About AgriSense


AgriSense, a Spurring a Transformation for Agriculture through Remote Sensing (STARS) Project

GeoODK- Components

STARS is an international initiative of multiple institutions lead by the University of Twente ITC in the Netherlands and is funded by the Bill and Melinda Gates Foundation. AgriSense works as part of STARS in Tanzania and Uganda and is lead by the University of Maryland (UMD) and Sokoine University of Agriculture (SUA). The project aims to use satellite remote sensing technologies for improving the basis and flow of information for agricultural monitoring and food security forecasting by the National Food Security Department in Ministry of Agriculture (MAFC) of Tanzania. AgriSense develops and deploys cutting-edge technologies for agricultural monitoring to the MAFC and other stakeholders.


GLAM - East Africa

GLAM East Africa- Components

The core technology of Agri-Sense Africa is an online system for the automated processing of MODIS satellite image time series and the production of NDVI time series graphs for the detection of low and high production areas in the country. The processing chain is implemented within the GLAM-East Africa portal (UMD Global Agricultural Monitoring System, customized for this project to Tanzania and East Africa). GLAM East Africa is a user-friendly, automated portal for MODIS and Landsat time series analysis to support reporting and crop condition monitoring; GLAM East Africa will be used to track crop condition throughout the growing season. GLAM allows MAFC staff to generate NDVI time-series graph and maps that give an overview of the crop conditions on the ground.

View GLAM » View news »

Tanzania Crop Monitor

Tanzania CropMonitor

The Tanzania Crop Monitor portal is a tool which allows regional and district analysts enter and submit their summary observations and reports on local crop conditions. The Crop Monitor for Tanzania will facilitate the development of a national monthly bulletin that will provide timely, coordinated national information, on crop conditions as the season develops in a format that is straight forward for uptake and usable to inform agricultural policies and planning. Informed decisions, such as on export restrictions, quantities of government stocks, distribution of grains and stocks, mobilization of food aid, targeted agricultural programs, will have direct implication on small holder farmers, who would benefit from more informed government decisions and policies.

View Crop Monitor »




The Geo-Wiki platform provides an innovative solution to the problem of interpreting large volumes of very high resolution imagery. As part of the AgriSense, IIASA took this application to Sokoine University of Agriculture (SUA), and invited students to attend training workshops. The students were trained on how to use the application for the classification of land use (i.e. cropland, woodland, both or neither), dominant field size and cultivation stage (cultivated, fallow, mixed, unclear or uncultivated). Learn More

Area Frame

Area Frame

Agri-Sense Africa is working on the design of area and point sample area frames for Kilosa district and the Morogoro region for within-season agriculture monitoring for food security and yield forecasting. Area sampling frames establish representative sample segments for field data collection using probability sampling techniques minimizing errors common in surveys and censuses that require extensive field work. Field data collection guided by a probability sampling protocol is more precise and better controlled, providing information that is credible, accurate and timely at a reduced cost. The stratification by agricultural intensity for designing the area sampling frame is based on very high resolution satellite imagery of the WordView-2 and Ikonos satellites provided by Digital Globe. During the pilot project data on crop type, area, condition, drought, pests and diseases and market prices will be carried out at each sampling location. The sampling frame lends itself to the collection of other parameters important for food security including household data and data on food storage from previous growing seasons.

Landsat Image Classification

Landsat Image Classification

A classification product created through Agrisense Africa is an agriculture mask for Tanzania at Landsat resolution. This nation-wide cropland mask at 30 m Landsat resolution was developed using composited Landsat tiles from a 2010 - 2013 time series. Decision tree classifier methods were used in the classification. Representative training areas were then collected for agriculture and no agriculture areas using appropriate indices to separate these classes. Validation was done using a random sample and Google Earth to compare Agriculture and No agriculture samples from the study area. This agriculture mask is an important part of agricultural monitoring providing the basis for GLAM- Tanzania NDVI analysis.

In-Situ Measurements



AgriSense is piloting the use of the GeoODK Android Application with representatives of the Food Security Department at MAFC, agricultural officers from the regional and district offices, and selected agricultural extension agents from within the Morogoro region. GeoODK’s smartphone based electronic field data collection system allows for faster and more efficient collection of data and fosters easier submission of field data to Government Ministry offices and decision-makers. The pilot project initially uses the traditional paper forms translated into electronic format. The GeoODK system will significantly increase the efficiency of data collection and delivery by directly submitting the data from the smartphone or tablet to the online database. The data are instantaneously accessible to any authorized entity within the government and can be used to integrate with other data sources and satellite remote sensing information. Field Extension officers are currently submitting electronic forms that give a general overview of the field conditions including pest and diseases. Learn more

View news »

Unmanned Aerial Vehicles (UAVs or Drones)


AgriSense uses Unmanned Aerial Vehicles (UAVs) to complement the field data collection by acquiring time series of extremely high resolution remote sensing imagery (4 cm resolution) over a number of 1x1 km test sites. The UAV imagery supports the satellite image interpretation. The project also explores the use of UAVs for cropping system mapping and crop condition assessments from multi-spectral UAV imagery. Both fixed-wing (small airplanes) and rotary wing (helicopter-type) UAVs are being used.