Using Sentinel-2 to Track Field-Level Tillage Practices at Regional Scales in Smallholder Systems

Weiqi Zhou,Shishpal Poonia,Balwinder Singh,Manish Kumar,Meha Jain,Preeti Rao,Mangi L Jat,Love Kumar Singh,Urs Schulthess,Deepak Bijarniya,Rajbir Singh

doi:10.3390/rs13245108

Abstract

Zero tillage is an important pathway to sustainable intensification and low-emission agriculture. However, quantifying the extent of zero tillage adoption at the field scale has been challenging, especially in smallholder systems where field sizes are small and there is limited ground data on zero tillage adoption. Remote sensing offers the ability to map tillage practices at large spatio-temporal scales, yet to date no studies have used satellite data to map zero tillage adoption in smallholder agricultural systems. In this study, we use Sentinel-2 satellite data, random forest classifiers, and Google Earth Engine to map tillage practices across India’s main grain producing region, the Indo-Gangetic Plains. We find that tillage practices can be classified with moderate accuracy (an overall accuracy of 75%), particularly in regions with relatively large field sizes and homogenous crop management practices. We find that models that use satellite data from only the first half of the growing season perform as well as models that use data throughout the growing season, allowing for the creation of within-season tillage maps. Finally, we find that our model can generalize well through time in the western IGP, with reductions in accuracy of only 5–10%. Our results highlight the ability of Sentinel-2 satellite data to map tillage practices at scale, even in smallholder systems where field sizes are small and cropping practices are heterogeneous.

Highlights

Zero tillage, one of the main components of conservation agriculture, has been shown to increase crop yields, increase farmer profits, reduce soil and water loss, and reduce environmental footprints in many agricultural systems across the globe [1,2,3,4,5]
As current estimates are based on household surveys with limited smallholder systems, as current estimates are based on household surveys with limited spatial and and temporal temporal coverage
ItIt has has historically historically been been challenging challenging to to map map field-level field-level tillage tillage systems practices in smallholder systems given that the size of fields is typically smaller than the practices in smallholder systems given that the size of fields is typically smaller than the spatial resolution resolution of of readily readily available available satellite satellite imagery, imagery, such such as as Landsat

Summary

Introduction

One of the main components of conservation agriculture, has been shown to increase crop yields, increase farmer profits, reduce soil and water loss, and reduce environmental footprints in many agricultural systems across the globe [1,2,3,4,5]. Small-scale case studies have suggested that zero tillage adoption is increasing in many agricultural systems worldwide [3,6]. There is limited understanding of the extent and location of zero tillage adoption at regional and national scales, in smallholder systems where ground or census data on tillage practices are typically unavailable. Remote sensing offers the ability to map zero tillage adoption at large spatio-temporal scales, and at the field scale (e.g., [7,8]). Despite the benefits of using satellite data to map tillage practices, to date no studies have used remote sensing to map field-level tillage practices in smallholder systems.

Methods

Results

Conclusion