Widespread Crop Failure in Moroto Karamoja
Imagery from NASA’s MODIS satellite and a recent field visit by Catherine Nakalembe reveal widespread crop failure in Moroto, Karamoja, this year. The local communities of subsistence farmers grow grain fed with minimal to no inputs. “We prepare the land in February, plant in March and wait for rain.” Rainfall in the region is erratic, weather forecasts are often inaccurate and there are no irrigation systems. Failed rains and “change in seasons” are the root cause of widespread crop failure and irreversible environmental destruction in the region.
A team lead by the University of Maryland visited the region during a field study in Moroto district from August 25 - September 9, 2014. Working with ESIPPS International Limited and the District Agricultural Officer Mr. John Olinga, the team found evidence of widespread crop failure in more than 70% at randomly selected locations in Rupa and Naduget sub-county. In July, time series of satellite data from Maryland’s Global Agriculture Monitoring System (GLAM) based on NASA’s MODIS satellite data indicated below average vegetation conditions (left) already long before the field visit. Late August to early harvesting time for sorghum, maize and beans - all staples in the region.
“The farmers tilled and sowed their land end of February expecting rain in the third week of March, unfortunately the rainfall started abruptly for 1-2 days in March…. nothing until the third week of May. Famers waited and struggled digging the ground but the rain didn’t arrive. Meager rains in June were not enough to support crop growth and in most parts of Rupa the sorghum is completely dry. Some of the crops recovered due to the rain in September, however, this critical rainfall happened far too late in the season. Farmers were supposed to be harvesting, but most of the sorghum had not even flowered. Serious food shortages were expected in Rupa and parts of Tapac, Nadunget and Katikekile sub-counties in Moroto district. This year some farmers have planted and replanted up to 3 times in some cases but the crop failure is widespread,” Olinga John, District Production Officer of Moroto, reported.
GLAM System data indicates below average conditions for most of East Africa. The areas shaded brown above, indicate below average condition. The figure to right is a closer look at the vegetation condition from early in the growing season. By April vegetation condition fell below average, a critical time in the growing season. The images below show the various conditions -September2014
A common survival strategy for drought years such as this is the sale of livestock; however, the entire northeastern region of Uganda is under quarantine due to an outbreak of Foot and Mouth Disease, which has affected a large number of animals in the region. Though the majority of the farmers cite animal sales as a livelihood alternative, this is not an option this year. Hence, the challenges to farmers and the whole population this year are enormous.
Although farmers agree that the rain season was “abnormal,” they continue to follow the same cropping calendar and have very little trust in weather forecasts. UMD is conducting research to develop an agriculture monitoring system for sites in Tanzania and Moroto District under the STARS project (Spurring a Transformation for Agriculture through Remote Sensing). STARS is led by the University of Twente in The Netherlands, and aims to identify how earth observation data products may help improve current information and decision support systems in the smallholder economies of sub-Saharan Africa and South Asia. With such a system in place, tailored to the local geography and farmer needs, early warning can be timelier and crop condition can be tracked at little to no cost by local governments.
AgriSense/STARS Tracks Impact of Late Rains on Crops in Central and Northern Tanzania
A primary goal of the University of Maryland (UMD)-lead AgriSense STARS project is developing and providing remote sensing observation tools in combination with new tools for field data collection to the Ministry of Agriculture, Food and Cooperatives of Tanzania (MAFC) whilst building capacity to use them. The novelty of AgriSense’s tools and methods are already being appreciated at MAFC. One of the tools is the GLAM-East Africa system, which provides satellite remote sensing observations from the MODIS sensor for tracking and monitoring crop conditions. The NDVI time series produced by the GLAM-East Africa system (UMD Global Agricultural Monitoring System, customized for this project to Tanzania and East Africa) showed unusually poor vegetation conditions in central and northern Tanzania three weeks into the main growing season, which in this area usually starts at the end of February. Figure A shows the NDVI time series graph for Same District in northern Tanzania in comparison to the long term average and the standard deviation. Figure B shows a UAV overflight image from March 18, 2015 over one of AgriSense’s study sites in the same district with bare earth and limited amounts of dry-planted fields, illustrating the poor conditions. Below-normal crop conditions and late development due to the delayed start of the rains are also evident in other regions across the country including Singda, Arusha, Simuyu and Morogoro and as such, the MAFC prospects of good harvest for the affected regions are low for this growing season.
This critical spatially explicit information provided through GLAM-East Africa can be used free of charge and crop analysts at MAFC can use it in their day-to-day monitoring. Catherine Nakalembe, a doctoral candidate at UMD and Jan Dempewolf, Assistant Research Professor at UMD, are training food security analysts at MAFC to use this system for complimenting, validating and confirming field reports.
AgriSense is a project at the University of Maryland on agricultural monitoring for food security in East Africa, lead by Jan Dempewolf and Inbal Becker-Reshef and funded by the Bill and Melinda Gates Foundation through an umbrella grant to the University of Twente/ITC in the Netherlands. AgriSense develops and deploys cutting-edge technologies for agricultural monitoring to the MAFC and other stakeholders. AgriSense is working closely with MAFC and with support from the implementing partner Sokoine University of Agriculture (SUA) in Morogoro on transforming agricultural monitoring for food security in Tanzania.
Figure: A shows significantly below average NDVI during the main growing season in Same District in northern Tanzania. The current season is shown in red, the long-term average from 2000-2015 is shown in purple and the dotted lines represent the standard deviation over the same time period. Figure B shows a natural-color UAV image obtained over the AgriSense study site in Same District with still bare soils three weeks into the main growing season.