A non-local model output statistics approach for the downscaling of CMIP5 GCMs for the projection of rainfall in Peninsular Malaysia

Noor, Muhammad Aslam; Ismail, Tarmizi; Ullah, Shahid; Iqbal, Zafar; Nawaz, Nadeem; Ahmed, Kamal

doi:10.2166/wcc.2019.041

Cited by 14 publications

(11 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ConvMOS is also showing rather stable performance as can be seen on the standard deviations in Table 1 despite its non-deterministic fitting procedure. We also ran this experiment with precipitation as the only climate predictor as some prior work has done [2,3,4,6] but found all methods to perform worse without additional predictors.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning for Climate Model Output Statistics

Steininger¹,

Ábel²,

Ziegler³

et al. 2021

Preprint

View full text Add to dashboard Cite

<p>Climate models are an important tool for the assessment of prospective climate change effects but they suffer from systematic and representation errors, especially for precipitation. Model output statistics (MOS) reduce these errors by fitting the model output to observational data with machine learning. In this work, we explore the feasibility and potential of deep learning with convolutional neural networks (CNNs) for MOS. We propose the CNN architecture ConvMOS specifically designed for reducing errors in climate model outputs and apply it to the climate model REMO. Our results show a considerable reduction of errors and mostly improved performance compared to three commonly used MOS approaches.</p>

show abstract

Section: Methodsmentioning

confidence: 99%

“…Like other previous MOS approaches [1,3,4] we preprocess our predictors for the standard MOS methods to reduce dimensionality and remove potentially unhelpful information. Like [3] and [4] we use supervised PCA [22]. First, we select the best predictors based on a univariate regression.…”

Section: Supervised Principal Component Analysismentioning

confidence: 99%

Deep Learning for Climate Model Output Statistics

Steininger¹,

Ábel²,

Ziegler³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The methodology opted in the paper was applied to Peninsular Malaysia, which is situated in South East Asia with a latitude 1.20° to 6.40° N and longitude 99.35° to 104.20° E (Noor et al, 2019a). Situated near the equator, the climate of Malaysia is humid and hot.…”

Section: Geography Of Peninsular Malaysiamentioning

confidence: 99%

“…In-situ observations are the most reliable precipitation data; however, they are insufficient to provide details of spatial rainfall distribution in most regions due to the sparse distribution of rain gauges (Dewan et al, 2019, Nashwan et al, 2019a. Satellite-driven products are emerging as a dependable source of high spatial and temporal resolution rainfall measurement globally (Bhatti et al, 2016, Noor et al, 2019a. Satellite-based precipitation products (SBP) have shown their potential in different hydro-climatic studies such as floods modelling, drought monitoring, water budgeting, and hydrological change assessment (Xie et al, 2011, Bitew et al, 2011, Yong et al, 2010, Nashwan et al, 2020.…”

Section: Introductionmentioning

confidence: 99%

Bias correction method of high-resolution satellite-based precipitation product for Peninsular Malaysia

Iqbal

Shahid

Ahmed

et al. 2022

Theor Appl Climatol

Self Cite

View full text Add to dashboard Cite

Satellite-based precipitation (SBP) is emerging as a reliable source for high-resolution rainfall estimates over the globe. However, uncertainty in SBP is still significant, limiting their use without evaluation and often without bias correction. The bias correction of SBP remained a challenge for atmospheric scientists. In this study, the performance of six SBPs, namely, SM2RAIN-ASCAT, IMERG, GsMap, CHIRPS, PERSIANN-CDS and PERSIANN-CSS in replicating observed daily rainfall at 364 stations over Peninsular Malaysia was evaluated. The bias of the most suitable SBP was corrected using a novel machine learning (ML)-based bias-correction method. The proposed bias-correction method consists of an ML classifier to correct the bias in estimating rainfall occurrence and an ML regression model to correct the amount of rainfall during rainfall events. The performance of different widely used ML algorithms for classification and regression were evaluated to select the suitable algorithms. IMERG showed better performance, showing a higher correlation coefficient (R 2 ) of 0.57 and Kling-Gupta Efficiency (KGE) of 0.5 compared to the other products. The performance of random forest (RF) was better than the knearest neighbourhood (KNN) for both classification and regression. RF classified the rainfall events with a skill score of 0.38 and estimated the rainfall amount of a rainfall event with the modified Index of Agreement (md) of 0.56. Comparison of IMERG and bias-corrected IMERG (BIMERG) revealed an average reduction in RMSE by 55% in simulating observed rainfall. The proposed bias correction method performed much better when compared with the conventional bias correction methods such as linear scaling and quantile regression. The BIMERG could reliably replicate the spatial distribution of heavy rainfall events, indicating its potential for hydro-climatic studies like flood and drought monitoring in the study area.

show abstract

“…Different GOF tests often provide contradictory results in ranking best fit distribution. Besides, various methods are used for the estimation of fitted PDF parameters such as L-moments, maximum likelihood estimator, generalized maximum likelihood, Bayesian methods, probability weighted moments, least square, and many others [23][24][25]. Estimated PDF parameter values vary significantly when different methods are used for the estimation of PDF parameters [26].…”

Section: Introductionmentioning

confidence: 99%

Assessing the Uncertainty Associated with Flood Features due to Variability of Rainfall and Hydrological Parameters

Sharafati

Khazaei

Nashwan

et al. 2020

Advances in Civil Engineering

View full text Add to dashboard Cite

An assessment of uncertainty in flood hydrograph features, e.g., peak discharge and flood volume due to variability in the rainfall-runoff model (HEC-HMS) parameters and rainfall characteristics, e.g., depth and duration, is conducted. Flood hydrographs are generated using a rain pattern generator (RPG) and HEC-HMS models through Monte Carlo simulation considering uncertainty in stochastic variables. The uncertainties in HEC-HMS parameters (e.g., loss, base flow, and unit hydrograph) are estimated using their probability distribution functions. The flood events are obtained by simulating runoff for rainfall events using the generated model parameters. The uncertainties due to rainfall and model parameters on generated flood hydrographs are evaluated using the relative coefficient of variation (RCV). The results reveal a higher RCV index for flood volume (RCV = 153) than peak discharge (RCV = 116) for a 12-hr rainfall duration. The average relative RCV (ARRCV) index computed for hydrological component (e.g., base flow, loss, or unit hydrograph) indicates the highest impact of rainfall depth on flood volume and peak. The results indicate that rainfall depth is the main source of uncertainty of flood peak and volume.

show abstract

A non-local model output statistics approach for the downscaling of CMIP5 GCMs for the projection of rainfall in Peninsular Malaysia

Cited by 14 publications

References 16 publications

Deep Learning for Climate Model Output Statistics

Deep Learning for Climate Model Output Statistics

Bias correction method of high-resolution satellite-based precipitation product for Peninsular Malaysia

Assessing the Uncertainty Associated with Flood Features due to Variability of Rainfall and Hydrological Parameters

Contact Info

Product

Resources

About