Title
Using spatial-temporal fields as inputs for numerical models: Global warming and its agricultural impacts
Speaker
Doug Nychka, Geophysical Statistics Project, National Center for Atmospheric Research
Abstract
A point sometimes missed in spatial statistics is that the estimated spatial fields may be an intermediate result, not an end in themselves. For example, to study the impact of changing weather patterns on agriculture one uses estimated meteorological fields, such as temperature and precipitation, as the inputs to numerical crop models. The resulting crop yields help to quantify the direct impact of a changing environment on the agricultural industry. This talk will present an overview of climate models and discuss how statisticians can play a role in assessing climate change predictions. As a specific example, we will present research constructing a space-time model for meteorological variables. These models, known as weather generators, involve an (interesting) mix of multivariate time series and spatial statistics techniques. A daily weather generator is presented that includes discrete components for precipitation occurence and is used to study the temporal and spatial distribution of corn yields in the Southeast US. Modifying the weather generator allows one to study different changes in climate.
Meet the speaker in Room 212 Cockins Hall at 4:30 p.m. Refreshments will be served.
