Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Seasonal and interannual variations in spatial autocorrelation statistics of tropical precipitation
Abstract
Spatial autocorrelation of tropical precipitation is examined by the application of Moran's I index to the 17-year CMAP data set. Global I indices, which express the overall spatial autocorrelation of precipitation for the entire tropics, have an annual cycle characterized by bimodalism. A higher degree of spatial organization of tropical precipitation occurs during both boreal and austral summers. While the annual cycle of mean precipitation appears to be more unimodal, global I indices, therefore, provide an alternative definition of precipitation seasonality in the tropics. The seasonal variations in global I indices are small in the equatorial region, but are very large in subtropical latitudes. Local I indices which specify the spatial autocorrelation of precipitation in a small geographical area are also calculated at each individual grid cell using a moving window approach. Generally, high local I indices are associated with large precipitation gradients and low local I indices are related to areas of precipitation maxima or minima. Due to this unique property, local I indices are very useful in the identification of the core locations of large scale convective features such as the ITCZ and SPCZ. The interannual variations of tropical precipitation and associated spatial autocorrelation are affected by ENSO-related atmospheric/oceanic forcings. Changes in the local spatial autocorrelation of tropical precipitation during an ENSO event are induced by the shifting of convective centers (change of local precipitation) and the corresponding increase or decrease of the precipitation gradient for a location. Regions where precipitation variables are affected by ENSO include most of the tropical Pacific, Indonesia, Amazonia, Northeast Brazil and West Africa. These ENSO-related connections are strongest in DJF, when the SOI leads the precipitation variable and weakest in JJA, when the precipitation variables lead the SOI. Due to this relationship, JJA precipitation statistics may prove to be a useful predictor of future ENSO activities.
Subject Area
Geography|Atmosphere
Recommended Citation
Siu, Wai-lok, "Seasonal and interannual variations in spatial autocorrelation statistics of tropical precipitation" (1997). ETD collection for University of Nebraska-Lincoln. AAI9734642.
https://digitalcommons.unl.edu/dissertations/AAI9734642