TY - JOUR
T1 - A multi-decadal assessment of the performance of gauge- and model-based rainfall products over Saudi Arabia: Climatology, anomalies and trends
AU - El Kenawy, Ahmed M.
AU - McCabe, Matthew
N1 - KAUST Repository Item: Exported on 2020-10-01
Acknowledgements: Research reported in this publication was supported by the King Abdullah University of Science and Technology (KAUST), Saudi Arabia. The authors are grateful to the Saudi Arabian Ministry of Water and Electricity (MOWE) for providing the daily rainfall data. We also acknowledge the many different universities and government-based research groups for providing their data sets for analysis.
PY - 2015/5/15
Y1 - 2015/5/15
N2 - Many arid and semi-arid regions have sparse precipitation observing networks, which limits the capacity for detailed hydrological modelling, water resources management and flood forecasting efforts. The objective of this work is to evaluate the utility of relatively high-spatial resolution rainfall products to reproduce observed multi-decadal rainfall characteristics such as climatologies, anomalies and trends over Saudi Arabia. Our study compares the statistical characteristics of rainfall from 53 observatories over the reference period 1965-2005, with rainfall data from six widely used gauge-based products, including APHRODITE, GPCC, PRINCETON, UDEL, CRU and PREC/L. In addition, the performance of three global climate models (GCMs), including CCSM4, EC-EARTH and MRI-I-CGCM3, integrated as part of the Fifth Coupled Model Intercomparison Project (CMIP5), was also evaluated. Results indicate that the gauge-based products were generally skillful in reproducing rainfall characteristics in Saudi Arabia. In most cases, the gauge-based products were also able to capture the annual cycle, anomalies and climatologies of observed data, although significant inter-product variability was observed, depending on the assessment metric being used. In comparison, the GCM-based products generally exhibited poor performance, with larger biases and very weak correlations, particularly during the summertime. Importantly, all products generally failed to reproduce the observed long-term seasonal and annual trends in the region, particularly during the dry seasons (summer and autumn). Overall, this work suggests that selected gauge-based products with daily (APHRODITE and PRINCETON) and monthly (GPCC and CRU) resolutions show superior performance relative to other products, implying that they may be the most appropriate data source from which multi-decadal variations of rainfall can be investigated at the regional scale over Saudi Arabia. Discriminating these skillful products is important not only for reducing uncertainty in climate, hydrological and environmental assessments but also for advancing model developments in the region.
AB - Many arid and semi-arid regions have sparse precipitation observing networks, which limits the capacity for detailed hydrological modelling, water resources management and flood forecasting efforts. The objective of this work is to evaluate the utility of relatively high-spatial resolution rainfall products to reproduce observed multi-decadal rainfall characteristics such as climatologies, anomalies and trends over Saudi Arabia. Our study compares the statistical characteristics of rainfall from 53 observatories over the reference period 1965-2005, with rainfall data from six widely used gauge-based products, including APHRODITE, GPCC, PRINCETON, UDEL, CRU and PREC/L. In addition, the performance of three global climate models (GCMs), including CCSM4, EC-EARTH and MRI-I-CGCM3, integrated as part of the Fifth Coupled Model Intercomparison Project (CMIP5), was also evaluated. Results indicate that the gauge-based products were generally skillful in reproducing rainfall characteristics in Saudi Arabia. In most cases, the gauge-based products were also able to capture the annual cycle, anomalies and climatologies of observed data, although significant inter-product variability was observed, depending on the assessment metric being used. In comparison, the GCM-based products generally exhibited poor performance, with larger biases and very weak correlations, particularly during the summertime. Importantly, all products generally failed to reproduce the observed long-term seasonal and annual trends in the region, particularly during the dry seasons (summer and autumn). Overall, this work suggests that selected gauge-based products with daily (APHRODITE and PRINCETON) and monthly (GPCC and CRU) resolutions show superior performance relative to other products, implying that they may be the most appropriate data source from which multi-decadal variations of rainfall can be investigated at the regional scale over Saudi Arabia. Discriminating these skillful products is important not only for reducing uncertainty in climate, hydrological and environmental assessments but also for advancing model developments in the region.
UR - http://hdl.handle.net/10754/565942
UR - http://onlinelibrary.wiley.com/doi/10.1002/joc.4374/abstract
UR - http://www.scopus.com/inward/record.url?scp=84956845078&partnerID=8YFLogxK
U2 - 10.1002/joc.4374
DO - 10.1002/joc.4374
M3 - Article
SN - 0899-8418
VL - 36
SP - 656
EP - 674
JO - International Journal of Climatology
JF - International Journal of Climatology
IS - 2
ER -