The potential of very high spatial resolution remote sensing in applications in smallholder agriculture

Smallholder farmers contribute to more than 50% of the world's annual production of cereals, meat and dairy products and they cultivate more than 80% of the total agricultural area in Asia and Sub-Saharan Africa. Conversely, it is estimated that more than 2.5 billion people depend directly on the agricultural sector. Earth Observation has been a tool for agronomists with ever-increasing capability, however, the case of agriculture in low-income countries imposes challenges, for instance in recognizing mixed- and inter-cropping, the small size of the farm plots and lack of sound crop management systems, which increase the uncertainty of information derived from remotely sensed data. The Spurring a Transformation for Agriculture through Remote Sensing (STARS) project aims to investigate the potential of Very High Spatial Resolution remote sensing in delivering data products that can better inform decisions around smallholder agriculture. Data are collected throughout the growing season via in-situ measurements, Unmanned Aerial Vehicles equipped with NIR and multispectral cameras and VHSR satellite images. One important STARS' objective is to analyze the spectral information of this multi-scale and multi-temporal dataset and establish implementation flows to support stakeholders in their decision making and economic development. In this study, we demonstrate the usefulness of VHSR images in mapping and monitoring smallholder crop fields in Bangladesh. UAV-based and satellite data at a variety of spatial scales are presented and the degree to which delineation of crop fields and related agronomical information can be extracted is discussed. This study aims to demonstrate potentials and limitations of the use of remote sensing for monitoring crop fields of smallholder farmers.