Projected Water Levels and Identified Future Floods: A Comparative Analysis for Mahaweli River, Sri Lanka
نویسندگان
چکیده
The Rainfall-Runoff (R-R) relationship is essential to the hydrological cycle. Sophisticated models can accurately investigate R-R relationships; however, they require many data. Therefore, machine learning and soft computing techniques have taken attention in environment of limited hydrological, meteorological, geological accuracy such depends on various parameters, including quality inputs outputs used algorithms. However, identifying a perfect algorithm still challenging. This study develops fuzzy logic-based called Cascaded-ANFIS predict runoff based rainfall. model was compared against three regression algorithms: Long Short-Term Memory, Grated Recurrent Unit, Neural Networks. These algorithms been selected due their outstanding performances similar studies. were tested Mahaweli River, longest Sri Lanka. results showcase that Cascaded-ANFIS-based outperforms other correlation coefficient each algorithm’s predictions 0.9330, 0.9120, 0.9133, 0.8915, 0.6811, 0.6734 for Cascaded-ANFIS, LSTM, GRU, RNN, Linear, Ridge, Lasso respectively. Hence, this concludes proposed 21% more accurate than second-best LSTM algorithm. In addition, Shared Socio-economic Pathways (SSP2-4.5 SSP5-8.5 scenarios) generate future rainfalls, forecast near-future mid-future water levels, identify potential flood events. forecasting indicate decrease events magnitudes both SSP2-4.5 scenarios. Furthermore, scenario shows drought weather from May August yearly. effectively be manage control resources mitigate damages.
منابع مشابه
El Niño–southern Oscillation Influences on the Mahaweli Streamflow in Sri Lanka
Despite advances over the last two decades in the capacity to predict the evolution of the El Niño–southern oscillation (ENSO) phenomenon and advances in understanding of the relationship between ENSO and climate, there has been little use of climate predictions for water resources management in the tropics. As part of an effort to develop such a prediction scheme, the ENSO influences on stream...
متن کاملFish Assemblage Structure of Two Contrasting Stream Catchments of the Mahaweli River Basin in Sri Lanka: Hallmarks of Human Exploitation and Implications for Conservation
Patterns of fish community composition in the Mahaweli ichthyological region of Sri Lanka were examined in agricultural tributaries of the Uma-oya catchment of the upper Mahaweli River in comparison to more pristine streams in a nature reserve in the Amban-ganga catchment. The Uma-oya catchment shows characteristics commonly observed in extensive agricultural exploitation such as impaired water...
متن کاملSri Lanka
Sri Lanka is a country with an approximate area of 66 thousand square kilometers (UNO, 2008). The population is 20,409,946 and the sex ratio (men per hundred women) is 98 (UNO, 2009). The proportion of the population under the age of 18 years is 28% and the proportion above age 60 is 8% (UNO, 2009). The literacy rate is 97% for men and 99% for women (UN Statistics, 2008). The life expectancy at...
متن کاملA bedside test for methemoglobinemia, Sri Lanka
PROBLEM Propanil is an aniline herbicide that is widely used for rice cultivation, but is also used for self-poisoning. Toxicity from propanil is largely due to methemoglobinemia. In resource-poor settings, the capacity to determine methemoglobin concentration is insufficient and prevents effective case management, which results in increased deaths from propanil poisoning. APPROACH Blood with...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE Access
سال: 2023
ISSN: ['2169-3536']
DOI: https://doi.org/10.1109/access.2023.3238717