The assessment is essentially based on a bibliographical survey at FAO, accompanied by sending questionnaires and missions to several countries to compile and validate collected data.
The bibliographical survey concentrated on statistics concerning:
forest cover;
growth in forest resources (afforestation and reforestation);
the major causes of generally accepted deforestation: human population, livestock, fuel-wood, fires.
The bibliographical survey found limitations in the availability and quality of information collected; in fact, the available data are rarely complete, homogenous or reliable.
A questionnaire on the availability of statistical information on forest cover and biomass was sent to the countries concerned in order to complete and validate the information collected. The few replies received within the time limit granted meant that the questionnaire had to be sent a second time, this time completed in advance with available information; only Lebanon and Yemen replied on both occasions. A visit by experts made it possible to make good the information gaps in the survey and in the questionnaire, by means of collecting the required data in the five large forestry countries in the zone: Algeria, Iran, Morocco, Syria and Tunisia. Turkey (outside the zone) was also visited.
The forest inventory data were standardised and adjusted to a common reference year (1990) by means of adjustment functions created at sub-regional level. In the absence of multi-date data, the adjustment functions (models) were based on the rate of forest cover observed at sub-national level, and a series of additional variables correlated to the rate of forest cover such as human population and livestock. The correlation between forest cover and the additional variables was tested by using a stepwise regression analysis to determine the best combination of variables as a function of the correlation coefficient (R2). For each sub-regional model, several trials were made to test the strength of the model's parameters with the data used.
2.1 NORTH AFRICA
The forest inventory data and/or assessments were used in estimating forest cover at sub-national level. These data were available for Algeria, Morocco and Tunisia for the reference years 1984 to 1990. As regards Egypt and Libya, no reliable inventory data were available and the data estimates from the previous assessment were retained [Interim report on the state of forest resources in the developing countries, Forest Resources Division, Forestry Department, Rome, Italy (FO:Misc/88/7)].
The following data were available for each province:
total population density and rate of increase;
rural population density;
livestock pressure (sheep equivalent/ha);
percentage of lands per ecological zone (derived from the ecological map of the International Vegetation Map Institute (ICIV - Toulouse, France);
Weck climatic index (WCI).
During the data analysis, it was noted that the correlation improved if a relative value was used for forest cover, where relative forest cover = natural forest area/potential forest area (derived from the ICIV's ecological map of Africa). It was also decided to exclude certain data from the analysis which deviated strongly from the general trend. These data included: (i) desertic zones; and (ii) zones which have suffered a very strong (positive) human influence, such as plantations or other man-made landscape (corresponding to vegetation types Nos78 and 79 “man-made landscapes” from the Africa Vegetation Map - Unesco/AETFAT/UNSO).
Two variables were found to be particularly significant in explaining the current rate of forest cover: the mountain moist areas (expressed in percentage of the lands), and the rural population density (results in Annex 1).
2.2 NEAR EAST
The methodology used for countries of the Near East is the same as that for the countries of North Africa, but the state of information on forest resources at sub-national level is less good.
Data on forest cover were available for Afghanistan, Iran and Saudi Arabia. For the other countries, only national estimates were available. Reference years vary from 1959 to 1987. Only one multi-date piece of information was available, that of Lebanon (1965 and 1990). Data on forest cover in Turkey at the provincial level were used to reinforce the set of data.
Bioclimatic data on the countries concerned was not available in the Geographic Information System of the 1990 forest resources assessment project. These data were taken from the bioclimatic map of the Mediterranean region (FAO/Unesco, 1962).
Only a few countries had performed a national forest inventory at the time of the study (Algeria in 1984, Lebanon and Jordan in 1965). National forest inventories were under way in Algeria, Morocco and Tunisia, but the results would not have been available by the end of the study. With the exception of Afghanistan, Lebanon and Turkey (outside the zone), no other country had multi-date information available which would have enabled forest resources monitoring to take place.
The results of the data collected may be summarised as follows:
classifications and definitions of forests and other wooded lands are badly defined and rarely respected, or unstable. The chronological data series is incomplete or is not the same from one country to the next. Therefore, no comparison is possible between data before and after the break; unless the development consisted of a regrouping or a burst and that one concordance table shows the passage from old to new divisions;
afforestation and reforestation areas are often confused and the reforestation considered as a net extension. This places a bias on the statistics, leading to an over-estimate of forest area. Furthermore, the rate of plantation success is generally unknown or is rarely considered in plantation inventories;
statistics on volumes of fuel-wood or round wood taken from the forest are widely under-estimated because they do not take illegal exploitation into consideration;
only four countries kept a forest inventory which generated information that could be used for estimating growing stock or vegetation biomass.
Information regarding the state of change in forest cover and biomass in the region will only be available on a reliable basis in the statistical plan if a significant effort is made to reinforce national capabilities in the area of forest inventory and forest resources monitoring.
These findings justify, if they needed justifying, the priority given to assessment and to the “systematic observation” of forest resources by the United Nations Conference on Environment and Development (UNCED, Rio de Janeiro, June 1992) in chapter 11 of its “Action 21” programme.
