Space-Time Interpolation and Uncertainty Assessment of an Extreme Precipitation Index Using Geostatistical Cosimulation

Author(s):  
Ana Cristina Costa ◽  
Amilcar Soares
2019 ◽  
Vol 34 (5) ◽  
pp. 1257-1276 ◽  
Author(s):  
Shawn M. Milrad ◽  
Eyad H. Atallah ◽  
John R. Gyakum ◽  
Rachael N. Isphording ◽  
Jonathon Klepatzki

Abstract The extreme precipitation index (EPI) is a coupled dynamic–thermodynamic metric that can diagnose extreme precipitation events associated with flow reversal in the mid- to upper troposphere (e.g., Rex and omega blocks, cutoff cyclones, Rossby wave breaks). Recent billion dollar (U.S. dollars) floods across the Northern Hemisphere midlatitudes were associated with flow reversal, as long-duration ascent (dynamics) occurred in the presence of anomalously warm and moist air (thermodynamics). The EPI can detect this potent combination of ingredients and offers advantages over model precipitation forecasts because it relies on mass fields instead of parameterizations. The EPI’s dynamics component incorporates modified versions of two accepted blocking criteria, designed to detect flow reversal during the relatively short duration of extreme precipitation events. The thermodynamic component utilizes standardized anomalies of equivalent potential temperature. Proof-of-concept is demonstrated using four high-impact floods: the 2013 Alberta Flood, Canada’s second costliest natural disaster on record; the 2016 western Europe Flood, which caused the worst flooding in France in a century; the 2000 southern Alpine event responsible for major flooding in Switzerland; and the catastrophic August 2016 Louisiana Flood. EPI frequency maxima are located across the North Atlantic and North Pacific mid- and high latitudes, including near the climatological subtropical jet stream, while secondary maxima are located near the Rockies and Alps. EPI accuracy is briefly assessed using pattern correlation and qualitative associations with an extreme precipitation event climatology. Results show that the EPI may provide potential benefits to flood forecasters, particularly in the 3–10-day range.


2008 ◽  
Vol 8 (4) ◽  
pp. 763-773 ◽  
Author(s):  
A. C. Costa ◽  
R. Durão ◽  
M. J. Pereira ◽  
A. Soares

Abstract. The topographic characteristics and spatial climatic diversity are significant in the South of continental Portugal where the rainfall regime is typically Mediterranean. Direct sequential cosimulation is proposed for mapping an extreme precipitation index in southern Portugal using elevation as auxiliary information. The analysed index (R5D) can be considered a flood indicator because it provides a measure of medium-term precipitation total. The methodology accounts for local data variability and incorporates space-time models that allow capturing long-term trends of extreme precipitation, and local changes in the relationship between elevation and extreme precipitation through time. Annual gridded datasets of the flood indicator are produced from 1940 to 1999 on 800 m×800 m grids by using the space-time relationship between elevation and the index. Uncertainty evaluations of the proposed scenarios are also produced for each year. The results indicate that the relationship between elevation and extreme precipitation varies locally and has decreased through time over the study region. In wetter years the flood indicator exhibits the highest values in mountainous regions of the South, while in drier years the spatial pattern of extreme precipitation has much less variability over the study region. The uncertainty of extreme precipitation estimates also varies in time and space, and in earlier decades is strongly dependent on the density of the monitoring stations network. The produced maps will be useful in regional and local studies related to climate change, desertification, land and water resources management, hydrological modelling, and flood mitigation planning.


2014 ◽  
Vol 35 (8) ◽  
pp. 1749-1760 ◽  
Author(s):  
Francesco Cioffi ◽  
Upmanu Lall ◽  
Ester Rus ◽  
Chandra Kiran B. Krishnamurthy

Author(s):  
Marina A. Volkova ◽  
◽  
Natalia N. Cheredko ◽  
Kirill I. Sokolov ◽  
Leontiy A. Ogurtsov ◽  
...  

2009 ◽  
Vol 9 (1) ◽  
pp. 241-250 ◽  
Author(s):  
R. Durão ◽  
M. J. Pereira ◽  
A. C. Costa ◽  
J. M. Côrte-Real ◽  
A. Soares

Abstract. Most of the actual studies and previews of future rainfall patterns, based on past observed records for Mediterranean climate areas, focus on the decline of the rainfall amounts over the years, and also on the increase of the frequency of heavy/intense rainfall events particularly in the winter season. These changes in heavy rainfall events may have severe implications and impacts on soil erosion resulting in increased soil degradation risks. The objective of the present work is to evaluate the spatial distribution of extreme precipitation events in Southern Portugal, using a geostatistical approach to assess the relationships between spatial and temporal extreme rainfall patterns. The used dataset comprises a set of 105 stations' records of daily precipitation within the period 1960–1999. Two indices of extreme precipitation were selected to be computed based on the daily precipitation observation series: one representing the frequency of extremely heavy precipitation events (R30) and another one characterizing flood events (R5D). The space-time patterns of the precipitation indices were evaluated and simulated using a geostatistical approach. Despite no significant temporal trends were detected on the calculated indices series, the space-time decadal patterns are becoming more continuous in the last two decades than the previous ones.


2002 ◽  
Author(s):  
J. B. Kennedy
Keyword(s):  

Sign in / Sign up

Export Citation Format

Share Document