During the last two decades, the frequency of extreme events has increased and future projections of climate show that this is likely to continue. World population is growing and we need to increase…

This is a $377,608 research grant awarded by the National Aeronautics and Space Administration (NASA) to the University of Maryland, College Park. The grant supports research to analyze the influence of extreme weather events on regional crop yields around the world using satellite data from sensors such as MODIS and VIIRS. The goal is to improve understanding of how extreme events impact global food production, especially for major commodity crops like wheat and corn. The research will leverage historical satellite data records and crop yield estimation methods to identify agricultural regions susceptible to extreme weather. This work will advance the current science of production estimation by calculating uncertainties in the crop yield estimates. The grant does not have a set-aside designation.

During the last two decades, the frequency of extreme events has increased and future projections of climate show that this is likely to continue. World population is growing and we need to increase food supply to cover the increased food demand. Policy-makers are thus very much concerned about long-term food security. In this context, we need to understand how extreme weather events affect global crop production, the periods that crops are more sensitive to them and how much production is impacted. We have reliable statistics in the US, but there's limited availability of reliable historical and independent crop statistics for other important countries/regions in the world with different varieties and climate characteristics and as a result, the global picture is unclear. Taking advantage of coarse resolution satellite data records and crop yield estimation methods (errors 5-10%), we can learn from past events and improve the knowledge needed to be able to react in the future. In recent years, agricultural monitoring using remote sensing data has gained increasing attention from the science community, due in part to the impact of extreme weather events on crop production around the world (2008, 2012, 2016) and improvements to the global satellite data record. Building on our experience with remote sensing of crop yield using coarse resolution sensors (modis, viirs and avhrr), this proposal will analyze the influence of extreme events on regional crop yield around the world, for the satellite record and its temporal and spatial variability. We will focus our analysis on the production of wheat and corn, two important global commodity crops and advance the current science of production estimation by calculating uncertainties of our crop yield estimates, as previous approaches usually do not report associated uncertainties. Although reliable crop production data from ground survey are available for the US, this is not the case for other countries and regions, during the c. 40-year satellite data record. The higher-quality modis and viirs-ERA data and meteorological data will be used as a baseline, to refine our methods for estimating crop production. We will then include a data mining exercise adding the recently enhanced avhrr long term data record (ltdr) records to increase the range of extreme weather events and provide a better understanding of their impact on crop production, identifying those agricultural regions which are more susceptible to extreme events. We will explore the impact of the timing and duration of extreme events on crop production in different parts of the growing season. Multiple satellite-derived features, such as vegetation indices and biophysical parameters estimated at a single date or accumulated over the crop growth period, will be analyzed, with an ndvi-based approach serving as a benchmark to connect them with crop yield at regional scale. Given that cropland areas and crop type change over time due to agricultural land use change, we will develop yearly crop masks built with modis data to drive the analysis and during the pre-modis period, using data from the landsat archive. The proposal will include a NASA terra aqua suomi NPP land discipline lead proposal from chris justice.