Future Climate of Colombo Downscaled with SDSM-Neural Network

Dorji, Singay; Herath, Srikantha; Mishra, Binaya Kumar

doi:10.3390/cli5010024

Cited by 21 publications

(7 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such long‐term severe drought conditions can have detrimental impacts on water resources and therefore on socio‐economic sectors (Heydari et al, 2018; Veijalainen et al, 2019). In addition, these drought conditions can also affect agricultural and food production systems in water‐limited regions due to the high dependency of these systems on water (Dorji et al, 2017; Paymard et al, 2019). Furthermore, all living organisms in the region can be seriously threatened by such long‐term severe drought conditions.…”

Section: Discussionmentioning

confidence: 99%

Past and future drought trends, duration, and frequency in the semi‐arid Urmia Lake Basin under a changing climate

Mirgol

Nazari

Etedali

et al. 2021

Meteorological Applications

View full text Add to dashboard Cite

Although the Urmia Lake has undergone remarkable drought conditions within the past decades mainly due to climate change, drought studies covering the entire Urmia Lake Basin and all drought aspects are lacking. The present study investigated the spatial and temporal drought conditions in the Urmia Lake Basin for the past and future (2021-2050 and 2051-2080) periods using five general circulation models (GCMs) under the IPCC (Intergovernmental Panel on Climate Change) scenarios RCP 2.6, RCP 4.5, and RCP 8.5. The standardized precipitation index (SPI) and standardized precipitation and evapotranspiration index (SPEI) were compared. The SPEI predicted more drought events than the SPI, and it seemed to be a more suitable drought index than the SPI for the basin. In general, the future periods would encounter less drought conditions in terms of significant drought trends and duration than the observed period under all scenarios, but the frequency of quarterly severe drought events in the future periods would be higher than in the observed period. Furthermore, the stations Urmia (western bank) and Tabriz and Maragheh (eastern banks) would face the highest frequency of different types of quarterly drought events in the future periods compared with the observed period. The predicted high frequencies of drought events for the future periods can intensify the current low water level situation of the Urmia Lake, which seriously threatens all types of ecosystems in the basin. Therefore, serious actions need to be taken into account for efficient ecosystem and water resources management in the basin.

show abstract

Section: Discussionmentioning

confidence: 99%

Past and future drought trends, duration, and frequency in the semi‐arid Urmia Lake Basin under a changing climate

Mirgol

Nazari

Etedali

et al. 2021

Meteorological Applications

View full text Add to dashboard Cite

show abstract

“…is study also came to a similar conclusion. Although neural networks have been popularly applied for downscaling [39,40], they were first tested on MOS. Compared with MLP and RF, the results of the MLP show a higher PCC and a lower RMSE.…”

Section: Statistical Metricmentioning

confidence: 99%

Downscaling and Projection of Multi-CMIP5 Precipitation Using Machine Learning Methods in the Upper Han River Basin

Chen

et al. 2020

Advances in Meteorology

View full text Add to dashboard Cite

Downscaling considerably alleviates the drawbacks of regional climate simulation by general circulation models (GCMs). However, little information is available regarding the downscaling using machine learning methods, specifically at hydrological basin scale. This study developed multiple machine learning (ML) downscaling models, based on a Bayesian model average (BMA), to downscale the precipitation simulation of 8 Coupled Model Intercomparison Project Phase 5 (CMIP5) models using model output statistics (MOS) for the years 1961–2005 in the upper Han River basin. A series of statistical metrics, including Pearson’s correlation coefficient (PCC), root mean squared error (RMSE), and relative bias (Rbias), were used for evaluation and comparative analyses. Moreover, the BMA and the best ML downscaling model were used to downscale precipitation in the 21st century under Representative Concentration Pathway 4.5 (RCP4.5) and RCP8.5 scenarios. The results show the following: (1) The performance of the BMA ensemble simulation is clearly better than that of the individual models and the simple mean model ensemble (MME). The PCC reaches 0.74, and the RMSE is reduced by 28%–60% for all the GCMs and 33% compared to the MME. (2) The downscaled models greatly improved station simulation performance. Support vector machine for regression (SVR) was superior to multilayer perceptron (MLP) and random forest (RF). The downscaling results based on the BMA ensemble simulation and SVR models were regarded as the best performing overall (PCC, RMSE, and Rbias were 0.82, 35.07, mm and −5.45%, respectively). (3) Based on BMA and SVR models, the projected precipitations show a weak increasing trend on the whole under RCP4.5 and RCP8.5. Specifically, the average rainfall during the mid- (2040–2069) and late (2070–2099) 21st century increased by 3.23% and 1.02%, respectively, compared to the base year (1971–2000) under RCP4.5, while they increased by 4.25% and 8.30% under RCP8.5. Additionally, the magnitude of changes during winter and spring was higher than that during summer and autumn. Furthermore, future work is recommended to study the improvement of downscaling models and the effect of local climate.

show abstract

“…The Statistical DownScaled Model (SDSM) projected an annual increase in the average temperature of 2.83°C in the 2080s for Colombo under Representation Concentration Pathways (RCP 8.5) compared to the 1961-1990 period. The annual increase in rainfall is 33% (Dorji et al 2017). From 2020 to 2100 under RCP 2.6, 4.5 and 8.5 scenarios, the future annual precipitation in the Hanwella sub-catchment of the Kelani River Basin will have variations between 2100 to 5130 mm (Dissanayake and Rajapakse 2017).…”

Section: Introductionmentioning

confidence: 99%

Analysis of Statistical Downscaling Model (SDSM) projected future rainfall in Northwestern, Western and Southern provinces of Sri Lanka

et al. 2022

View full text Add to dashboard Cite

Even though an extensive amount of climate change studies have been carried out in different parts of the world, Sri Lanka is one of the least focused countries in this regard. Climate projections are important and encouraged to manage the futuristic adverse impacts. Identifying this research gap, future rainfall projections were carried out in three provinces in Sri Lanka, i.e. Northwestern (Puttalam and Kurunegala), Western (Katunayake and Colombo), and Southern province (Galle and Hambantota). The Canadian Earth System Model (CanESM2) under the Representation Concentration Pathways (RCP8.5) was downscaled using the Statistical DownScaling Model (SDSM). Non-parametric tests, including Mann-Kendall (MK) test and the Sen's Slope estimator, were used to determine the significance of trends and magnitude of the slope of the historical trends (1990-2019) and future projected trends (2020-2100). The trends were analyzed for four major seasons in Sri Lanka, including First Inter-monsoon (FIM), Southwest monsoon (SWM), Second Intermonsoon (SIM), and the Northeast monsoon (NEM). The standard error and model bias at rainfall stations were 0.014-0.034 mm and 1-1.1 respectively, which are acceptable when compared to previous studies. Several significant rainfall trends were identified, including positive trends in the mid-future (2041-2070), and negative trends in the far-future (2071-2100). In addition, rainfall indices, including Rx5day, R20mm, Consecutive dry days (CDD), and Consecutive wet days (CWD) were tested in future projected and historical rainfalls. The results of the present study will be useful for policymakers for decision-making processes in water resources management and agriculture.

show abstract

Future Climate of Colombo Downscaled with SDSM-Neural Network

Cited by 21 publications

References 25 publications

Past and future drought trends, duration, and frequency in the semi‐arid Urmia Lake Basin under a changing climate

Past and future drought trends, duration, and frequency in the semi‐arid Urmia Lake Basin under a changing climate

Downscaling and Projection of Multi-CMIP5 Precipitation Using Machine Learning Methods in the Upper Han River Basin

Analysis of Statistical Downscaling Model (SDSM) projected future rainfall in Northwestern, Western and Southern provinces of Sri Lanka

Contact Info

Product

Resources

About