The potential of satellite-observed crop phenology to enhance yield gap assessments in smallholder landscapes

Abstract

Many of the undernourished people on the planet obtain their entitlements to food via agricultural-based livelihood strategies, often on underperforming croplands and smallholdings. In this context, expanding cropland extent is not a viable strategy for smallholders to meet their food needs. Therefore, attention must shift to increasing productivity on existing plots and ensuring yield gaps do not widen. Thus, supporting smallholder farmers to sustainably increase the productivity of their lands is one part of a complex solution to realizing universal food security. However, the information (e.g., location and causes of cropland underperformance) required to support measures to close yield gaps in smallholder landscapes are often not available. This paper reviews the potential of crop phenology, observed from satellites carrying remote sensing sensors, to fill this information gap. It is suggested that on a theoretical level phenological approaches can reveal greater intra-cropland thematic detail, and increase the accuracy of crop extent maps and crop yield estimates. However, on a practical level the spatial mismatch between the resolution at which crop phenology can be estimated from satellite remote sensing data and the scale of yield variability in smallholder croplands inhibits its use in this context. Similarly, the spatial coverage of remote sensing-derived phenology offers potential for integration with ancillary spatial datasets to identify causes of yield gaps. To reflect the complexity of smallholder cropping systems requires ancillary datasets at fine spatial resolutions which, often, are not available. This further precludes the use of crop phenology in attempts to unpick the causes of yield gaps. Research agendas should focus on generating fine spatial resolution crop phenology, either via data fusion or through new sensors (e.g., Sentinel-2) in smallholder croplands. This has potential to transform the applied use of remote sensing in this context.

Cropland yield divergence over Africa and its implication for mitigating food insecurity

“…Multiple sensors, especially the Moderate Resolution Imaging Spectroradiometer (MODIS) have generated time-series of remote sensing imagery that enable monitoring of the intra-annual, and inter-annual, dynamics of vegetation growth. The repeat coverage of remote sensing enables extracting the key points of crop growth period at pixel level to increase the accuracy of simulating crop yields (Duncan et al 2015b). Satellite data also enable appropriate representation of spatially heterogeneous agricultural systems.…”

Crop Area Mapping Using 100-m Proba-V Time Series

“…This information has significant environmental, policy, agricultural and economic implications for most national governments, since crop production figures are used for determining the amount of food to import or export at the end of the growing season [1,2]. The error introduced to crop production estimation from general agricultural land cover maps is minimized with accurate crop extent maps [1,[3][4][5]. For remote sensing-based crop production estimates, the ideal approach would be to combine biomass proxies and crop maps.…”

“…Discriminating croplands from non-croplands and identifying different crop types can be achieved with remote sensing-based crop growth monitoring and in particular with indices that quantify the distinct green-up and senescence of the crop cycle [5]. Since different crops show different spectral responses depending on their maturity stage, the temporal dimension of remote sensing data is most…”

The Accuracy of Self-Reported Crop Yield Estimates and Their Ability to Train Remote Sensing Algorithms

“…There is a long history in remote sensing of mapping agricultural characteristics. For example, at regional and global scales, satellite data have been used to map the extent of croplands (Waldner et al, 2016), crop management practices (Bégu et al, 2018), biomass and yield Jain et al, 2016), crop phenology (Duncan et al, 2015), and crop stress (Kannan et al, 2017;Paliwal et al, 2019). Recent advancements in remote sensing, including cloud computing, the increased use of machine learning, and finer spatial, temporal, and spectral resolution data have only increased what is possible to map over the last decade (Ma et al, 2019).…”