Approach to soil productivity assessment over Canada’s agricultural land using Earth Observation technology

Accurate information on soil quality is needed for decision-making at different levels and scales. It also supports models for predicting agricultural productivity and developing beneficial soil and crop management practices for agri-environmental sustainability. Crop biomass accumulation and final yield are representations of soil and environmental conditions as well as agricultural management practices. Different systems have been developed in Canada to rate the general agricultural land capability, or to assess the suitability of land for specific crops. These expert systems are based on soil polygons in which the spatial distribution of soil subcomponents is not explicitly delineated. Earth Observation (EO) technology, with its capability to acquire spatial-temporal information of crop growth conditions over large areas, provides a useful tool for soil productivity assessment. This paper introduces approaches and activities used in mapping soil productivity and estimating crop yield using EO at Agriculture and Agri-Food Canada. High resolution optical EO data from different sensors have been used to map spatial-temporal variability of biophysical descriptors within crop fields, and these outputs have been integrated into radiation use efficiency models and crop growth models for biomass and yield estimation at field and sub-field scales. At the regional scale, time series MODIS products have been used for deriving crop phenology parameters, mapping general crop types, and estimating crop yield. Our experience shows that multi-temporal optical remote sensing data is useful for crop productivity mapping at different scales, and the information on soil productivity can be incorporated into digital soil mapping approaches used for updating and upgrading the spatial representation of soil properties.