Abstract. The increased frequency and magnitude of extreme rainfall events due to anthropogenic climate change, and decadal and multi-decadal climate variability question the stationary climate assumption. The possible violation of stationarity in climate can cause erroneous estimation of design rainfalls derived from extreme rainfall frequency analysis. This may result in significant consequences for infrastructure and flood protection projects since design rainfalls are essential input for design of these projects. Therefore, there is a need to conduct frequency analysis of extreme rainfall events in the context of non-stationarity, when nonstationarity is present in extreme rainfall events. A methodology consisting of threshold selection, extreme rainfall data (peaks over threshold data) construction, trend and nonstationarity analysis, and stationary and non-stationary generalised Pareto distribution (GPD) models was developed in this paper to investigate trends and non-stationarity in extreme rainfall events, and potential impacts of climate change and variability on intensity-frequency-duration (IFD) relationships. The methodology developed was successfully implemented using rainfall data from an observation station in Melbourne (Australia) for storm durations ranging from 6 min to 72 h. Although statistically significant trends were detected in extreme rainfall data for storm durations of 30 min, 3 h and 48 h, statistical non-stationarity tests and non-stationary GPD models did not indicate non-stationarity for these storm durations and other storm durations. It was also found that the stationary GPD models were capable of fitting extreme rainfall data for all storm durations. Furthermore, the IFD analysis showed that urban flash flood producing hourly rainfall intensities have increased over time.
Abstract. The increased frequency and magnitude of extreme rainfall events due to anthropogenic climate change, and decadal and multi-decadal climate variability question the stationary climate assumption. The possible violation of stationarity in climate can cause erroneous estimation of design rainfalls derived from extreme rainfall frequency analysis. This may result in significant consequences for infrastructure and flood protection projects since design rainfalls are essential input for design of these projects. Therefore, there is a need to conduct frequency analysis of extreme rainfall events in the context of non-stationarity, when non-stationarity is present in extreme rainfall events. A methodology consisting of, threshold selection, extreme rainfall data (peaks over threshold data) construction, trend and non-stationarity analysis, and stationary and non-stationary Generalized Pareto Distribution (GPD) models was developed in this paper to investigate trends and non-stationarity in extreme rainfall events, and potential impacts of climate change and variability on Intensity–Frequency–Duration (IFD) relationships. The developed methodology was successfully implemented using rainfall data from an observation station in Melbourne (Australia) for storm durations ranging from 6 min to 72 h. Although statistically significant trends were detected in extreme rainfall data for storm durations of 30 min, and 3 and 48 h, statistical non-stationarity tests and non-stationary GPD models did not indicate non-stationarity for these storm durations and other storm durations. It was also found that the stationary GPD models were capable of fitting extreme rainfall data for all storm durations. Furthermore, the IFD analysis showed that urban flash flood producing hourly rainfall intensities have increased over time.
Abstract:The objective of this research is the assessment of the efficiency of a non-linear regression technique in predicting long-term seasonal rainfall. The non-linear models were developed using the lagged (past) values of the climate drivers, which have a significant correlation with rainfall. More specifically, the capabilities of SEIO (South-eastern Indian Ocean) and ENSO (El Nino Southern Oscillation) were assessed in reproducing the rainfall characteristics using the non-linear regression approach. The non-linear models developed were tested using the individual data sets, which were not used during the calibration of the models. The models were assessed using the commonly used statistical parameters, such as Pearson correlations (R), root mean square error (RMSE), mean absolute error (MAE) and index of agreement (d). Three rainfall stations located in the Australian Capital Territory (ACT) were selected as a case study. The analysis suggests that the predictors which has the highest correlation with the predictands do not necessarily produce the least errors in rainfall forecasting. The non-linear regression was able to predict seasonal rainfall with correlation coefficients varying from 0.71 to 0.91. The outcomes of the analysis will help the watershed management authorities to adopt efficient modelling technique by predicting long-term seasonal rainfall.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.