| description abstract | Monthly precipitation at 46 stations located throughout the state of Texas was examined over a continuous 62-year period from 1923 to 1984. Precipitation data were subjected to spectral, empirical orthogonal function (EOF) and correlation analyses. Focus was on the dominant EOF (E1), which explains 30% to 45% of all precipitation variance. The time-dependent coefficient associated with E1 closely resembles a statewide average precipitation index. This time-dependent coefficient undergoes large month-to-month fluctuations; however, these fluctuations are, for the most part, aperiodic. Other than slight month-to-month persistence during winter and spring, monthly precipitation anomalies cannot be predicted or anticipated based on time-series or spectral analysis. A long-term monthly mean sea level pressure dataset is composited over anomalously wet and anomalously dry months covering the 62 years. A signal is found in the sea level pressure composites, which is best defined during winter months. Anomalously wet (dry) months in Texas are associated with a northward (southward) shift of high pressure. Simultaneous correlations between monthly statewide precipitation and temperature indicate strong negative (greater than ?0.60) correlations during the warm season. However, lag correlations suggest that precipitation is controlling temperature. Methods to forecast monthly precipitation in Texas remain elusive. | |
