Generating Cropland Extent of Ethiopia from High-Resolution Imagery

Principal Investigator

Andy Nelson, University of Twente

Collaborator

Steffen Fritz, International Institute for Applied Systems Analysis

Geographic area of interest

Ethiopia

Description

Efforts to better target research and extension for sustainable agriculture require basic information on the extent and location of fields, which are the fundamental land management unit on which decisions are made regarding what, when and how to grow crops. Effective targeting of interventions in staple and cash crop systems requires accurate spatial information on where specific crops are grown. The spatial extent of cropland, at the level of detail where individual fields can be discriminated, is poorly defined, particularly in regions of the world where field sizes are small (1 ha or less), where cropping systems are extensive and diverse and where agricultural landscapes are fragmented. This is particularly true in developing countries and emerging economies where efforts are most needed to sustainably intensify agricultural systems to close prevalent yield gaps, boost productivity and improve livelihoods. Without this basic information on the location and extent of the agricultural land - which is the main source of income and calories for millions of low income families - there is a limit to how much impact or effectiveness can be expected from agricultural investments. Simply put, field boundary information is an essential layer of geospatial information for recommendation domains in agriculture. Furthermore, field boundaries are the building blocks for Land Information Systems that maintain information on land tenure and which support tenure security, good governance and investment opportunities in the agricultural sector. In cases like Ethiopia where Land Information Systems are poor or where such data are not readily available, then other methods must be used to derive fundamental information on cropland area and field boundaries.

Project goals

• Generate comprehensive field boundary information over a number of high resolution images across a diverse and representative set of croplands in Ethiopia
• Compare this crowd sourced field boundary information with ground collected boundary information and assess the accuracy, completeness, validity and consistency of the crowd-based data
• Make recommendations for where (semi-)automated methods would be more appropriate and where manual field boundary delineation is required for the range of observed cropland environments

Key Achievements (last update: Nov 2016)

None

Publications/presentations/workshops

None

Status

Will start from January, 2017