1.3. Analysis of the data included in the Compendium

The basic objective of preparing this Compendium is not to draw any conclusion or analyse the state of the agricultural environment but only to provide basic data to policy makers and analysts. In the past, decision-makers have analysed the economic impact of various programmes and policies without taking into consideration environmental aspects. However, they may like to know the answers to questions such as: (a) What areas need extension support (e.g. use of fertilizer, crop rotation programmes) to improve the quality of land, and what are crop type and farming practices suitable for a region? (b) If the livestock population versus pasture land ratio is suitable for a region and, if not, what changes are required in farming practice for correcting the adverse impact on land? It was noted by the FAO Statistics Division that although agencies like UNEP provide detailed information to review the state of the environment, comprehensive information on agricultural related indicators is not readily available. The information contained in the Compendium is intended to fill this gap, and can be used to study the inter- and intra-regional state of the agricultural environment.

Before looking into the indicators, the nature and limitations of the data should be understood. The indicator series is based on the data compiled by FAO Statistics Division that form a statistical time series starting with the year 1961. The statistics are published in the Production Yearbook and Fertilizer Yearbook and also presented on the FAO website. Some of the basic characteristics of these data-sets are given below:

1. A leading difficulty faced by the Statistics Division arises from the huge quantity of data required to create a global data-set covering all countries and all items, necessary to undertake global and regional analysis. Not only is there a huge quantity of annual figures relating to various elements to collect or estimate, evaluate and record but also the magnitude of the figures - in terms of units - varies between a few thousands to millions depending upon the country.
   
   
2. These data are collected in numerous ways, primarily by sending tailored questionnaires to member countries. Other sources are international and national publications, including general yearbooks, agricultural yearbooks, pocket-books, periodicals, FAO Country Representatives' reports, and country visits by ESS statisticians. In some cases, official figures released by the countries can be found in these publications long before they reach FAO through the questionnaires. It is not always easy to ascertain which of the figures reported by the various sources is the most recent or the most reliable, although the national Yearbook data are supposed to be the most accurate. However, they report historical data up to a period of one to three years behind the current year.
   
   
3. The data are evaluated for consistency within the country and between countries in line with the definitions and coverage of the data recommended by FAO. Furthermore, although there are so many sources of data, it should be noted that most sources deal with the limited data according to their own statistical system. In other words, too much information for certain commodities and too little for others. This results in missing data and gaps in the series. In the absence of any figure from official or unofficial sources, estimates for missing data are made on the basis of various assumptions. However, the estimation of a complete time series, when neither official nor unofficial data are available, presents more difficulties. All these, of course, are the matters which deserve attention and which receive only limited resources.

As an example, the land use classification adopted in this connection is linked to the FAO's recommendations for censuses of agriculture. The FAO land use statistics are more linked to the use of land for agriculture and are historically based on point estimates derived from data collected in periodic agricultural censuses and estimated in intercensual years using a variety of information and data sources. The analysis on data received from countries indicates that the land use definition given by the FAO is not adopted by the majority of countries in collecting and compiling data. Some of the problems countries faced in collection, compilation and presentation of data on land use are given below:

1. There is no universally accepted concept for some of the items used in the land use classifications. Definitions used by reporting countries vary considerably and items classified under the same category often relate to different kinds of land. Three important points in this connection are as follows:
   
   
   1. Most countries define arable land as the land that is potentially cultivable, whereas FAO's definition refers to land under temporary crops, meadow and pasture. This classification also includes fallow land. Fallow land is generally of two types: land that has been left idle in the current crop season to improve the productivity of the land; and land that is left fallow for a longer time period and for which no cultivation activity has been planned. Many countries do not distinguish between these two types of fallow area and count them under arable land.
   2. Area under shifting cultivation presents another problem for collection of data on arable land. This activity can be defined as cultivation involving the removal or burning of vegetation for pasturage or other purposes to support human life. This practice is being done in both fallow areas (which are of a permanent nature) as well as in forest areas. Many countries include this category of land under arable land. Some countries (such as Mali, Niger) also include area under shifting cultivation in the harvested area but exclude it from arable land.
   3. Some countries refer to wooded land as what statistician prefers to call 'woodland'. In most developed countries, animals graze in these areas. Statistically, the areas where animals graze are classified as pasture while those involved in resource assessment classify them (on the basis of satellite imagery) as wooded land. Similarly there are also problems with the classification of area under some plantation crops. For example, should rubber plantations be classified as wooded land or woodland?
      
      
2. Furthermore, the land use database is not always up-to-date for several reasons. In many countries, there is no established statistical system to generate such data. Gaps have been filled by FAO estimates using other information and assumptions while keeping the time series data reliable. These problems are highlighted to point out that the indicators derived using country data should be seen to reflect the general magnitude of the data rather than an exact measurement. It is possible to compile a more appropriate indicator for each individual country by taking into account their farming customs; however, this has not been attempted as it would present problems in inter-country comparison and aggregation. Therefore, in the present compendium this global data-set has been included without making adjustments for each individual country in order to present an overview of the state of environment.