Global Strategy issues Handbook on Remote Sensing for Agricultural Statistics

In the context of the Global Strategy to improve Agricultural and Rural Statistics, remote sensing has been identified as a point of reference for Master Sampling Frames and methodological improvements in design and estimation terms, as a prime contributor to the localization and geocoding of the sampling units, as a way to achieve sustainability and as a core data provider for indicators linked to land uses and covers.
With the adoption of the 2030 SDGs and the importance of the agro-environment indicators, its contribution has even been reinforced.

The recently published handbook on Remote Sensing for Agricultural Statistics provides guidelines on the use of remote sensing in the context of agricultural statistics. Since the mid-1970s, remote sensing has been considered a promising technique for improving agricultural statistics. Various applications of remote sensing have taken place on all continents and today, several approaches may be considered mature enough to contribute to the sustainability of agricultural statistics.

This handbook seeks to enable interested readers to comprehend whether remote sensing can answer their needs, and if its adoption can improve timeliness, coverage, precision and/or costs in a sustainable manner.   Its structure reflects the diversity and complexity of the domain of agricultural statistics, as well as of the technicalities of remote sensing:

  • An agricultural statistical information system is composed of several layers, each corresponding to different core statistical topics and societal needs. Remote sensing can be particularly efficient in improving Global Strategy core items linked to crop areas, yields and productions. Its role is highly versatile, potentially ranging from optimization of sampling design to the facilitation of the fieldwork of enumerators, quality assurance and even data production. Societal needs can be separated into two components: (1) the production forecasts from early season to pre-harvest time, for food security monitoring and (2) classical agricultural statistics, of which the continuity and consistency over time will allow policymakers to plan and evaluate agricultural policy and its positive effect on total factor productivity, farmer income and rural development.
  • The techniques associated with remote sensing raise issues pertaining to the sensors (optical or radar), image resolution (30 cm to 5 m) and revisiting time (one hour to 16 days); to (non-)open access and the (generally prohibitive) associated prices; and to the software and hardware available for image analysis (open-source or commercial software, local or cloud computing). This aspect will require managers to identify the time, resources and staff competences required to move from experimentation to operational activities.