Browse
Abstract
Statistical downscaling (SD) of climate change projections is a key piece for impact and adaptation studies, due to its low computational expense compared to dynamical downscaling, which allows to explore uncertainties through the generation of large ensembles. SD has been extensively evaluated and applied in the extratropics, but few examples exist in tropical regions. In this study several state-of-the-art methods belonging to different families have been evaluated for maximum/minimum daily temperature and daily accumulated precipitation (both from the ERA5 reanalysis at 0.25°) in two regions with very different climates: Spain (Mid-latitudes) and Central America (Tropics). Some key assumptions of SD have been tested: the strength of the predictors/predictand links, the skill of different approaches and the extrapolation capability of each method. It has been found that relevant predictors are different in both regions, as well as the behavior of statistical methods. For temperature, most methods perform significantly better in Spain than in Central America, where Transfer Function methods present important extrapolation problems, probably due to the low variability of the training sample (present climate). In both regions, Model Output Statistics (MOS) methods have achieved the best results for temperature. In Central America Transfer Function (TF) methods have achieved better results than MOS methods in the evaluation in the present climate, but they do not preserve trends in the future. For precipitation, MOS methods and the machine learning method eXtreme Gradient Boost have achieved the best results in both regions. Additionally, it has been found that although the use of humidity indexes as predictors improve results for the downscaling of precipitation, future trends given by statistical methods are very sensitive to the use of one or another index. Three indexes have been compared: relative humidity, specific humidity and dew point depression. The use of the specific humidity has been found to seriously deviate trends given by the downscaled projections from those given by raw Global Climate Models in both regions.
Abstract
Statistical downscaling (SD) of climate change projections is a key piece for impact and adaptation studies, due to its low computational expense compared to dynamical downscaling, which allows to explore uncertainties through the generation of large ensembles. SD has been extensively evaluated and applied in the extratropics, but few examples exist in tropical regions. In this study several state-of-the-art methods belonging to different families have been evaluated for maximum/minimum daily temperature and daily accumulated precipitation (both from the ERA5 reanalysis at 0.25°) in two regions with very different climates: Spain (Mid-latitudes) and Central America (Tropics). Some key assumptions of SD have been tested: the strength of the predictors/predictand links, the skill of different approaches and the extrapolation capability of each method. It has been found that relevant predictors are different in both regions, as well as the behavior of statistical methods. For temperature, most methods perform significantly better in Spain than in Central America, where Transfer Function methods present important extrapolation problems, probably due to the low variability of the training sample (present climate). In both regions, Model Output Statistics (MOS) methods have achieved the best results for temperature. In Central America Transfer Function (TF) methods have achieved better results than MOS methods in the evaluation in the present climate, but they do not preserve trends in the future. For precipitation, MOS methods and the machine learning method eXtreme Gradient Boost have achieved the best results in both regions. Additionally, it has been found that although the use of humidity indexes as predictors improve results for the downscaling of precipitation, future trends given by statistical methods are very sensitive to the use of one or another index. Three indexes have been compared: relative humidity, specific humidity and dew point depression. The use of the specific humidity has been found to seriously deviate trends given by the downscaled projections from those given by raw Global Climate Models in both regions.
Abstract
Accurate and reliable precipitation product on regular grids is essential for understanding trends and variability within climate studies, for weather forecasting, and in hydrology and agrometeorology applications. However, the construction of high-resolution spatiotemporal precipitation grid product is challenging for complex terrain with sparse rain gauge networks and when only coarse spatial resolutions of satellite data are available. The objective of this study was to consequently provide a practical method to create grid precipitation product by merging accurate quantitative observations from weather stations with continuous spatial information and from satellite-based estimate product. The new gridded precipitation product exhibits a monthly temporal resolution and a spatial resolution of 0.01° for the Tianshan Mountains, while extending back to 1981. To overcome the limitation of low densities and sparse distributions of meteorological stations in the complex terrain of the Tianshan Mountains, a suitable interpolation of ANUSPLIN was used to interpolate grid precipitation based on in situ data. The interpolation grid precipitation was then merged with the satellite precipitation product developed by the U.S. Geological Survey and the Climate Hazards Group. After evaluation and validation using withheld stations and comparison to reference datasets, the result indicated that the merged product exhibit considerable promise for application in complex terrain. The method can be widely applied and is expected to construct precipitation product with high spatial and temporal resolution by merging multiple precipitation data sources.
Abstract
Accurate and reliable precipitation product on regular grids is essential for understanding trends and variability within climate studies, for weather forecasting, and in hydrology and agrometeorology applications. However, the construction of high-resolution spatiotemporal precipitation grid product is challenging for complex terrain with sparse rain gauge networks and when only coarse spatial resolutions of satellite data are available. The objective of this study was to consequently provide a practical method to create grid precipitation product by merging accurate quantitative observations from weather stations with continuous spatial information and from satellite-based estimate product. The new gridded precipitation product exhibits a monthly temporal resolution and a spatial resolution of 0.01° for the Tianshan Mountains, while extending back to 1981. To overcome the limitation of low densities and sparse distributions of meteorological stations in the complex terrain of the Tianshan Mountains, a suitable interpolation of ANUSPLIN was used to interpolate grid precipitation based on in situ data. The interpolation grid precipitation was then merged with the satellite precipitation product developed by the U.S. Geological Survey and the Climate Hazards Group. After evaluation and validation using withheld stations and comparison to reference datasets, the result indicated that the merged product exhibit considerable promise for application in complex terrain. The method can be widely applied and is expected to construct precipitation product with high spatial and temporal resolution by merging multiple precipitation data sources.
Abstract
This study quantifies the potential effect of the lee vortex on the fine particulate matter (PM2.5) pollution deterioration under the complex topography in Taiwan using observational data. We select the lee-vortex days that favor the development of the lee vortices in northwestern Taiwan under the southeasterly synoptic winds. We then define the enhancement index that discerns the areas with the high occurrence frequencies of the PM2.5 enhancement under the flow regime relative to the seasonal background concentrations. Under the lee-vortex days, the center of western Taiwan exhibits enhancement indices higher than 0.65. In addition, during the consecutive lee-vortex days, the index characterizes a northward shift in the PM2.5-enhanced areas under the subtle transition of the background wind directions. The areas with indices higher than 0.65 expand on the second day in northwestern Taiwan; the number of stations exhibiting indices higher than 0.8 increases by threefold from the first to the second day. The idealized numerical simulations using the Taiwan vector vorticity equation cloud-resolving model (TaiwanVVM) explicitly resolve the structures of leeside circulations and the associated pollutant transport, and their evolutions are highly sensitive to the background winds.
Significance Statement
Our study investigates the challenging question of local circulation patterns affected by mountain orography and the associated pollutant transport. We analyzed long-term balloon sounding and ground station observations to select the weather regime favoring the formation of lee vortices on Taiwan island. We then quantified the areas with a frequent enhancement of particulate pollutants. The long-term trend of the lee-vortex days exhibited a significant increase in the past decades. The pollution enhancement areas are highly consistent with the regions dominated by leeside local circulation. Together with the idealized high-resolution simulations, we identified that the detailed evolution of the lee vortices is highly sensitive to the subtle changes in background wind direction and hence the redistribution of local pollution.
Abstract
This study quantifies the potential effect of the lee vortex on the fine particulate matter (PM2.5) pollution deterioration under the complex topography in Taiwan using observational data. We select the lee-vortex days that favor the development of the lee vortices in northwestern Taiwan under the southeasterly synoptic winds. We then define the enhancement index that discerns the areas with the high occurrence frequencies of the PM2.5 enhancement under the flow regime relative to the seasonal background concentrations. Under the lee-vortex days, the center of western Taiwan exhibits enhancement indices higher than 0.65. In addition, during the consecutive lee-vortex days, the index characterizes a northward shift in the PM2.5-enhanced areas under the subtle transition of the background wind directions. The areas with indices higher than 0.65 expand on the second day in northwestern Taiwan; the number of stations exhibiting indices higher than 0.8 increases by threefold from the first to the second day. The idealized numerical simulations using the Taiwan vector vorticity equation cloud-resolving model (TaiwanVVM) explicitly resolve the structures of leeside circulations and the associated pollutant transport, and their evolutions are highly sensitive to the background winds.
Significance Statement
Our study investigates the challenging question of local circulation patterns affected by mountain orography and the associated pollutant transport. We analyzed long-term balloon sounding and ground station observations to select the weather regime favoring the formation of lee vortices on Taiwan island. We then quantified the areas with a frequent enhancement of particulate pollutants. The long-term trend of the lee-vortex days exhibited a significant increase in the past decades. The pollution enhancement areas are highly consistent with the regions dominated by leeside local circulation. Together with the idealized high-resolution simulations, we identified that the detailed evolution of the lee vortices is highly sensitive to the subtle changes in background wind direction and hence the redistribution of local pollution.
Abstract
Removing nonweather echoes is a critical component of the quality control (QC) chain used in the context of radar data assimilation for numerical weather prediction, quantitative precipitation estimation, and nowcasting applications. Recent studies show that using a simple QC method based on the depolarization ratio (DR) performs remarkably well in many situations. Nonetheless, this method may misclassify echoes in regions affected by nonuniform beamfilling or melting particles. This study presents an updated version of this QC used to remove nonweather echoes that uses the DR-based classification together with a set of physically based rules for correcting misclassifications of hail, nonuniform beamfilling, and melting particles. The potential of the new QC is evaluated using a continental-scale monitoring framework that compares the radar observations after QC with the precipitation occurrence derived from aviation routine weather reports (METARs). For this evaluation, the study uses the radar data and the METARs available over North America during the summer of 2019 and winter of 2020. In addition, the study demonstrates the usefulness of the monitoring framework to determine the optimal QC configuration. Some practical limitations of using the METAR-derived precipitation to assess radar data quality are also discussed.
Abstract
Removing nonweather echoes is a critical component of the quality control (QC) chain used in the context of radar data assimilation for numerical weather prediction, quantitative precipitation estimation, and nowcasting applications. Recent studies show that using a simple QC method based on the depolarization ratio (DR) performs remarkably well in many situations. Nonetheless, this method may misclassify echoes in regions affected by nonuniform beamfilling or melting particles. This study presents an updated version of this QC used to remove nonweather echoes that uses the DR-based classification together with a set of physically based rules for correcting misclassifications of hail, nonuniform beamfilling, and melting particles. The potential of the new QC is evaluated using a continental-scale monitoring framework that compares the radar observations after QC with the precipitation occurrence derived from aviation routine weather reports (METARs). For this evaluation, the study uses the radar data and the METARs available over North America during the summer of 2019 and winter of 2020. In addition, the study demonstrates the usefulness of the monitoring framework to determine the optimal QC configuration. Some practical limitations of using the METAR-derived precipitation to assess radar data quality are also discussed.
Abstract
October-September runoff increased 6% and 17% in the Upper (UMRB) and Lower (LMRB) Missouri River Basin, respectively, in a recent (1990-2019) compared to past (1960-1989) climate. The runoff increases were unanticipated, given various projections for semi-permanent drought and/or aridification in the North American Great Plains. Here, five transient coupled climate model ensembles are used to diagnose the effects of natural internal variability and anthropogenic climate change on the observed runoff increases, and to project UMRB and LMRB runoff to the mid-21st century.
The runoff increases observed in the recent compared to the past climate were not due to anthropogenic climate change, but resulted mostly from an extreme occurrence of internal multi-decadal variability. High runoff resulted from large, mostly internally-generated, precipitation increases (6% in the UMRB and 5% in the LMRB) that exceeded simulated increases due to climate change forcing alone (0-2% inter-model range). The precipitation elasticity of runoff, which relates runoff sensitivity to precipitation differences in the recent compared to the past climate, led to 1-3-fold and 2-4-fold amplifications of runoff versus precipitation in the UMRB and LMRB, respectively. Without the observed precipitation increases in the recent compared to the past climate, effects of human-induced warming of about 1°C would alone have most likely induced runoff declines of 7% and 13% in the UMRB and LMRB, respectively. Ensemble model simulations overwhelmingly project lower UMRB and LRMB runoff by 2050 compared to 1990-2019, a change found to be insensitive to whether individual realizations experienced high flows in the recent climate.
Abstract
October-September runoff increased 6% and 17% in the Upper (UMRB) and Lower (LMRB) Missouri River Basin, respectively, in a recent (1990-2019) compared to past (1960-1989) climate. The runoff increases were unanticipated, given various projections for semi-permanent drought and/or aridification in the North American Great Plains. Here, five transient coupled climate model ensembles are used to diagnose the effects of natural internal variability and anthropogenic climate change on the observed runoff increases, and to project UMRB and LMRB runoff to the mid-21st century.
The runoff increases observed in the recent compared to the past climate were not due to anthropogenic climate change, but resulted mostly from an extreme occurrence of internal multi-decadal variability. High runoff resulted from large, mostly internally-generated, precipitation increases (6% in the UMRB and 5% in the LMRB) that exceeded simulated increases due to climate change forcing alone (0-2% inter-model range). The precipitation elasticity of runoff, which relates runoff sensitivity to precipitation differences in the recent compared to the past climate, led to 1-3-fold and 2-4-fold amplifications of runoff versus precipitation in the UMRB and LMRB, respectively. Without the observed precipitation increases in the recent compared to the past climate, effects of human-induced warming of about 1°C would alone have most likely induced runoff declines of 7% and 13% in the UMRB and LMRB, respectively. Ensemble model simulations overwhelmingly project lower UMRB and LRMB runoff by 2050 compared to 1990-2019, a change found to be insensitive to whether individual realizations experienced high flows in the recent climate.
Abstract
The southern Great Plains (SGP) is defined by hydrometeorological swings between dry and wet extremes. These swings exacerbate the climatological gradients of moisture (from east to west) and temperature (from south to north), which can impact the agricultural production of the region. Thus, it is key to understand extremes to sustainably maintain agricultural success in the region. This study investigates the wet extremes, or extreme precipitation events, that have become more prominent in the last two decades. Data from 108 U.S. Historical Climatology Network stations were analyzed for the 1950–2020 period to detect changes in the frequency and magnitude of extreme precipitation events. Results show that changes in the magnitude of extreme precipitation are isolated and scattered across the SGP, with only the winter season showing regional shifts in extreme precipitation magnitude. Changes in the frequency of extreme precipitation events were noted across the entire SGP, although the changes in frequency are more notable in the eastern SGP than in the western SGP. Analysis shows that the increased number of events detected is driven more, but not exclusively, by the increasing spatial extent of individual extreme precipitation events than by an increased number of events. Overall, these results depict the changing nature of extreme precipitation within the SGP and differences in extreme precipitation between the eastern and western SGP.
Abstract
The southern Great Plains (SGP) is defined by hydrometeorological swings between dry and wet extremes. These swings exacerbate the climatological gradients of moisture (from east to west) and temperature (from south to north), which can impact the agricultural production of the region. Thus, it is key to understand extremes to sustainably maintain agricultural success in the region. This study investigates the wet extremes, or extreme precipitation events, that have become more prominent in the last two decades. Data from 108 U.S. Historical Climatology Network stations were analyzed for the 1950–2020 period to detect changes in the frequency and magnitude of extreme precipitation events. Results show that changes in the magnitude of extreme precipitation are isolated and scattered across the SGP, with only the winter season showing regional shifts in extreme precipitation magnitude. Changes in the frequency of extreme precipitation events were noted across the entire SGP, although the changes in frequency are more notable in the eastern SGP than in the western SGP. Analysis shows that the increased number of events detected is driven more, but not exclusively, by the increasing spatial extent of individual extreme precipitation events than by an increased number of events. Overall, these results depict the changing nature of extreme precipitation within the SGP and differences in extreme precipitation between the eastern and western SGP.
Abstract
In this study, the possible climate change impacts on irrigated corn production in the lower Mississippi delta (LMD) region were analyzed. The observed daily maximum and minimum air temperature, wind speed, relative humidity (RH), and precipitation from 1960 to 2018 were used in the analysis. The length of the growing season, evapotranspiration (ET), and crop yield estimates from the precalibrated Root Zone Water Quality Model (RZWQM) were also used. Trend analyses were performed on growing-season averages for temperature, RH, and wind speed; growing-season totals for precipitation and ET; daily values of minimum and maximum temperature; and averages of RH and wind speed at critical corn growth stages. The last day of spring freezing (LDSF) and days with an average daily air temperature Ta of more than 35°C during corn silking were also included in the analysis. The trend analysis was performed using the modified Mann–Kendall test, Pettitt test, and Sen’s slope at a significance level of p ≤ 0.05. The results from our study pointed out increases in minimum Ta , increases in the number of days with Ta exceeding 35°C during the corn silking stage, increases in RH and decreases in ET, advancement of the LDSF by 2 weeks, and 8% reductions in corn yield that could be attributed to changes in climate. If the observed trends in climate (climate variability and change) and yield reductions continue in the region, it could be challenging to grow the corn crop in the LMD profitably.
Significance Statement
In recent years, growing corn and soybeans instead of cotton has been gaining popularity in the lower Mississippi delta (LMD) region. Our study is aimed to understand how climate change in the recent past (1960–2018) has affected irrigated corn production in LMD. This will help us to better understand the expected consequences of the future climate. We used observations of daily weather data from 1960 to 2018 and corn yield, water balance, and crop yield results from a computer model designed to simulate corn production. Tools were used to analyze trends and patterns in the data and results. Our analysis pointed out significant changes in weather, water balance, and yield decline for irrigated corn in the LMD.
Abstract
In this study, the possible climate change impacts on irrigated corn production in the lower Mississippi delta (LMD) region were analyzed. The observed daily maximum and minimum air temperature, wind speed, relative humidity (RH), and precipitation from 1960 to 2018 were used in the analysis. The length of the growing season, evapotranspiration (ET), and crop yield estimates from the precalibrated Root Zone Water Quality Model (RZWQM) were also used. Trend analyses were performed on growing-season averages for temperature, RH, and wind speed; growing-season totals for precipitation and ET; daily values of minimum and maximum temperature; and averages of RH and wind speed at critical corn growth stages. The last day of spring freezing (LDSF) and days with an average daily air temperature Ta of more than 35°C during corn silking were also included in the analysis. The trend analysis was performed using the modified Mann–Kendall test, Pettitt test, and Sen’s slope at a significance level of p ≤ 0.05. The results from our study pointed out increases in minimum Ta , increases in the number of days with Ta exceeding 35°C during the corn silking stage, increases in RH and decreases in ET, advancement of the LDSF by 2 weeks, and 8% reductions in corn yield that could be attributed to changes in climate. If the observed trends in climate (climate variability and change) and yield reductions continue in the region, it could be challenging to grow the corn crop in the LMD profitably.
Significance Statement
In recent years, growing corn and soybeans instead of cotton has been gaining popularity in the lower Mississippi delta (LMD) region. Our study is aimed to understand how climate change in the recent past (1960–2018) has affected irrigated corn production in LMD. This will help us to better understand the expected consequences of the future climate. We used observations of daily weather data from 1960 to 2018 and corn yield, water balance, and crop yield results from a computer model designed to simulate corn production. Tools were used to analyze trends and patterns in the data and results. Our analysis pointed out significant changes in weather, water balance, and yield decline for irrigated corn in the LMD.
Abstract
This study investigates the impact of future climate warming on tropical cyclones (TC) and extratropical cyclones (ETC) using the database for Policy Decision-Making for Future Climate Change (d4PDF) large ensemble simulations. Cyclone tracking was performed using the neighbor enclosed area tracking algorithm (NEAT), and TC and ETCs were identified over the western North Pacific Ocean (WNP). For cyclone frequency, it was revealed that, although a slight underestimation of the total number of TCs and ETCs in both the WNP and near Hokkaido, Japan, exists, the d4PDF reproduced the spatial distribution of both TC and ETC tracks well when compared with observations/reanalysis. The 4-K warming scenarios derived from six different sea surface temperature warming patterns showed robust decreases in TC frequency in the tropical WNP and a slight reduction in ETCs near Japan. Next, precipitation characteristics for TCs or ETCs in the vicinity of Hokkaido were examined using 5-km-mesh regional climate ensemble simulations. Four representative cyclone locations near Hokkaido are identified using K-means clustering and revealed distinct precipitation characteristics between clusters, with higher TC-associated precipitation than ETC-associated precipitation and the heaviest precipitation in the southern portion of the prefecture. The 4-K warming scenarios revealed increased precipitation for all cyclone placements for both TCs and ETCs. Last, average cyclone intensity, translation speed, and size were examined. It was shown that TCs in future climates are more intense, propagate more slowly, and are smaller in terms of enclosed vorticity area as they approach Hokkaido. For ETCs, mean intensity does not change much; they travel slightly faster, and become smaller.
Abstract
This study investigates the impact of future climate warming on tropical cyclones (TC) and extratropical cyclones (ETC) using the database for Policy Decision-Making for Future Climate Change (d4PDF) large ensemble simulations. Cyclone tracking was performed using the neighbor enclosed area tracking algorithm (NEAT), and TC and ETCs were identified over the western North Pacific Ocean (WNP). For cyclone frequency, it was revealed that, although a slight underestimation of the total number of TCs and ETCs in both the WNP and near Hokkaido, Japan, exists, the d4PDF reproduced the spatial distribution of both TC and ETC tracks well when compared with observations/reanalysis. The 4-K warming scenarios derived from six different sea surface temperature warming patterns showed robust decreases in TC frequency in the tropical WNP and a slight reduction in ETCs near Japan. Next, precipitation characteristics for TCs or ETCs in the vicinity of Hokkaido were examined using 5-km-mesh regional climate ensemble simulations. Four representative cyclone locations near Hokkaido are identified using K-means clustering and revealed distinct precipitation characteristics between clusters, with higher TC-associated precipitation than ETC-associated precipitation and the heaviest precipitation in the southern portion of the prefecture. The 4-K warming scenarios revealed increased precipitation for all cyclone placements for both TCs and ETCs. Last, average cyclone intensity, translation speed, and size were examined. It was shown that TCs in future climates are more intense, propagate more slowly, and are smaller in terms of enclosed vorticity area as they approach Hokkaido. For ETCs, mean intensity does not change much; they travel slightly faster, and become smaller.
