Cropland Data to Improve Yield Forecasting in Support of Food Security Status, Tanzania
By Hawa Adinani
The use of satellite images in crop monitoring is developing rapidly in larger-scale farming along with the use of remote sensing data i.e vegetation index analysis. However, there is an increasing need for ground truthing data to make sure the monitoring of crops is precise.
In collaboration with Collecte Localisation Satellites (CLS), OpenMap Development Tanzania has conducted a survey to collect field harmonized training data (also called ground truth data) for the classification of crop types and the provision of unbiased crop area estimates and the validation of the crop type maps and crop mask in 3 regions of Tanzania (Dodoma, Manyara and Tanga).
The mapping is part of the Group on Earth Observations Global Agricultural Monitoring Initiative (GEOGLAM) with the broad goal of providing open, timely, and science-driven information on crop conditions in support of market transparency and early warning of production shortfalls. The first campaign was carried out from February to March 2021 and the second in December 2021. The regions have been selected to represent the diverse nature of crops.
OMDTZ organized the field activities in the mentioned regions and collected crop data by using OpenDataKit — a free android application for data collection. The team visited crop fields identified and collect relevant information including;
- The context of the sample with field characteristics i.e geolocation, presence of cropland, crop type, crop stage, irrigation/rainfed type and cropping pattern.
- The crop characteristics such as a photo indicating the details of the crops like crop stage or field preparation.
The collected data will help in creating high-resolution cropland masks that will improve yield forecasting and support the status of food security. The cropland data can be used by the ministry of agriculture and other relevant sectors in providing appropriate agricultural support i.e infrastructure and technical advice to optimize productivity based on the available data and ensure increased yields and food security.