Uncertainties of gridded precipitation observations in characterizing spatio‐temporal drought and wetness over Vietnam

Abstract

Precipitation is an important climate variable to investigate extreme events (e.g. drought and flood) as well as to develop robust strategies for water resources planning and management. Lack of adequate and robust information on precipitation poses great difficulties in understanding the observed climate as well as to validate climate model outputs. To overcome this limitation gridded precipitation data sets have been constructed to supplement the lack of in situ data. This study compares five popular gridded precipitation data sets to evaluate their performance in terms of drought and wetness over Vietnam. These five gridded data sets include: (1) Asian Precipitation Highly Resolved Observational Data Integration Towards the Evaluation of Water Resources (APHRODITE), (2) Climate Precipitation Center (CPC), (3) Climate Research Unit (CRU), (4) Global Precipitation Climatology Center (GPCC) and (5) University of Delaware (UDEL). The recently developed gridded precipitation observational data ‘VnGP’ from Vietnam is used as the reference data set to assess the performance of these five gridded precipitation products. The Standardized Precipitation Index (SPI) is used to quantify drought and wetness. GPCC and APHRODITE performed reasonably well in reproducing spatial and temporal precipitation patterns. GPCC performs consistently better than APHRODITE in all the statistical tests. Except for UDEL, other gridded data sets able to exhibit the characteristics of drought/wetness (e.g. the percentage of drought events and severity) during strong El Nino Southern Oscillation (ENSO) events. However, higher uncertainty exists to quantify drought inter‐arrival time in most of the data sets. Furthermore, trend analysis was performed to evaluate the comparative performance of gridded data sets to quantify drought (wet) spells at annual and seasonal time scales. Although the gauge‐based and hybrid satellite–gauge merged products use partly ground truth data, the different interpolation techniques and merging algorithms may contribute to large uncertainties.

The impact of the resolution of meteorological data sets on catchment‐scale precipitation and drought studies

“…However, station density is strongly varying in space and time with Europe having one of the highest station densities. The GPCC has been used to characterize droughts both on a global (Spinoni et al, ) and a regional scale (Raziei et al, ) and was recently found to be more precise than comparable products (Vu et al, ). For this study we selected the daily product for consistency with the other data sets. Continental: precipitation data set of the European Climate Assessment & Dataset (Haylock et al, ), hereafter called E‐OBS.…”

“…Researchers commonly use the data source that is most easily available while still appropriate regarding resolution and coverage. However, values derived from different sources can vary considerably (Yatagai et al, ; Rauthe et al, ; Frick et al, ; Vu et al, ). Comparative assessments of gridded meteorological data have found differences that are relevant for catchment‐scale hydrological analysis and modelling.…”

“…Vu et al . () compared several low‐resolution data sets for Vietnam and found that station density is the most important factor for a good representation of precipitation patterns. Various studies have suggested that particularly extreme values—for drought, in terms of low precipitation and dry periods—might be influenced by the interpolation scheme and resolution of data and therefore precipitation statistics of station data and gridded products will differ (e.g., Ensor and Robeson, ; Gervais et al, ; Wibig et al, ).…”

Trends and Extremes of Drought Episodes in Vietnam Sub-Regions during 1980–2017 at Different Timescales

“…In terms of drought impacts, these have been quantified for well-known dry periods, such as the extremely strong drought event in 1982-1983, which had severe impacts on the environmental and socio-economic sectors of Vietnam [49]. Another example is the 1997-1998 drought, when the total losses in terms of agricultural production were about 400 million US dollars [44,50]. Furthermore, the drought episode that occurred in 2003 caused the main impacts on the coffee sector of Vietnam, thereby emphasizing the role of the detection of drought as an important issue [29].…”

Uncertainty Analysis of Standardized Precipitation Index Due to the Effects of Probability Distributions and Parameter Errors

“…As to the input data, a large number of studies have concerned about the uncertainties associated with input data sources, data length, and sampling uncertainty in the calculation of drought index (e.g., Hao et al, 2014Hao et al, , 2016Liu et al, 2014;Hu et al, 2015;Katiraie-Boroujerdy et al, 2016;Vergni et al, 2017;Zambrano et al, 2017;Beyaztas et al, 2018;Carbone et al, 2018). Vu et al (2018) compared the performance of rain-gauge data and gridded precipitation data in the calculation of SPI over Vietnam. Carbone et al (2018) demonstrated that the SPI estimates derived from 30-year record have considerably more uncertainty than those from the 60-year record.…”