Identifying El Niño–Southern Oscillation influences on rainfall with classification models: implications for water resource management of Sri Lanka

Abstract

Abstract. Seasonal to annual forecasts of precipitation patterns are very important for water infrastructure management. In particular, such forecasts can be used to inform decisions about the operation of multipurpose reservoir systems in the face of changing climate conditions. Success in making useful forecasts is often achieved by considering climate teleconnections such as the El Niño–Southern Oscillation (ENSO) and Indian Ocean Dipole (IOD) as related to sea surface temperature variations. We present a statistical analysis to explore the utility of using rainfall relationships in Sri Lanka with ENSO and IOD to predict rainfall to the Mahaweli and Kelani River basins of the country. Forecasting of rainfall as the classes flood, drought, and normal is helpful for water resource management decision-making. Results of these models give better accuracy than a prediction of absolute values. Quadratic discrimination analysis (QDA) and classification tree models are used to identify the patterns of rainfall classes with respect to ENSO and IOD indices. Ensemble modeling tool Random Forest is also used to predict the rainfall classes as drought and not drought with higher skill. These models can be used to forecast the areal rainfall using predicted climate indices. Results from these models are not very accurate; however, the patterns recognized provide useful input to water resource managers as they plan for adaptation of agriculture and energy sectors in response to climate variability.

Recent rainfall trend over Sri Lanka (1987–2017)

“…Seasonal analysis showed increase in rainfall during FIM, SIM and NEM, but there is decrease in rainfall during SWM. SWM rain brings more rainfall to wet zone (De Silva and Hornberger, ) and it is the source of water of main perennial rivers in Sri Lanka and it contribute in hydro power generation and irrigating larger area of dry zone Sri Lanka. Thus decrease in rainfall during SWM period likely to affect the economy of the country.…”

“…Cyclic climate teleconnections such as El Niño‐Southern Oscillation (ENSO), Indian Ocean Dipole (IOD) and Quasi‐Biennial Oscillation may also influence these trends. It is reported that during El Niño event rainfall increases for the first three months of the Maha season (October–December), and decreases during the last three months (January–March) (Ropelewski and Halpert, ; De Silva and Hornberger, ). Similarly, La Niña events enhance the rainfall during the SWM (Whitaker et al ., ; De Silva and Hornberger, ), whereas during El Niño events the SWM rainfall is reduced (Chandimala and Zubair, ; De Silva and Hornberger, ).…”

“…IOD also play an important role in modulating the precipitation over Sri Lanka. There is high probability of occurrence of drought when both IOD and MEI (multivariate ENSO index) are positive and non‐occurrence of drought when both IOD and MEI are negative (De Silva and Hornberger, ). Therefore, these rainfall factors influence the variation of the rainfall trend over the Sri Lanka.…”

Prediction and Validation of Rainfall Classes for Vaigai River Catchment using El Nino

“…The quadratic discriminate function is derived by taking the logarithmic values, as given in (7). For class k(π k ), prior probabilities are calculated by taking the frequency of data points of class k in the training data, as in (8). If we consider N number of points totally in the training observations, then N k represents the number of observations corresponding to the k th class.…”

“…Studies made by Yadav et al [7] found that the inter annual variability of North West India Winter Precipitation is influenced by Arctic Oscillation / North Atlantic Oscillation (AO/NAO) and El Nino -Southern Oscillation (ENSO) phenomena. T. De Silva M et al [8] have made an attempt to forecast the rainfall of Mahaweli and Kelani River basins of Sri Lanka. Cao et al [9] investigated the rainy-season precipitation in China based on the influence of five El Nino-Southern Oscillations indices.…”