Search

Reanalysis datasets provide a continuous picture of the past climate for every point on Earth. They are especially useful in areas with few direct observations, such as Siberia. However, to ensure these datasets are sufficiently accurate they need to be validated against readings from meteorological stations. Here, we analyse how values of six climate variables—the minimum, mean and maximum 2‐metre air temperature, snow depth (SD), total precipitation and wind speed (WSP)—from three reanalysis datasets—ERA‐Interim, ERA5 and ERA5‐Land—compare against observations from 29 meteorological stations across Siberia and the Russian Far East on a daily timescale from 1979 to 2019. All three reanalyses produce values of the mean and maximum daily 2‐metre air temperature that are close to those observed, with the average absolute bias not exceeding 1.54°C. However, care should be taken for the minimum 2‐metre air temperature during the summer months—there are nine stations where correlation values are <0.60 due to inadequate night‐time cooling. The reanalysis values of SD are generally close to those observed after 1992, especially ERA5, when data from some of the meteorological stations began to be assimilated, but the reanalysis SD should be used with caution (if at all) before 1992 as the lack of assimilation leads to large overestimations. For low daily precipitation values the reanalyses provide good approximations, however they struggle to attain the extreme high values. Similarly, for the 10‐metre WSP; the reanalyses perform well with speeds up to 2.5 ms−1 but struggle with those above 5.0 ms−1. For these variables, we recommend using ERA5 over ERA‐Interim and ERA5‐Land in future research. ERA5 shows minor improvements over ERA‐Interim, and, despite an increased spatial resolution, there is no advantage to using ERA5‐Land. [ABSTRACT FROM AUTHOR]

This study evaluates the reliability of ERA5 and ERA5‐Land reanalysis datasets in describing the mean daily air temperature of four climate domains in mainland Portugal. The reanalysis datasets were compared with ground observations from 94 meteorological stations (1980–2021). Overall, the results demonstrated a good degree of correlation between the observed and reanalysis data on both a daily and seasonal scale. Both the latitudinal distribution of the air temperature and the moderating effect of the Atlantic Ocean are well described. However, in the case of Portugal, the ERA5‐Land was shown to be considerably more effective at describing the mean daily air temperature than ERA5. The results also indicated that, in general, the reanalysis methodologies perform better when applied to air temperature simulation in flatter regions as opposed to regions with high‐altitude and complex terrain. The study further suggests that ERA5 and ERA5‐Land reanalysis should be used with caution in the case of short‐term environmental studies. In fact, relevant differences were shown to exist between the reanalyses and the observed daily mean air temperature datasets for certain specific years. Overall, considering the RMSE between the ERA5‐Land reanalysis datasets for mean daily air temperature and the observed datasets there is a 28% probability of locally having a mean RMSE <1.5°C, 52% probability of having a mean RMSE >1.5°C and <2.0°C, and 16% probability of having a RMSE >2.0°C and <3.0°C. These conclusions will hopefully contribute to improving our understanding of the uncertainty sources in relation to ERA5 and ERA5‐Land reanalysis data for different climate domains. [ABSTRACT FROM AUTHOR]

Long-term observation series of hydrological and climatic parameters should be obtained to solve the problems of assessing and forecasting the hydrological regime of river basins. In the absence of surface measurements of these parameters, reanalysis data can be used. The paper presents a comparison of the runoff magnitude and the consistency of runoff dynamics based on the observational data of gauging stations and the ERA5-Land and TerraClimate reanalysis data for seven rivers of Western Siberia (the left-bank tributaries of the Yenisei located in the zones of taiga and forest-tundra). Spearman's rank correlation coefficients for seasonal and annual runoff between the data of reanalysis and the actual data are equal to 0.35–0.98. The normalized root-mean-square error of annual runoff according to the reanalysis ranges from 13 to 21%. The values of errors and correlation coefficients depend on geographic location and area of the basins and vary during a season. [ABSTRACT FROM AUTHOR]

In mountainous regions, the scarcity of air temperature (Ta) measurements is a major limitation for hydrological and crop monitoring. An alternative to in situ measurements could be to downscale the reanalysis Ta data provided at high-temporal resolution. However, the relatively coarse spatial resolution of these products (i.e., 9 km for ERA5-Land) is unlikely to be directly representative of actual local Ta patterns. To address this issue, this study presents a new spatial downscaling strategy of hourly ERA5-Land Ta data with a three-step procedure. First, the 9 km resolution ERA5 Ta is corrected at its original resolution by using a reference Ta derived from the elevation of the 9 km resolution grid and an in situ estimate over the area of the hourly Environmental Lapse Rate (ELR). Such a correction of 9 km resolution ERA5 Ta is trained using several machine learning techniques, including Multiple Linear Regression (MLR), Support Vector Regression (SVR), and Extreme Gradient Boosting (Xgboost), as well as ancillary ERA5 data (daily mean, standard deviation, hourly ELR, and grid elevation). Next, the trained correction algorithms are run to correct 9 km resolution ERA5 Ta, and the corrected ERA5 Ta data are used to derive an updated ELR over the area (without using in situ Ta measurements). Third, the updated hourly ELR is used to disaggregate 9 km resolution corrected ERA5 Ta data at the 30-meter resolution of SRTM's Digital Elevation Model (DEM). The effectiveness of this method is assessed across the northern part of the High Atlas Mountains in central Morocco through (1) k-fold cross-validation against five years (2016 to 2020) of in situ hourly temperature readings and (2) comparison with classical downscaling methods based on a constant ELR. Our results indicate a significant enhancement in the spatial distribution of hourly local Ta. By comparing our model, which included Xgboost, SVR, and MLR, with the constant ELR-based downscaling approach, we were able to decrease the regional root mean square error from approximately 3 ∘ C to 1.61 ∘ C, 1.75 ∘ C, and 1.8 ∘ C, reduce the mean bias error from −0.5 ∘ C to null, and increase the coefficient of determination from 0.88 to 0.97, 0.96, and 0.96 for Xgboost, SVR, and MLR, respectively. [ABSTRACT FROM AUTHOR]

An increase in extreme temperature events could have a significant impact on terrestrial ecosystems. Reanalysis temperature data are an important data set for extreme temperature estimation in mountainous areas with few meteorological stations. The ability of ERA5-Land reanalysis data to capture the extreme temperature index published by the Expert Team on Climate Change Detection and Indices (ETCCDI) was evaluated by using the observational data from 17 meteorological stations in the Qilian Mountains (QLM) during 1979–2017. The results show that the ERA5-Land reanalysis temperature data can capture well for the daily maximum temperature, two warm extremes (TXx and TX90p) and one cold extreme (FD0) in the QLM. ERA5-Land’s ability to capture temperature extremes is best in summer and worst in spring and winter. In addition, ERA5-Land can capture trends in all extreme temperature indices except the daily temperature range (DTR). The main bias of ERA5-Land is due to the difference in elevation between the ground observation station and the ERA5-Land grid point. The simulation accuracy of ERA5-Land increases with the decrease of elevation difference. The results can provide a reference for the study of local extreme temperature by using reanalysis data.

Reanalysis temperature products are important datasets for temperature estimates over high-elevation areas with few meteorological stations. In this study, surface 2 m air temperature data from 17 meteorological stations from 1979 to 2017 in the Qilian Mountains (QLM) are used for comparison with the newest reanalysis product: ERA5-Land derived from the European Centre for Medium-Range Weather Forecasts (ECMWF). In general, the ERA5-Land temperature product can reproduce the observation variation at different time scales very well. A high monthly correlation coefficient that ranges from 0.978 to 0.998 suggests that ERA5-Land reanalysis temperature could capture the observations very well. However, attention should be paid before using ERA5-Land at individual sites because of the average root-mean-square-error (RMSE) of 2.2°C of all stations. The biases between ERA5-Land temperature and observations are mainly caused by the elevation differences between ERA5-Land grid points and meteorological sites. The annual mean temperature shows a significant warming trend (0.488°C/decade) from 1979 to 2017 based on the observations. ERA5-Land reanalysis temperature captures the increasing trend very well (0.379°C/decade). The biggest positive warming trends of observations and ERA5-Land are both found in summer with values of 0.574°C/decade and 0.496°C/decade, respectively. We suggest that ERA5-Land generally reproduces the temperature trend very well for observations and is reliable for scientific research over the QLM.

In mountainous regions, the scarcity of air temperature (Ta) measurements is a major limitation for hydrological and crop monitoring. An alternative to in situ measurements could be to downscale the reanalysis Ta data provided at high-temporal resolution. However, the relatively coarse spatial resolution of these products (i.e., 9 km for ERA5-Land) is unlikely to be directly representative of actual local Ta patterns. To address this issue, this study presents a new spatial downscaling strategy of hourly ERA5-Land Ta data with a three-step procedure. First, the 9 km resolution ERA5 Ta is corrected at its original resolution by using a reference Ta derived from the elevation of the 9 km resolution grid and an in situ estimate over the area of the hourly Environmental Lapse Rate (ELR). Such a correction of 9 km resolution ERA5 Ta is trained using several machine learning techniques, including Multiple Linear Regression (MLR), Support Vector Regression (SVR), and Extreme Gradient Boosting (Xgboost), as well as ancillary ERA5 data (daily mean, standard deviation, hourly ELR, and grid elevation). Next, the trained correction algorithms are run to correct 9 km resolution ERA5 Ta, and the corrected ERA5 Ta data are used to derive an updated ELR over the area (without using in situ Ta measurements). Third, the updated hourly ELR is used to disaggregate 9 km resolution corrected ERA5 Ta data at the 30-meter resolution of SRTM’s Digital Elevation Model (DEM). The effectiveness of this method is assessed across the northern part of the High Atlas Mountains in central Morocco through (1) k-fold cross-validation against five years (2016 to 2020) of in situ hourly temperature readings and (2) comparison with classical downscaling methods based on a constant ELR. Our results indicate a significant enhancement in the spatial distribution of hourly local Ta. By comparing our model, which included Xgboost, SVR, and MLR, with the constant ELR-based downscaling approach, we were able to decrease the regional root mean square error from approximately 3 ∘C to 1.61 ∘C, 1.75 ∘C, and 1.8 ∘C, reduce the mean bias error from −0.5 ∘C to null, and increase the coefficient of determination from 0.88 to 0.97, 0.96, and 0.96 for Xgboost, SVR, and MLR, respectively.