Issues and guidelines for the emerging use of GPS and PDAs in agricultural statistics in developing countries

The Global Strategy to Improve Agriculture and Rural Statistics adopted by the United Nations Statistical Commission recognises the importance of geo-referencing statistical units and use of geospatial information for the construction of master sample frames which is a key element for integration of agriculture into national statistical systems, one of the three pillars of the Strategy. The use of handheld geo-referencing devices (global positioning system and personal data assistant equipped with GPS) is also emerging as an effective means of reducing the cost of crop area data collection and improving crop area data quality particularly in the African context. The crop area (area planted and area harvested) is among the key variables included in the core data set defined in the Global Strategy which have to be produced by all countries on a regular basis. The availability of reliable estimates of crop area is an important requirement of any country agricultural statistics system, as this variable is an important parameter for estimating the production, together with crop yield. These three variables (area, yield and production) are among any minimum set of key indicators for planning, monitoring and evaluating agricultural and rural development programmes aiming at food security and raising the living standards of the population in a country. However, in many developing countries, collecting reliable data on crop area still remains a challenge for agricultural statisticians. This was confirmed by a recent stakeholder survey conducted in the framework of the preparation of the implementation plan for Africa of the Global Strategy. The results of the survey clearly showed that improving the methodologies for estimation of crop area, yield and production remains the highest priority for research in Africa. Different methods are currently used for collecting crop area data in developing countries, including field reporting system, eye estimation, interview of the farmers, objective measurement methods. All these methods have their limitations in terms of reliability of crop area data, but the objective method of measuring areas is considered to give the most reliable data. However, measuring areas is difficult, costly and time-consuming in developing countries, due to specific problematic situations. The shape of the fields, is not always polygonal but often a curvilinear closed 2 figure, which has to be reduced to a polygon, with a small number of sides (e.g. less than 20) of an equivalent area. Measuring errors can be introduced by the surveyor or are inherent to the equipment used; thus the investigators have to be well trained on surveying techniques and on the proper use of the necessary equipment. In hilly regions, when crops are grown on slopes which could be quite abrupt (more than 20%), the evaluation of the crop area is not simple. The crop area should not be the physical area measured on the slope (the inclined plan) but its projection on the horizontal plane. This is due to the fact that plants and trees grow vertically and not perpendicularly to the slope and thus require for their growth some kind of vertical cylinder of soil. If the crop area is measured on the slope and not projected horizontally, crop areas could be significantly over-estimated. In addition to difficulties associated with measuring the area of a given plot, there are difficulties in determining the share of specific crops in presence of mixed or associated cropping which is a common practice in African countries. In some cases, two or more different temporary and/or permanent crops simultaneously grow on the same field or plot. The problem of estimating crop areas in these situations gets more and more complicated as the number of crops in the mixture or in association increases and especially when the proportions of the different crops in the mixture vary from field to field. Moreover, the vegetative...