistiques/chiffres/economie/agriculture/index.jsp

DGSEI. 2008. L’agriculture. Available at:http://statbel.fgov.be/fr/statistiques/chiffres/economie/agriculture/index.jsp

DGSEI. 2009. L’agriculture. Available at:http://statbel.fgov.be/fr/statistiques/chiffres/economie/agriculture/index.jsp

European Soil Bureau. 2005. Soil Atlas of Europe. Brussels, European Commission. 128 p.

Federal Public Service Economy (FPS Economy). 2007. Panorama de l’économie belge. Brussels. 257 pp.

FPS Economy. 2008. L’agriculture en Belgique en chiffres. Chiffres clés de l’agriculture belge 2008. Directorate-General Statistics and Economic Information. Brussels. 24 p.

FPS Economy – DGSEI. 2008. L’accroissement de population le plus important depuis 1965. La Belgique s’approche des 11 millions. Communiqué de presse, 28 août 2008 : 3 pp.

FPS Economy – DGSEI. 2009. Population. Available at: http://statbel.fgov.be/fr/statistiques/chiffres/population/index.jsp

Genot, L. (ed.) 2005. Les ressources génétiques des animaux d'élevage en Belgique. Contribution de la Belgique au Premier Rapport sur l'État des Ressources Zoogénétiques dans le Monde. Rapport national à la FAO. 58 pp. http://agriculture.wallonie.be/apps/spip_wolwin/IMG/pdf/RapportNationalFAO.pdf

Goor, F. 2006. L’agriculture. Etat de l’environnement wallon: 39-59.

Haas G., Wetterich F. and Köpke U. 2001. Comparing intensive, extensified and organic grassland farming in southern Germany by process life cycle assessment. Agriculture, Ecosystems and Environment, 83, 43-53.

Hanset, R. 1998. Emergence and selection of the Belgian Blue breed. Belgian Blue Herd- Book. Ciney, Blgium, University of Liege. 16 pp.

Houins, G. 2007. Belgian approach of the safety of the food chain: The Federal Agency for the Safety of the Food Chain (FASFC) from 2002 to 2006. Brussels, Federal Agency for the Safety of the Food Chain. 19 p. Also available at http://www.favv.be/publications-en/Belgian-approach-safety-food-chain.asp

Houins G. 2009. The Belgian Food Safety Agency in 2008. Facts and figures. Federal Agency for the Safety of the Food Chain: 63 pp.

Institut Géographique National (IGN). 2001. Carte adminsitrative de Belgique. Available at http://www.ngi.be/images/4/adm/adm6_f.htm

ISF (International Seed Federation). 2007. Certified (C) and uncertified (UC) seed production (tons) of selected forage and turf species. Sowing season 2006. Available at: http://www.worldseed.org/cms/medias/file/ResourceCenter/SeedStatistics/ForageandTurfSeedMarket/

Lecomte H., Florkin P., Morimont J.P. & Thirion M. 2000a. La forêt wallonne : état de la ressource à la fin du 20^ème siècle. 3. La forêt wallonne en quelques chiffres. Division de la Nature et des Forêts, Direction des Ressources forestières, Cellule Inventaire des Ressources forestières de Wallonie: 19 pp.

Lecomte H., Florkin P., Morimont J.P. & Thirion M. 2000b. La forêt wallonne : état de la ressource à la fin du 20^ème siècle. 4. Les caractéristiques du milieu forestier. Division de la Nature et des Forêts, Direction des Ressources forestières, Cellule Inventaire des Ressources forestières de Wallonie: 9 pp.

Lecomte H., Florkin P., Morimont J.P. & Thirion M. 2000c. La forêt wallonne : état de la ressource à la fin du 20^ème siècle. 5. Les peuplements de la forêt wallonne. Division de la Nature et des Forêts, Direction des Ressources forestières, Cellule Inventaire des Ressources forestières de Wallonie: 33 pp.

Le Roi, A. and Walot, Th. 2009. Évaluation des mesures agro-environnementales en Région wallonne. EVAGRI 2006 – 2008. Partie 6. Evolution du programme agri-environnemental en Wallonie. Synthèse cartographique et statistique. GIREA : 40 pp.

Météo Nature. 2009. Dossiers météo. Available at:http://www.meteonature.com/index/index.php?option=com_content&view=section&id=8&Itemid=85

National Institute of Statistics. 1999 in Genot (ed.) 2005. Les ressources génétiques des animaux d'élevage en Belgique. Contribution de la Belgique au Premier Rapport sur l'État des Ressources Zoogénétiques dans le Monde. Rapport national à la FAO. 58 pp. Available at: http://agriculture.wallonie.be/apps/spip_wolwin/IMG/pdf/RapportNationalFAO.pdf

Organisation for Economic Co-operation and Development (OECD). 2008. Environmental Performance of Agriculture in OECD countries since 1990. Belgium Country Section, pp. 210-­242. Paris.

Peeters, A. & Kopec, S. 1996. Production and productivity of cutting grasslands in temperate climates of Europe. Grassland Science in Europe, 1: 59-73.

Peeters, A., Parente, G. & Le Gall, A. 2006. Temperate legumes: key species for sustainable temperate mixtures. Grassland Science in Europe, 11: 205-220.

Peeters, A., Lafontaine, R.M., Beudels, R., Devillers, P., Nolte, S., Buysse, J. & Van Huylenbroeck, G. 2009. Evaluation de l’impact sur la biodiversité du développement de cultures pour biocarburants, notamment de plantes génétiquement modifiées, en Belgique. Service Public Fédéral - Santé publique, Sécurité de la chaîne alimentaire et environnement. 136 p.

Royal Meteorological Institute. 2009. Le climat de la Belgique. Available at: http://www.meteo.be/meteo/view/fr/123769-Le+climat+de+la+Belgique.html

The World Factbook. 2009. Available at https://www.cia.gov/library/publications/the-world-factbook/geos/be.html

Wong D. 2005. World Forage, Turf and Legume Seed Markets. Available at: www1.agric.gov.ab.ca/%24department/newslett.nsf/pdf/fsu6885/%24file/worldforage.pdf

Further reading

Andries, A. 1950. L’appréciation dans la pratique de la valeur agricole des herbages, par l’examen de leur composition botanique. Revue de l’Agriculture, 12: 15-19.

Andries, A. 1982. Rôle du trèfle violet dans la production fourragère belge, évolution et perspectives. Fourrages, 90: 27-37.

Andries, A. & Carlier, L. 1980. Systèmes de pâturage : pâturage continu / pâturage rotatif (résultats de cinq années de recherche). Communic. RVP 450: 32 pp.

Andries, A., Limbourg, P., Toussaint, B. & Carlier, L. 1981. Exploitation des prairies. 1) Amélioration et resemis des prairies. 2) La fumure azotée, facteur de production quantitative et qualitative de l'herbe. Revue de l'Agriculture, 34(3): 469-489.

Andries, A. & Van Slijken, A. 1965. The effect of different methods of improving the botanical composition of permanent pasture. Neth. J. Agric. Sci., 13(2): 114-119.

Andries, A. & Van Slijken, A. 1969. Fumure phospho-potassique sur herbage. Rapport de 5 ans de recherches concernant la fumure phospho-potassique en fonction du mode d'exploitation de l'herbage. Revue de l'Agriculture, 9: 1197-1229.

Bienfait, J.M., Van Eenaeme, C. & Limbourg, P. 1978. Comparaison de quatre systèmes de pâturage différant par les chargements, les doses de complément énergétique et les niveaux de fumure azotée en fonction de la productivité de l'herbe et de la production laitière sur les pâturages d'Ardennes belges. Fourrages, 74: 19-28.

De Cauwer, B., Reheul, D., Nijs, I. & Milbau, A. 2006. Dry matter yield and herbage quality of field margin vegetation as a function of vegetation development and management regime. NJAS - Wageningen Journal of Life Sciences, 54(1): 37-60.

Dieguez, F., Hornick, J.L., Istasse, L. & Dufrasne, I. 2001. Désintensification raisonnée du pâturage par la vache allaitante Blanc Bleu Belge. Fourrages, 165: 61-71.

Dufrasne, I., Gielen, M., Limbourg, P., Brundseaux, C. & Istasse, L. 1995. Production bovine allaitante en Belgique : effets de l’intensification et de la complémentation des veaux au pâturage. Fourrages, 141: 91-104.

Eurostat. 2008. Agricultural statistics. Main results 2006-2007. European Commission, Pocketbooks. 147 p.

Lambert, J. & Latour, G. 1969. La prairie temporaire de fauche : ses limites et ses potentialités dans la province du Luxembourg belge. Fourrages, 37: 1-13.

Lambert, R., Peeters, A. & Toussaint, B. 1998. Persistance du trèfle dans les associations fourragères fauchées : importance du choix variétal. Fourrages, 152: 511-517.

Lecomte, Ph., Stilmant, D., Seutin, Y. & Dardenne, R. 1998. Variabilité des quantités et de la qualité des ensilages en balles enrubannées récoltés dans une exploitation. Fourrages, 156: 517-525.

Legrand, E., Legrain, A. & Geerens, P. 1989. L’ensilage d'herbe. Mécanisation balles rondes sous plastique étirable. Note technique 5/53. Gembloux, Station de Génie rural, CRA. 49 p.

Limbourg, P., Decruyenaere, V., Parache, P. & Stilmant, D. 2001. Fumier composté sur prairie pâturée par des taurillons de type à viande. Fourrages, 167: 329-335.

Limbourg, P., Lambert, J. & Toussaint, B. 1983. Le trèfle blanc en Belgique : observations sur son comportement et perspectives d'avenir. Fourrages, suppl. 94 & 95: 29-47.

Marsin, J.M. 2006. Le secteur de l’agriculture en région wallonne. Etat de l’Environnement wallon. 26 p.

Meul, M., Nevens, F., Reheul, D. & Hofman, G. 2007. Energy use efficiency of specialised dairy, arable and pig farms in Flanders. Agriculture, Ecosystems & Environment, 119(1-2): 135-144.

Miserque, O., Oestges, O. & Tissot, S. 1998. Témoignage sur l'évolution des systèmes de récolte en zone herbagère de Wallonie (Belgique). Fourrages, 155: 293-304.

Miserque, O., Tissot, S. & Oestges, O. 1997. Coût des techniques de récolte des fourrages en Belgique. Fourrages, 149: 81-93.

Nevens, F. & Reheul, D. 2001. Crop rotation versus monoculture; yield, N yield and ear fraction of silage maize at different levels of mineral N fertilization. Netherlands Journal for Agricultural Sciences, 49(4): 405-425.

Nevens, F. & Reheul, D. 2002. The nitrogen- and non-nitrogen-contribution effect of ploughed grass leys on the following arable forage crops: determination and optimum use. European Journal of Agronomy, 16(1): 57-74

Nevens, F. & Reheul, D. 2003. Permanent grassland and 3-year leys alternating with 3 years of arable land: 31 years of comparison. European Journal of Agronomy, 19(1): 77-90.

Nevens, F., Verbruggen, I., Reheul, D. & Hofman, G. 2006. Farm gate nitrogen surpluses and nitrogen use efficiency of specialized dairy farms in Flanders: Evolution and future goals. Agricultural Systems, 88(2-3): 142-155.

Peeters, A. 2004. Wild and Sown Grasses. Profiles of a Temperate Species Selection: Ecology, Biodiversity and Use. Oxford, UK, Blackwell Publishers and Rome, Italy, FAO. 311 p.

Peeters, A. 2009. Importance, evolution, environmental impact and future challenges of grasslands and grassland-based systems in Europe. Grassland Science, 55(3): 115-125.

Peeters, A. & Janssens, F. 1998. Species-rich grasslands: diagnostic, restoration and use in intensive livestock production systems. Grassland Science in Europe, 3: 375-393.

Peeters, A. & Lambert, J. 1990. Application agronomique d’une typologie des prairies intensifiées. Fourrages, 124: 357-369.

Reheul, D., Nevens, F., Hofman, G., Van Cleemput, O., Bogaert, N., Vermoesen, A., Carlier, L., Michiels, J., Verbrugge, I., Kuyken, E., Martens, K., Van Huylenbroeck, G., Jacobs, G. & Martens, L. 1998. Naar een duurzame grasland en groenvoederuitbating. Onderzoek naar de integratie van landbouwkundige en ecologische doelstellingen bij grasland- en groenvoederwinning. Brussels, Nationaal Centrum voor Grasland- en groenvoederonderzoek (ed.) Ministerie van Middenstand en Landbouw. 221 p.

Rigot, J. 1977. Evolution de la production fourragère en Belgique : passé et avenir. Fourrages, 72: 123-143.