Authors: Michael Cecil*, Clark University
Topics: Agricultural Geography, Geographic Information Science and Systems, Africa
Keywords: maize, yield, DSSAT, Zambia, agriculture
Session Type: Poster
Start / End Time: 5:20 PM / 7:00 PM
Room: Napoleon Foyer/Common St. Corridor, Sheraton, 3rd Floor
Presentation File: Download
Variability in crop yield has important implications for food security in sub-Saharan Africa. In the context of climate change adaptation, much of the focus on agricultural decision-making has been on inter-annual decisions, such as crop selection and how much land should be farmed. However, farmers also make significant decisions during the course of individual seasons in response to intra-annual variability in precipitation. This study quantifies the effect within-season management decisions have on yields relative to the effect of intra-seasonal weather variability.
In this study, we attribute maize yield variability to three factors: (1) land management (planting date, fertilization); (2) intra-seasonal climate variability; (3) soil and topography, using the Decision Support System for Agrotechnology Transfer (DSSAT) cropping system model. We quantify the percent of variability in maize yield that can be attributed to factors within farmers' control (planting date, fertilization, and potentially cultivar selection), versus the variability that is caused by within season weather (e.g. drought or flood events), or soil conditions. This study uses historical weather data in Zambia's Southern Provinces to simulate a wide range of variability in yearly climate.
We also plan to incorporate other data sources, including farmer responses to mobile phone surveys (SMS data), environmental pod sensors, and agricultural sensor data to create more realistic simulations. The SMS data will be used to compare farmers' perceptions of precipitation variability to other data sources and also compare expected and simulated maize yield.