Submitted to: Climate and Weather Research Workshop Proceedings
Publication Type: Proceedings
Publication Acceptance Date: November 7, 1995
Publication Date: N/A
Interpretive Summary: The linked ocean-atmosphere system, El Nino - Southern Oscillation (ENSO), is shown to influence precipitation in the southwestern United States and northern Mexico, especially during the fall and winter. Knowledge of this kind allows scientists to understand more accurately how ENSO affects the timing of precipitation in this region. These findings will help weather forecasters predict precipitation and river flows, and will be valuable to water managers as they plan for residential and irrigation needs and operate reservoirs for power generation and recreation.
Technical Abstract: The responses of parameters of a Markov chain-mixed exponential daily precipitation model to seasonally-varying perturbations of the Southern Oscillation Index (SOI) were evaluated, extending the work of Woolhiser et al. (1993). The perturbations on model parameters are defined by G'(t) = G(t) + bG(k)SOI(t-lG(k)), where G(t) is the unperturbed model parameter, SOI(t-lG(k)) is the antecedent SOI lagged by lG(k) days by precipitation day t, bG(k) is a coefficient and k is an index for seasons (months, 1,...,12). The algebraic sign of the coefficient, bG(k), of the linear SOI perturbation term may be used to infer increases or decreases in modeled precipitation. Several months at each of three stations in Arizona, USA and Sonora, Mexico have significant increases in the log likelihood function for both occurrence process and amount per wet day due to SOI perturbations. Negative SOI is shown to be linked to increase modeled precipitation during the winter season (November through April).