Browse
Abstract
High-resolution numerical weather prediction (NWP) systems present a strong potential to provide meteorological information in alpine terrain for diverse applications. However, they still suffer from biases highly detrimental for practical purposes. In this study, we investigate the origin of a significant wintertime screen-level temperature bias in forecasts of the AROME-France NWP system in high-altitude, snow-covered alpine terrain. For this purpose, a thorough set of meteorological and snow observations from two high-altitude instrumental sites is used. Targeted numerical simulations are carried out to disentangle the contributions to this bias coming from atmospheric fields, from the snow scheme, and from the coupling between the snowpack and the atmosphere. At both sites, the wind speed and incoming longwave radiation appear significantly negatively biased in AROME in the winter season. Using targeted offline simulations, we show that the simulation errors in these screen-level fields contribute to an average of 67% of the screen-level temperature bias of AROME, while the contribution of errors in the incoming shortwave radiation is negligible. Additionally, the screen-level temperature of AROME is not majorly impacted by changes in the complexity and especially the vertical layering of the snow model. However, it appears particularly sensitive to the parameterization of turbulent fluxes in stable conditions. Evidence suggest that these findings could at least partially be generalized to the whole AROME-France alpine domain. Hence, reducing the high-altitude, winter screen-level temperature bias in AROME may in great part proceed from improving the simulation of atmospheric fields and eliminating some bias compensations in the model.
Abstract
High-resolution numerical weather prediction (NWP) systems present a strong potential to provide meteorological information in alpine terrain for diverse applications. However, they still suffer from biases highly detrimental for practical purposes. In this study, we investigate the origin of a significant wintertime screen-level temperature bias in forecasts of the AROME-France NWP system in high-altitude, snow-covered alpine terrain. For this purpose, a thorough set of meteorological and snow observations from two high-altitude instrumental sites is used. Targeted numerical simulations are carried out to disentangle the contributions to this bias coming from atmospheric fields, from the snow scheme, and from the coupling between the snowpack and the atmosphere. At both sites, the wind speed and incoming longwave radiation appear significantly negatively biased in AROME in the winter season. Using targeted offline simulations, we show that the simulation errors in these screen-level fields contribute to an average of 67% of the screen-level temperature bias of AROME, while the contribution of errors in the incoming shortwave radiation is negligible. Additionally, the screen-level temperature of AROME is not majorly impacted by changes in the complexity and especially the vertical layering of the snow model. However, it appears particularly sensitive to the parameterization of turbulent fluxes in stable conditions. Evidence suggest that these findings could at least partially be generalized to the whole AROME-France alpine domain. Hence, reducing the high-altitude, winter screen-level temperature bias in AROME may in great part proceed from improving the simulation of atmospheric fields and eliminating some bias compensations in the model.
Abstract
An ideal spatial interpolation approach is indispensable for obtaining high-quality gridded climatic data in mountainous regions with scarce observations, particularly for the Hengduan Mountains Region (HMR) with dense longitudinal ranges and gorges. However, there is much controversy about the applicability of thin plate smooth spline (TPSS), cokriging, and inverse distance weighting (IDW) in mountainous regions. Here, we use the daily observations of temperature and precipitation at 125 stations in HMR and its surroundings from 1961 to 2018 and adopt three interpolation methods to map the annual average temperature and precipitation at a resolution of 500 m in HMR. Then, we assess the applicability of three interpolation methods in HMR from the perspectives of interpolation accuracy and effects. The evaluation implies a satisfactory interpolation accuracy of TPSS with the highest correlation and lowest error, whether for temperature (R 2 = 0.92, RMSE = 1.2°C) or precipitation (R 2 = 0.54, RMSE = 165.9 mm). In addition, the TPSS could better display the temperature (precipitation) gradient along elevation and depict dry valleys’ high-temperature and low-precipitation characteristics. Moreover, the satisfactory interpolation performance of TPSS mainly benefits from the screening of optimal TPSS model that varied primarily with the regional topography feature and meteorological observation density. The uncertainty of gridded climate datasets has become an urgent problem to solve in the complex terrain. This research illustrates the satisfactory applicability of TPSS for climatic spatial interpolation in HMR, providing theoretical support for high-precision interpolation in complex terrain, hopefully improving the regional weather forecasts and disaster warnings.
Abstract
An ideal spatial interpolation approach is indispensable for obtaining high-quality gridded climatic data in mountainous regions with scarce observations, particularly for the Hengduan Mountains Region (HMR) with dense longitudinal ranges and gorges. However, there is much controversy about the applicability of thin plate smooth spline (TPSS), cokriging, and inverse distance weighting (IDW) in mountainous regions. Here, we use the daily observations of temperature and precipitation at 125 stations in HMR and its surroundings from 1961 to 2018 and adopt three interpolation methods to map the annual average temperature and precipitation at a resolution of 500 m in HMR. Then, we assess the applicability of three interpolation methods in HMR from the perspectives of interpolation accuracy and effects. The evaluation implies a satisfactory interpolation accuracy of TPSS with the highest correlation and lowest error, whether for temperature (R 2 = 0.92, RMSE = 1.2°C) or precipitation (R 2 = 0.54, RMSE = 165.9 mm). In addition, the TPSS could better display the temperature (precipitation) gradient along elevation and depict dry valleys’ high-temperature and low-precipitation characteristics. Moreover, the satisfactory interpolation performance of TPSS mainly benefits from the screening of optimal TPSS model that varied primarily with the regional topography feature and meteorological observation density. The uncertainty of gridded climate datasets has become an urgent problem to solve in the complex terrain. This research illustrates the satisfactory applicability of TPSS for climatic spatial interpolation in HMR, providing theoretical support for high-precision interpolation in complex terrain, hopefully improving the regional weather forecasts and disaster warnings.
Abstract
Wind-driven snow transport has important implications for spatial–temporal heterogeneity of snow distribution and snowpack evolution in mountainous areas, such as the French Alps. Due to the paucity of near-surface observations, our knowledge on the spatiotemporal variability of blowing snow occurrences is rather limited. Based on multiyear in situ observations, the spatial–temporal variability in the occurrence of blowing snow events in the French Alps was presented to investigate potential links with ambient meteorological conditions. Statistical analysis of the observations demonstrates that blowing snow events are frequently observed with substantial spatiotemporal variability. Most stations experienced snow transport one out of every five days throughout winter, and the corresponding cumulative hours with blowing snow occurrence accounted for 8% of the month in winter. Blowing snow events generally last 4–8 h in winter and early spring. The likelihood of blowing snow occurrences increases with wind speed but with divergent patterns across snow types. The frequency of blowing snow occurrences with concurrent snowfall is substantially higher than that without concurrent snowfall, although high spatiotemporal variability was found. The considerable variation in snow transport frequency can be explained by contrasting meteorological conditions, local climate, snowpack properties, and topography (elevation and aspect). The temperature-based empirical scheme failed to recognize individual occurrence of blowing snow events because of the significantly overestimated threshold wind speeds, highlighting the importance of validation using in situ observations. Our results contribute to the understanding of spatiotemporal occurrence of blowing snow events and facilitate the development of blowing snow models.
Abstract
Wind-driven snow transport has important implications for spatial–temporal heterogeneity of snow distribution and snowpack evolution in mountainous areas, such as the French Alps. Due to the paucity of near-surface observations, our knowledge on the spatiotemporal variability of blowing snow occurrences is rather limited. Based on multiyear in situ observations, the spatial–temporal variability in the occurrence of blowing snow events in the French Alps was presented to investigate potential links with ambient meteorological conditions. Statistical analysis of the observations demonstrates that blowing snow events are frequently observed with substantial spatiotemporal variability. Most stations experienced snow transport one out of every five days throughout winter, and the corresponding cumulative hours with blowing snow occurrence accounted for 8% of the month in winter. Blowing snow events generally last 4–8 h in winter and early spring. The likelihood of blowing snow occurrences increases with wind speed but with divergent patterns across snow types. The frequency of blowing snow occurrences with concurrent snowfall is substantially higher than that without concurrent snowfall, although high spatiotemporal variability was found. The considerable variation in snow transport frequency can be explained by contrasting meteorological conditions, local climate, snowpack properties, and topography (elevation and aspect). The temperature-based empirical scheme failed to recognize individual occurrence of blowing snow events because of the significantly overestimated threshold wind speeds, highlighting the importance of validation using in situ observations. Our results contribute to the understanding of spatiotemporal occurrence of blowing snow events and facilitate the development of blowing snow models.
Abstract
Errors associated with the location of precipitation in QPFs present challenges when used for hydrologic prediction, particularly in small watersheds. This work builds on a past study that systematically shifted QPFs prior to inputting them into a hydrologic model to generate streamflow ensembles. In the original study, which used static, predetermined shifting distances, flood detection improved, but false alarms increased due to large ensemble spread. The present research tests a more informed approach by randomly selecting shift directions and distances based on the distribution of displacement errors from a sample of QPFs. Precipitation forecasts were taken from the High-Resolution Rapid Refresh Ensemble (HRRRE), and streamflow predictions were generated using the Weather Research and Forecasting hydrological modeling system, version 5.1.1, in a National Water Model 2.0 configuration. A 63-member streamflow ensemble was generated using the 9 original HRRRE and 54 shifted HRRRE members. Two ensemble updating schemes were tested in which ensemble member weights were adjusted using precipitation location and QPF displacement present at convective initiation. The ensembles using QPF shifted based on climatological spatial errors showed higher probabilistic forecasting skill, while having comparable dichotomous forecasting skill to the original HRRRE ensemble. Other methods of selecting nine ensemble members from the full 63-member suite did not show significant improvement. Flood peak timing showed frequent errors, with average timing errors around five hours early. Larger watersheds tended to have better skill metric scores than smaller basins, with increased skill added by the shifting of QPF.
Abstract
Errors associated with the location of precipitation in QPFs present challenges when used for hydrologic prediction, particularly in small watersheds. This work builds on a past study that systematically shifted QPFs prior to inputting them into a hydrologic model to generate streamflow ensembles. In the original study, which used static, predetermined shifting distances, flood detection improved, but false alarms increased due to large ensemble spread. The present research tests a more informed approach by randomly selecting shift directions and distances based on the distribution of displacement errors from a sample of QPFs. Precipitation forecasts were taken from the High-Resolution Rapid Refresh Ensemble (HRRRE), and streamflow predictions were generated using the Weather Research and Forecasting hydrological modeling system, version 5.1.1, in a National Water Model 2.0 configuration. A 63-member streamflow ensemble was generated using the 9 original HRRRE and 54 shifted HRRRE members. Two ensemble updating schemes were tested in which ensemble member weights were adjusted using precipitation location and QPF displacement present at convective initiation. The ensembles using QPF shifted based on climatological spatial errors showed higher probabilistic forecasting skill, while having comparable dichotomous forecasting skill to the original HRRRE ensemble. Other methods of selecting nine ensemble members from the full 63-member suite did not show significant improvement. Flood peak timing showed frequent errors, with average timing errors around five hours early. Larger watersheds tended to have better skill metric scores than smaller basins, with increased skill added by the shifting of QPF.
Abstract
Portions of the northeastern United States (NE) have experienced drought every year since 2016. The U.S. Drought Monitor (USDM) has played an important role in drought characterization and management by providing weekly drought maps across the entire United States, including the NE. Unfortunately, the USDM lacks consistency between input variables leading to difficulties in defining boundaries between drought categories. This paper evaluates the National Water Model’s (NWM) ability to model streamflow and soil moisture, two important hydrological products that are frequently incorporated in drought indices. Using a 26-yr NWM retrospective simulation, comparisons were conducted between NWM output and observations of streamflow and soil moisture, as well as between drought categories derived from the NWM and observations and the USDM. Results indicate that NWM provides moderate predictions of streamflow at NE stations when comparing to historical observations, that NWM streamflow estimators are generally upwardly biased, and performance is worse at lower streamflow magnitudes. The NWM’s ability to predict soil moisture is worse than streamflow, with again a positive bias at most sites and strong variations in anomaly correlation across sites. When predicting drought categories, NWM streamflow is as strong a predictor of USDM drought categories as observed streamflow. Extending the NWM streamflow series using a maintenance of variance technique and only past records provides slight improvements over drought categories derived from the entire 26-yr retrospective simulation. Output from the NWM appears to have some skill in characterizing drought in the NE and provides a spatial resolution to improve the designation of drought boundaries.
Abstract
Portions of the northeastern United States (NE) have experienced drought every year since 2016. The U.S. Drought Monitor (USDM) has played an important role in drought characterization and management by providing weekly drought maps across the entire United States, including the NE. Unfortunately, the USDM lacks consistency between input variables leading to difficulties in defining boundaries between drought categories. This paper evaluates the National Water Model’s (NWM) ability to model streamflow and soil moisture, two important hydrological products that are frequently incorporated in drought indices. Using a 26-yr NWM retrospective simulation, comparisons were conducted between NWM output and observations of streamflow and soil moisture, as well as between drought categories derived from the NWM and observations and the USDM. Results indicate that NWM provides moderate predictions of streamflow at NE stations when comparing to historical observations, that NWM streamflow estimators are generally upwardly biased, and performance is worse at lower streamflow magnitudes. The NWM’s ability to predict soil moisture is worse than streamflow, with again a positive bias at most sites and strong variations in anomaly correlation across sites. When predicting drought categories, NWM streamflow is as strong a predictor of USDM drought categories as observed streamflow. Extending the NWM streamflow series using a maintenance of variance technique and only past records provides slight improvements over drought categories derived from the entire 26-yr retrospective simulation. Output from the NWM appears to have some skill in characterizing drought in the NE and provides a spatial resolution to improve the designation of drought boundaries.
Abstract
Using abundant rainfall gauge measurements and Global Precipitation Mission (GPM) data, spatial patterns of rainfall diurnal cycles and their seasonality over high mountain Asia (HMA) were examined. Spatial distributions of rainfall diurnal cycles over the HMA have a prominent seasonality regulated by circulations at different spatiotemporal scales, within which large regional contrasts are embedded. Rainfall diurnal variability is relatively weak in the premonsoon season, with larger amplitude over the western HMA, the southeastern HMA, as well as southern periphery regions, characterized by a dominant late afternoon to morning rainfall preference. The pattern of rainfall spatial distributions is closely related to the midlatitude westerlies. Both the mean rainfall and amplitudes of diurnal cycles become more pronounced with the advance of monsoon season but weaken during postmonsoon. The widespread late afternoon to night pattern over HMA migrating with seasonal atmospheric circulation is consistent with the lifetime of convective systems, which become active from the afternoon due to radiative heating and decay during the night. Stationary terrain-dependent night-to-morning rainfall patterns are visible in those east–west-orientated valleys over HMA and the Qaidam basin throughout the seasons. This salient geographical dependence is associated with local circulation produced by the strong differential thermal conditions over mountains and valleys, which can lift the warm moist air at the mouth of the valley and trigger nocturnal convection.
Significance Statement
The main purpose of this study is to explore how spatial patterns of rainfall diurnal cycles over high mountain Asia vary with the seasons. Our results show that the widespread late afternoon to night rainfall over high mountain Asia migrating with seasonal atmospheric circulation is consistent with the lifetime of convective systems. Stationary terrain-dependent night-to-morning rainfall patterns are visible in those east–west-orientated valleys over high mountain Asia and the Qaidam basin throughout the seasons. These results highlight the importance of large-scale atmospheric circulation and local circulation on precipitation, which is critical for water resources over high mountain Asia.
Abstract
Using abundant rainfall gauge measurements and Global Precipitation Mission (GPM) data, spatial patterns of rainfall diurnal cycles and their seasonality over high mountain Asia (HMA) were examined. Spatial distributions of rainfall diurnal cycles over the HMA have a prominent seasonality regulated by circulations at different spatiotemporal scales, within which large regional contrasts are embedded. Rainfall diurnal variability is relatively weak in the premonsoon season, with larger amplitude over the western HMA, the southeastern HMA, as well as southern periphery regions, characterized by a dominant late afternoon to morning rainfall preference. The pattern of rainfall spatial distributions is closely related to the midlatitude westerlies. Both the mean rainfall and amplitudes of diurnal cycles become more pronounced with the advance of monsoon season but weaken during postmonsoon. The widespread late afternoon to night pattern over HMA migrating with seasonal atmospheric circulation is consistent with the lifetime of convective systems, which become active from the afternoon due to radiative heating and decay during the night. Stationary terrain-dependent night-to-morning rainfall patterns are visible in those east–west-orientated valleys over HMA and the Qaidam basin throughout the seasons. This salient geographical dependence is associated with local circulation produced by the strong differential thermal conditions over mountains and valleys, which can lift the warm moist air at the mouth of the valley and trigger nocturnal convection.
Significance Statement
The main purpose of this study is to explore how spatial patterns of rainfall diurnal cycles over high mountain Asia vary with the seasons. Our results show that the widespread late afternoon to night rainfall over high mountain Asia migrating with seasonal atmospheric circulation is consistent with the lifetime of convective systems. Stationary terrain-dependent night-to-morning rainfall patterns are visible in those east–west-orientated valleys over high mountain Asia and the Qaidam basin throughout the seasons. These results highlight the importance of large-scale atmospheric circulation and local circulation on precipitation, which is critical for water resources over high mountain Asia.
Abstract
Although several flow routing (FR) algorithms are developed for hydrological modeling, it is still uncertain how the selection of algorithms may affect model results. This study aims to explore the similarity and dissimilarity in model results among different FR algorithms characterized by single flow direction (SD) and multiple flow direction (MD). The Coupled Hydro-Ecological Simulation System (CHESS) was incorporated with six different FR algorithms (D8, D∞, MD∞, MD8, MFD-md, and RMD∞) and then applied for modeling ecohydrological processes for a semiarid mountainous watershed in the western United States during 1991–2012. Comparisons were made between the model results at the catchment and the grid scale. After slightly adjusting one of the most sensitive soil parameters, all algorithms behave similarly in simulating stream hydrographs. When averaged for the watershed, the modeled ecohydrological variables mostly do not differ significantly (<5%) among the six FR algorithms. Nevertheless, the simulated ecohydrological variables are spatially more autocorrelated under the more dispersive MD algorithms. In addition, there exist significant (>5%) cell-level differences in modeled soil moisture among different FR algorithms, with propagated influences on the simulated evapotranspiration and vegetation growth variables. In hillslopes, the cell-level differences in model results tend to increase significantly as the flows move to the streams. Overall, this study proves that the watershed-level differences in model results among FR algorithms are low after model calibration, while significant differences still occur at the cell level. Thus, observational data are essential for testing which routing algorithm captures better the reality of local ecohydrological processes.
Significance Statement
The consideration of flow routing is essential for accurately simulating land surface ecohydrological processes. However, less is known about how the selection of flow routing algorithms may affect the model results. Based on model experiments, we found that the model results under different algorithms do not significantly differ from each other when averaged for the watershed. However, significant differences in model results exist at the individual cell level. These findings are useful for guiding future modeling-related research and also suggest the importance of field studies for testing which routing algorithm can better represent local ecohydrological processes.
Abstract
Although several flow routing (FR) algorithms are developed for hydrological modeling, it is still uncertain how the selection of algorithms may affect model results. This study aims to explore the similarity and dissimilarity in model results among different FR algorithms characterized by single flow direction (SD) and multiple flow direction (MD). The Coupled Hydro-Ecological Simulation System (CHESS) was incorporated with six different FR algorithms (D8, D∞, MD∞, MD8, MFD-md, and RMD∞) and then applied for modeling ecohydrological processes for a semiarid mountainous watershed in the western United States during 1991–2012. Comparisons were made between the model results at the catchment and the grid scale. After slightly adjusting one of the most sensitive soil parameters, all algorithms behave similarly in simulating stream hydrographs. When averaged for the watershed, the modeled ecohydrological variables mostly do not differ significantly (<5%) among the six FR algorithms. Nevertheless, the simulated ecohydrological variables are spatially more autocorrelated under the more dispersive MD algorithms. In addition, there exist significant (>5%) cell-level differences in modeled soil moisture among different FR algorithms, with propagated influences on the simulated evapotranspiration and vegetation growth variables. In hillslopes, the cell-level differences in model results tend to increase significantly as the flows move to the streams. Overall, this study proves that the watershed-level differences in model results among FR algorithms are low after model calibration, while significant differences still occur at the cell level. Thus, observational data are essential for testing which routing algorithm captures better the reality of local ecohydrological processes.
Significance Statement
The consideration of flow routing is essential for accurately simulating land surface ecohydrological processes. However, less is known about how the selection of flow routing algorithms may affect the model results. Based on model experiments, we found that the model results under different algorithms do not significantly differ from each other when averaged for the watershed. However, significant differences in model results exist at the individual cell level. These findings are useful for guiding future modeling-related research and also suggest the importance of field studies for testing which routing algorithm can better represent local ecohydrological processes.
Abstract
Most heavy precipitation events and extreme flooding over the U.S. Pacific coast can be linked to prevalent atmospheric river (AR) conditions. Thus, reliable quantitative precipitation estimation with a rich spatiotemporal resolution is vital for water management and early warning systems of flooding and landslides over these regions. At the same time, high-quality near-real-time measurements of AR precipitation remain challenging due to the complex topographic features of land surface and meteorological conditions of the region: specifically, orographic features occlude radar measurements while infrared-based algorithms face challenges, differentiating between both cold brightband (BB) precipitation and the warmer nonbrightband (NBB) precipitation. It should be noted that the latter precipitation is characterized by greater orographic enhancement. In this study, we evaluate the performance of a recently developed near-real-time satellite precipitation algorithm: Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks (PERSIANN) Dynamic Infrared–Rain Rate-Now (PDIR-Now). This model is primarily dependent on infrared information from geostationary satellites as input; consequently, PDIR-Now has the advantage of short data latency, 15–60-min delay between observation to precipitation product delivery. The performance of PDIR-Now is analyzed with a focus on AR-related events for cases dominated by NBB and BB precipitation over the Russian River basin. In our investigations, we utilize S-band (3-GHz) precipitation profilers with Joss/Parsivel disdrometer measurements at the Middletown and Santa Rosa stations to classify BB and NBB precipitation events. In general, our analysis shows that PDIR-Now is more skillful in retrieving precipitation rates over both BB and NBB events across the topologically complex study area as compared to PERSIANN-Cloud Classification System (CCS). Also, we discuss the performance of well-known operational near-real-time precipitation products from 2017 to 2019. Conventional categorical and volumetric categorical indices, as well as continuous statistical metrics, are used to show the differences between various high-resolution precipitation products such as Multi-Radar Multi-Sensor (MRMS).
Abstract
Most heavy precipitation events and extreme flooding over the U.S. Pacific coast can be linked to prevalent atmospheric river (AR) conditions. Thus, reliable quantitative precipitation estimation with a rich spatiotemporal resolution is vital for water management and early warning systems of flooding and landslides over these regions. At the same time, high-quality near-real-time measurements of AR precipitation remain challenging due to the complex topographic features of land surface and meteorological conditions of the region: specifically, orographic features occlude radar measurements while infrared-based algorithms face challenges, differentiating between both cold brightband (BB) precipitation and the warmer nonbrightband (NBB) precipitation. It should be noted that the latter precipitation is characterized by greater orographic enhancement. In this study, we evaluate the performance of a recently developed near-real-time satellite precipitation algorithm: Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks (PERSIANN) Dynamic Infrared–Rain Rate-Now (PDIR-Now). This model is primarily dependent on infrared information from geostationary satellites as input; consequently, PDIR-Now has the advantage of short data latency, 15–60-min delay between observation to precipitation product delivery. The performance of PDIR-Now is analyzed with a focus on AR-related events for cases dominated by NBB and BB precipitation over the Russian River basin. In our investigations, we utilize S-band (3-GHz) precipitation profilers with Joss/Parsivel disdrometer measurements at the Middletown and Santa Rosa stations to classify BB and NBB precipitation events. In general, our analysis shows that PDIR-Now is more skillful in retrieving precipitation rates over both BB and NBB events across the topologically complex study area as compared to PERSIANN-Cloud Classification System (CCS). Also, we discuss the performance of well-known operational near-real-time precipitation products from 2017 to 2019. Conventional categorical and volumetric categorical indices, as well as continuous statistical metrics, are used to show the differences between various high-resolution precipitation products such as Multi-Radar Multi-Sensor (MRMS).
Abstract
The land surface model is extensively used to simulate turbulence fluxes and hydrological and momentum variables at the land–atmosphere interface. In this study, the Community Land Model, version 5 (CLM5), driven by the 0.1° × 0.1° Chinese Meteorological Forcing Dataset (CMFD) and the field-surveyed soil parameters, is used to simulate land surface processes during 1979–2018. Various high-quality land surface datasets are adopted to assess the model simulations. In general, the CLM5 well captures the monthly variations of 0–10-cm soil moisture in subregions, particularly in the Tibetan Plateau, with an anomaly correlation coefficient between 0.56 and 0.88. However, the simulated soil moisture shows overall wet biases in the whole country, resulting from several reasons. The model simulation is skillful in replicating both the magnitude and spatial pattern when they are compared with the MODIS snow cover dataset. Compared with in situ measured soil temperature in multiple soil layers within 320-cm soil depth from 1980 to 2018, the simulations accurately capture spatial patterns, vertical profiles, and long-term warming trends. For land surface energy components, the simulations have a highly temporal correlation with the observation of Chinese Flux Observation and Research Network (ChinaFLUX) cropland and grassland sites, except for four forest sites, where biases exist in both atmospheric forcing variables and surface vegetation phenology in the model default input dataset. In summary, this study reveals the overall capability of CLM5 in reproducing land surface energy fluxes and hydrological variables over conterminous China, and the validation results may also provide some references for future model improvement and application.
Significance Statement
The offline Community Land Model, version 5 (CLM5), driven by a 0.1° × 0.1° (∼10 km) horizontal resolution atmospheric forcing dataset and a set of field-surveyed soil parameters, are used to simulate the land surface hydrological and heat fluxes in continental China for 1980–2018. The simulated hydrological variables and energy fluxes are validated with various sources of high-quality observation-based datasets. From our systematic evaluations, the current CLM5 high–resolution simulation accurately captures the spatial patterns and temporal variations in most of the water and energy balance components, although biases exist in some simulated variables. Overall, this study reveals the capability of the offline CLM5 simulation in conterminous China and provides the reference for future model improvement and application.
Abstract
The land surface model is extensively used to simulate turbulence fluxes and hydrological and momentum variables at the land–atmosphere interface. In this study, the Community Land Model, version 5 (CLM5), driven by the 0.1° × 0.1° Chinese Meteorological Forcing Dataset (CMFD) and the field-surveyed soil parameters, is used to simulate land surface processes during 1979–2018. Various high-quality land surface datasets are adopted to assess the model simulations. In general, the CLM5 well captures the monthly variations of 0–10-cm soil moisture in subregions, particularly in the Tibetan Plateau, with an anomaly correlation coefficient between 0.56 and 0.88. However, the simulated soil moisture shows overall wet biases in the whole country, resulting from several reasons. The model simulation is skillful in replicating both the magnitude and spatial pattern when they are compared with the MODIS snow cover dataset. Compared with in situ measured soil temperature in multiple soil layers within 320-cm soil depth from 1980 to 2018, the simulations accurately capture spatial patterns, vertical profiles, and long-term warming trends. For land surface energy components, the simulations have a highly temporal correlation with the observation of Chinese Flux Observation and Research Network (ChinaFLUX) cropland and grassland sites, except for four forest sites, where biases exist in both atmospheric forcing variables and surface vegetation phenology in the model default input dataset. In summary, this study reveals the overall capability of CLM5 in reproducing land surface energy fluxes and hydrological variables over conterminous China, and the validation results may also provide some references for future model improvement and application.
Significance Statement
The offline Community Land Model, version 5 (CLM5), driven by a 0.1° × 0.1° (∼10 km) horizontal resolution atmospheric forcing dataset and a set of field-surveyed soil parameters, are used to simulate the land surface hydrological and heat fluxes in continental China for 1980–2018. The simulated hydrological variables and energy fluxes are validated with various sources of high-quality observation-based datasets. From our systematic evaluations, the current CLM5 high–resolution simulation accurately captures the spatial patterns and temporal variations in most of the water and energy balance components, although biases exist in some simulated variables. Overall, this study reveals the capability of the offline CLM5 simulation in conterminous China and provides the reference for future model improvement and application.
Abstract
Root-zone soil moisture (RZSM) is an important variable in land–atmosphere interactions, notably affecting the global climate system. Contrary to satellite-based acquisition of surface soil moisture, RZSM is generally obtained from model-based simulations. In this study, in situ observations from the Naqu and Pali networks that represent different climatic conditions over the Tibetan Plateau (TP) and a triple collocation (TC) method are used to evaluate model-based RZSM products, including Global Land Evaporation Amsterdam Model (GLEAM) (versions 3.5a and 3.5b), Global Land Data Assimilation System (GLDAS) (versions 2.1 and 2.2), and the fifth-generation European Centre for Medium-Range Weather Forecasts reanalysis (ERA5). The evaluation results based on in situ observations indicate that all products tend to overestimate but could generally capture the temporal variation, and ERA5 exhibits the best performance with the highest R (0.875) and the lowest unbiased RMSE (ubRMSE; 0.015 m3 m−3) against in situ observations in the Naqu network. In the TC analysis, similar results are obtained: ERA5 has the best performance with the highest TC-derived R (0.785) over the entire TP, followed by GLEAM v3.5a (0.746) and GLDAS-2.1 (0.682). Meanwhile, GLEAM v3.5a and GLDAS-2.1 outperform GLEAM v3.5b and GLDAS-2.2 over the entire TP, respectively. Besides, possible error causes in evaluating these RZSM products are summarized, and the effectiveness of TC method is also evaluated with two dense networks, finding that TC method is reliable since TC-derived R is close to ground-derived R, with only 6.85% mean relative differences. These results using both in situ observations and TC method may provide a new perspective for the soil moisture product developers to further enhance the accuracy of model-based RZSM over the TP.
Significance Statement
The purpose of this study is to better understand the quality and applicability of GLEAM, GLDAS, and ERA5 RZSM products over the TP using both in situ observations and the triple collocation (TC) method, making it better applied to climate and hydrological research. This study provides four standard statistical metrics evaluation based on in situ observations, as well as the reliable metric, that is, correlation coefficient (R) derived from TC method, and highlights that TC-based evaluation could supplement the ground-based validation, especially over the data-scarce TP region.
Abstract
Root-zone soil moisture (RZSM) is an important variable in land–atmosphere interactions, notably affecting the global climate system. Contrary to satellite-based acquisition of surface soil moisture, RZSM is generally obtained from model-based simulations. In this study, in situ observations from the Naqu and Pali networks that represent different climatic conditions over the Tibetan Plateau (TP) and a triple collocation (TC) method are used to evaluate model-based RZSM products, including Global Land Evaporation Amsterdam Model (GLEAM) (versions 3.5a and 3.5b), Global Land Data Assimilation System (GLDAS) (versions 2.1 and 2.2), and the fifth-generation European Centre for Medium-Range Weather Forecasts reanalysis (ERA5). The evaluation results based on in situ observations indicate that all products tend to overestimate but could generally capture the temporal variation, and ERA5 exhibits the best performance with the highest R (0.875) and the lowest unbiased RMSE (ubRMSE; 0.015 m3 m−3) against in situ observations in the Naqu network. In the TC analysis, similar results are obtained: ERA5 has the best performance with the highest TC-derived R (0.785) over the entire TP, followed by GLEAM v3.5a (0.746) and GLDAS-2.1 (0.682). Meanwhile, GLEAM v3.5a and GLDAS-2.1 outperform GLEAM v3.5b and GLDAS-2.2 over the entire TP, respectively. Besides, possible error causes in evaluating these RZSM products are summarized, and the effectiveness of TC method is also evaluated with two dense networks, finding that TC method is reliable since TC-derived R is close to ground-derived R, with only 6.85% mean relative differences. These results using both in situ observations and TC method may provide a new perspective for the soil moisture product developers to further enhance the accuracy of model-based RZSM over the TP.
Significance Statement
The purpose of this study is to better understand the quality and applicability of GLEAM, GLDAS, and ERA5 RZSM products over the TP using both in situ observations and the triple collocation (TC) method, making it better applied to climate and hydrological research. This study provides four standard statistical metrics evaluation based on in situ observations, as well as the reliable metric, that is, correlation coefficient (R) derived from TC method, and highlights that TC-based evaluation could supplement the ground-based validation, especially over the data-scarce TP region.
