Estimating smallholder crops production at village level from Sentinel-2 time series in Mali's cotton belt

Lambert, M J and Traore, P C S and Blaes, X and Baret, P and Defourny, P (2018) Estimating smallholder crops production at village level from Sentinel-2 time series in Mali's cotton belt. Remote Sensing of Environment (TSI), 216. pp. 647-657. ISSN 00344257

[img]
Preview
PDF (It is an Open Access article) - Published Version
Download (3MB) | Preview

Abstract

In Mali's cotton belt, spatial variability in management practices, soil fertility and rainfall strongly impact crop productivity and the livelihoods of smallholder farmers. To identify crop growth conditions and hence improve food security, accurate assessment of local crop production is key. However, production estimates in heterogeneous smallholder farming systems often rely on labor-intensive surveys that are not easily scalable, nor exhaustive. Recent advances in high-resolution earth observation (EO) open up new possibilities to work in heterogeneous smallholder systems. This paper develops a method to estimate individual crop production at farm-to-community scales using high-resolution Sentinel-2 time series and ground data in the commune of Koningue, Mali. Our estimation of agricultural production relies on (i) a supervised, pixel-based crop type classification inside an existing cropland mask, (ii) a comparison of yield estimators based on spectral indices and derived leaf area index (LAI), and (iii) a Monte Carlo approach combining the resulting unbiased crop area estimate and the uncertainty on the associated yield estimate. Results show that crop types can be mapped from Sentinel-2 data with 80% overall accuracy (OA), with best performances observed for cotton (Fscore 94%), maize (88%) and millet (83%), while peanut (71%) and sorghum (46%) achieve less. Incorporation of parcel limits extracted from very high-resolution imagery is shown to increase OA to 85%. Obtained through inverse radiative transfer modeling, Sen2-Agri estimates of LAI achieve better prediction of final grain yield than various vegetation indices, reaching R2 of 0.68, 0.62, 0.8 and 0.48 for cotton, maize, millet and sorghum respectively. The uncertainty of Monte Carlo production estimates does not exceed 0.3% of the total production for each crop type.

Item Type: Article
Divisions: Research Program : West & Central Africa
CRP: CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS)
Uncontrolled Keywords: Smallholder crops, Sentinel-2, crop production, smallholder agriculture, cotton belt, Mali, crop yield, agricultural production
Subjects: Others > Smallholder Farmers
Others > GIS Techniques/Remote Sensing
Others > Smallholder Agriculture
Others > African Agriculture
Others > Mali
Depositing User: Mr Ramesh K
Date Deposited: 27 Aug 2018 10:53
Last Modified: 27 Aug 2018 10:55
URI: http://oar.icrisat.org/id/eprint/10845
Official URL: http://dx.doi.org/10.1016/j.rse.2018.06.036
Projects: UNSPECIFIED
Funders: FNRS (Fonds National de Recherche Scientifique belge)
Acknowledgement: This research was funded by a FNRS (Fonds National de Recherche Scientifique belge) grant for M-J Lambert. The field campaign realized by M-J Lambert in October and November 2016 was financed by the FNRS short stay abroad. The success of the field campaign largely depended upon the collaborations with ICRISAT (in particular Pierre Sibiry Traoré, Issa Kassogue and Kadiatou Touré) and the AMEDD NGO (Birama Sissoko, Gilbert Dembélé, and all the field operators). This publication was made possible (in part) by the STARS project, an integrated effort to improve our understanding of the use of remote sensing technology in monitoring smallholder farming supported by the Bill and Melinda Gates Foundation. Additional support was provided by the CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS) (please refers to https://ccafs.cgiar.org/donors for information about the fundings). The authors also acknowledge the critical provision of digitized field vector layers by the Université de Sherbrooke.
Links:
View Statistics

Actions (login required)

View Item View Item