Premium
Downscaling climate change of mean climatology and extremes of precipitation and temperature: Application to a Mediterranean climate basin
Author(s) -
Zhang Rong,
CorteReal João,
Moreira Madalena,
Kilsby Chris,
Burton Aidan,
Fowler Hayley J.,
Blenkinsop Stephen,
Birkinshaw Stephen,
Forsythe Nathan,
Nunes João P.,
Sampaio Elsa
Publication year - 2019
Publication title -
international journal of climatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.58
H-Index - 166
eISSN - 1097-0088
pISSN - 0899-8418
DOI - 10.1002/joc.6122
Subject(s) - downscaling , climatology , environmental science , precipitation , climate change , context (archaeology) , mediterranean climate , cru , meteorology , geology , geography , paleontology , oceanography , archaeology
Downscaling is usually necessary for robust hydrological impact assessments. This may be undertaken using a wide range of methods, including a combination of dynamical and statistical‐stochastic downscaling. This study uses the Spatial–Temporal Neyman‐Scott Rectangular Pulses model—RainSimV3, the precipitation‐conditioned daily weather generator—ICAAM‐WG, and the change factor approach for downscaling synthetic climate scenarios for robust hydrological impact assessment at middle‐sized basins. The ICAAM‐WG was developed based on the concept of the Climate Research Unit daily weather generator (CRU‐WG), motivated by the need for improved representation of heat waves by downscaling methods given the positive feedback between low soil moisture and high air temperature. We demonstrated the validity of the proposed methodology in the 705‐km 2 Mediterranean climate basin in southern Portugal. The results show that, for the control period 1980–2010, both RainSimV3 and ICAAM‐WG reproduced not only the mean climatology, but also extreme wet and low precipitation events, as well as the extremes of temperature and heat waves. We found that downscaling with ICAAM‐WG (SIM6), which uses second‐order autoregressive processes for the simulation of temperature during consecutive dry and wet days, outperformed ICAAM‐WG (SIM4), which used only first‐order autoregressive processes, leading to improved simulation of heat waves. ICAAM‐WG (SIM6) well reproduced observed heatwave extremes with return periods of up to 30 years; however, ICAAM‐WG (SIM4) overestimated these extremes substantially. This indicates the importance of incorporating second‐order autoregressive processes in the simulation of heatwave length. In the context of climate warming, the proposed methodology provides a tool to improve downscaled projections of future extremes with confidence intervals for not only wet events but also dry spells and heat waves.