Search

The sensitive ecosystem of the central Himalayan (CH) region, which is experiencing enhanced stress from anthropogenic forcing, requires adequate atmospheric observations and an improved representation of the Himalaya in the models. However, the accuracy of atmospheric models remains limited in this region due to highly complex mountainous topography. This article delineates the effects of spatial resolution on the modeled meteorology and dynamics over the CH by utilizing the Weather Research and Forecasting (WRF) model extensively evaluated against the Ganges Valley Aerosol Experiment (GVAX) observations during the summer monsoon. The WRF simulation is performed over a domain (d01) encompassing northern India at 15 km × 15 km resolution and two nests (d02 at 5 km × 5 km and d03 at 1 km × 1 km) centered over the CH, with boundary conditions from the respective parent domains. WRF simulations reveal higher variability in meteorology, e.g., relative humidity (RH = 70.3 %–96.1 %) and wind speed (WS = 1.1–4.2 m s−1), compared to the ERA-Interim reanalysis (RH = 80.0 %–85.0 %, WS = 1.2–2.3 m s−1) over northern India owing to the higher resolution. WRF-simulated temporal evolution of meteorological variables is found to agree with balloon-borne measurements, with stronger correlations aloft (r = 0.44–0.92) than those in the lower troposphere (r = 0.18–0.48). The model overestimates temperature (warm bias by 2.8 ∘C) and underestimates RH (dry bias by 6.4 %) at the surface in d01. Model results show a significant improvement in d03 (P = 827.6 hPa, T = 19.8 ∘C, RH = 92.3 %), closer to the GVAX observations (P = 801.4 hPa, T = 19.5 ∘C, RH = 94.7 %). Interpolating the output from the coarser domains (d01, d02) to the altitude of the station reduces the biases in pressure and temperature; however, it suppresses the diurnal variations, highlighting the importance of well-resolved terrain (d03). Temporal variations in near-surface P, T, and RH are also reproduced by WRF in d03 to an extent (r>0.5). A sensitivity simulation incorporating the feedback from the nested domain demonstrates the improvement in simulated P, T, and RH over the CH. Our study shows that the WRF model setup at finer spatial resolution can significantly reduce the biases in simulated meteorology, and such an improved representation of the CH can be adopted through domain feedback into regional-scale simulations. Interestingly, WRF simulates a dominant easterly wind component at 1 km × 1 km resolution (d03), which is missing in the coarse simulations; however, the frequency of southeasterlies remains underestimated. The model simulation implementing a high-resolution (3 s) topography input (SRTM) improved the prediction of wind directions; nevertheless, further improvements are required to better reproduce the observed local-scale dynamics over the CH.

Future projections of evapotranspiration (ET) are of critical importance for agricultural and freshwater management and for predicting land–atmosphere feedbacks on the climate system. However, ET from phase 5 of the Coupled Model Intercomparison Project (CMIP5) simulations exhibits substantial biases, bolstering little confidence in future ET projections. Despite poor predictive skill and large bias of ET from the global climate models, the information content necessary to calculate ET offline is available in the models’ archived outputs: temperature T, water vapor pressure e, atmospheric pressure P, and surface net radiation R. A relatively simple three-source energy balance model [Penman–Monteith (PM)], along with the mean annual cycle of remotely sensed vegetation properties, can then be used to reconstruct ET with a substantially reduced bias relative to in situ turbulent heat flux measurements. This methodology is used here to reconstruct ET projections from 2006 through 2100 over North America using output from selected CMIP5 models and to attribute projected ET trends to specific atmospheric controls. CMIP5 ET exhibits substantial bias in annual ET relative to in situ flux measurements across North America (38%–73%; 2006–15), but ET reconstructed from the CMIP5 meteorology with the PM method greatly reduces this bias (−8% to +14%). Present-day North American ET is more sensitive to changes in atmospheric demand for ET (temperature and water vapor pressure) than energy limitation (net radiation), and to a lesser extent vegetation properties (leaf area index). Accordingly, ET is projected to increase 0.26–0.87 mm yr−1 yr−1 over North America through 2100 driven primarily by trends in temperature.

Flash droughts are the rapid intensification of drought conditions generally associated with increased temperatures and decreased precipitation on short time scales. Consequently, flash droughts are responsible for reduced soil moisture which contributes to diminished agricultural yields and lower groundwater levels. Drought management, especially flash drought in the United States is vital to address the human and economic impact of crop loss, diminished water resources and increased wildfire risk. In previous research, climate change scenarios show increased growing season (i.e. frost-free days) and drying in soil moisture over most of the United States by 2100. Understanding projected flash drought is important to assess regional variability, frequency and intensity of flash droughts under future climate change scenarios. Data for this work was produced with the Weather Research and Forecasting (WRF) model. Initial and boundary conditions for the model were supplied by CCSM4, GFDL-ESM2G, and HadGEM2-ES and based on the 8.5 Representative Concentration Pathway (RCP8.5). The WRF model was downscaled to a 12 km spatial resolution for three climate time frames: 1995-2004 (Historical), 2045-2054 (Mid), and 2085-2094 (Late). A key characteristic of flash drought is the rapid onset and intensification of dry conditions. For this, we identify onset with vapor pressure deficit during each time frame. Known flash drought cases during the Historical run are identified and compared to flash droughts in the Mid and Late 21st century.

This study develops a neural-network-based approach for emulating high-resolution modeled precipitation data with comparable statistical properties but at greatly reduced computational cost. The key idea is to use combination of low- and high-resolution simulations (that differ not only in spatial resolution but also in geospatial patterns) to train a neural network to map from the former to the latter. Specifically, we define two types of CNNs, one that stacks variables directly and one that encodes each variable before stacking, and we train each CNN type both with a conventional loss function, such as mean square error (MSE), and with a conditional generative adversarial network (CGAN), for a total of four CNN variants. We compare the four new CNN-derived high-resolution precipitation results with precipitation generated from original high-resolution simulations, a bilinear interpolater and the state-of-the-art CNN-based super-resolution (SR) technique. Results show that the SR technique produces results similar to those of the bilinear interpolator with smoother spatial and temporal distributions and smaller data variabilities and extremes than the original high-resolution simulations. While the new CNNs trained by MSE generate better results over some regions than the interpolator and SR technique do, their predictions are still biased from the original high-resolution simulations. The CNNs trained by CGAN generate more realistic and physically reasonable results, better capturing not only data variability in time and space but also extremes such as intense and long-lasting storms. The new proposed CNN-based downscaling approach can downscale precipitation from 50 to 12 km in 14 min for 30 years once the network is trained (training takes 4 h using 1 GPU), while the conventional dynamical downscaling would take 1 month using 600 CPU cores to generate simulations at the resolution of 12 km over the contiguous United States.

Dynamical downscaling with high-resolution regional climate models may offer the possibility of realistically reproducing precipitation and weather events in climate simulations. As resolutions fall to order kilometers, the use of explicit rather than parametrized convection may offer even greater fidelity. However, these increased resolutions both allow and require increasingly complex diagnostics for evaluating model fidelity. In this study we focus on precipitation evaluation and analyze five 2-month-long dynamically downscaled model runs over the continental United States that employ different convective and microphysics parameterizations, including one high-resolution convection-permitting simulation. All model runs use the Weather Research and Forecasting Model driven by National Center for Environmental Prediction reanalysis data. We show that employing a novel rainstorm identification and tracking algorithm that allocates essentially all rainfall to individual precipitation events (Chang et al. in J Clim 29(23):8355–8376, 2016) allows new insights into model biases. Results include that, at least in these runs, model wet bias is driven by excessive areal extent of individual precipitating events, and that the effect is time-dependent, producing excessive diurnal cycle amplitude. This amplified cycle is driven not by new production of events but by excessive daytime enlargement of long-lived precipitation events. We further show that in the domain average, precipitation biases appear best represented as additive offsets. Of all model configurations evaluated, convection-permitting simulations most consistently reduced biases in precipitation event characteristics.

The aim of this study is to examine projections of extreme temperatures over the continental United States (CONUS) for the 21st century using an ensemble of high spatial resolution dynamically downscaled model simulations with different boundary conditions. The downscaling uses the Weather Research and Forecast model at a spatial resolution of 12 km along with outputs from three different Coupled Model Intercomparison Project Phase 5 global climate models that provide boundary conditions under two different future greenhouse gas (GHG) concentration trajectories. The results from two decadal-length time slices (2045–2054 and 2085–2094) are compared with a historical decade (1995–2004). Probability density functions of daily maximum/minimum temperatures are analyzed over seven climatologically cohesive regions of the CONUS. The impacts of different boundary conditions as well as future GHG concentrations on extreme events such as heat waves and days with temperature higher than 95°F are also investigated. The results show that the intensity of extreme warm temperature in future summer is significantly increased, while the frequency of extreme cold temperature in future winter decreases. The distribution of summer daily maximum temperature experiences a significant warm-side shift and increased variability, while the distribution of winter daily minimum temperature is projected to have a less significant warm-side shift with decreased variability. Using “business-as-usual” scenario, 5-day heat waves are projected to occur at least 5–10 times per year in most CONUS and ≥95°F days will increase by 1–2 months by the end of the century.

This study uses Weather Research and Forecast (WRF) model to evaluate the performance of six dynamical downscaled decadal historical simulations with 12-km resolution for a large domain (7200 × 6180 km) that covers most of North America. The initial and boundary conditions are from three global climate models (GCMs) and one reanalysis data. The GCMs employed in this study are the Geophysical Fluid Dynamics Laboratory Earth System Model with Generalized Ocean Layer Dynamics component, Community Climate System Model, version 4, and the Hadley Centre Global Environment Model, version 2-Earth System. The reanalysis data is from the National Centers for Environmental Prediction-US. Department of Energy Reanalysis II. We analyze the effects of bias correcting, the lateral boundary conditions and the effects of spectral nudging. We evaluate the model performance for seven surface variables and four upper atmospheric variables based on their climatology and extremes for seven subregions across the United States. The results indicate that the simulation’s performance depends on both location and the features/variable being tested. We find that the use of bias correction and/or nudging is beneficial in many situations, but employing these when running the RCM is not always an improvement when compared to the reference data. The use of an ensemble mean and median leads to a better performance in measuring the climatology, while it is significantly biased for the extremes, showing much larger differences than individual GCM driven model simulations from the reference data. This study provides a comprehensive evaluation of these historical model runs in order to make informed decisions when making future projections.

Climate models robustly imply that some significant change in precipitation patterns will occur. Models consistently project that the intensity of individual precipitation events increases by approximately 6-7%/K, following the increase in atmospheric water content, but that total precipitation increases by a lesser amount (1-2 %/K in the global average in transient runs). Some other aspect of precipitation events must then change to compensate for this difference. We develop here a new methodology for identifying individual rainstorms and studying their physical characteristics - including starting location, intensity, spatial extent, duration, and trajectory - that allows identifying that compensating mechanism. We apply this technique to precipitation over the contiguous U.S. from both radar-based data products and high-resolution model runs simulating 80 years of business-as-usual warming. In model studies, we find that the dominant compensating mechanism is a reduction of storm size. In summer, rainstorms become more intense but smaller, in winter, rainstorm shrinkage still dominates, but storms also become less numerous and shorter duration. These results imply that flood impacts from climate change will be less severe than would be expected from changes in precipitation intensity alone. We show also that projected changes are smaller than model-observation biases, implying that the best means of incorporating them into impact assessments is via "data-driven simulations" that apply model-projected changes to observational data. We therefore develop a simulation algorithm that statistically describes model changes in precipitation characteristics and adjusts data accordingly, and show that, especially for summertime precipitation, it outperforms simulation approaches that do not include spatial information., This work has been submitted for publication. Copyright in this work may be transferred without further notice, and this version may no longer be accessible

Regional climate models (RCMs) are a standard tool for downscaling climate forecasts to finer spatial scales. The evaluation of RCMs against observational data is an important step in building confidence in the use of RCMs for future projection. In addition to model performance in climatological means and marginal distributions, a model's ability to capture spatiotemporal relationships is important. This study develops two approaches: (1) spatial correlation/variogram for a range of spatial lags, with total monthly precipitation and nonseasonal precipitation components used to assess the spatial variations of precipitation, and (2) spatiotemporal correlation for a wide range of distances, directions, and time lags, with daily precipitation occurrence used to detect the dynamic features of precipitation. These measures of spatial and spatiotemporal dependence are applied to a high-resolution RCM run and to the National Center for Environmental Prediction (NCEP)-U.S. Department of Energy Atmospheric Model Intercomparison Project II reanalysis data (NCEP-R2), which provide initial and lateral boundary conditions for the RCM. The RCM performs significantly better than NCEP-R2 in capturing both the spatial variations of total and nonseasonal precipitation components and the spatiotemporal correlations of daily precipitation occurrences, which are related to dynamic behavior of precipitating systems. The improvements are apparent not only at resolutions finer than that of NCEP-R2 but also when the RCM and observational data are aggregated to the resolution of NCEP-R2.

The Great Lakes eco-region is one of the largest sources of fresh water in North America; however it is chronically exposed to heavy metal loadings such as mercury. In this study a comprehensive model evaluation was conducted to determine mercury loadings to the Great Lakes. The study also evaluated the relative impact of anthropogenic mercury emissions from China, regional and global sources on deposition to the Great Lakes. For the 2005 study period, CMAQ 4.7.1 model estimated a total of 6.4 ± 0.5 metric tons of mercury deposited in the Great Lakes. The total deposition breakdown showed a net loading for Lake Superior of 1906 ± 246 kg/year which is the highest of all the lakes. Lake Michigan followed with 1645 ± 203 kg/year and 1511 ± 107 kg/year in Lake Huron. The lowest total deposition was seen in Lakes Erie and Ontario amassing annual totals of 837 ± 107 kg and 506 ± 63 kg, respectively. Wet and dry deposition of mercury were both significant pathways and exhibited strong seasonal variability with higher deposition occurring in the warmer months (June–November) and the lowest in winter. Wet deposition of RGM significantly influenced the deposition proportions accounting for roughly 90% of all mercury deposited. Of the three emission sources (global background, integrated planning management (IPM) and Chinese), global background concentrations represented the maximum impact to deposition loading in the Great Lakes, except for Lake Erie and parts of Lake Michigan. There was minimal seasonality for the global background, but differences in percentage contribution between dry (28–97%) and wet deposition (43–98%) was predicted. The contributions were seen mainly in the northern sections of the Great Lakes further away from IPM point sources. These findings suggest strong localized impact of IPM sources on the southernmost lakes. Deposition as a result of emissions from China exhibited seasonality in both wet and dry deposition and showed significant contributions ranging from 0.2 to 9%.

The lake-breeze circulation that forms over Lake Michigan during the summer influences the Chicago, Illinois, metropolitan area’s weather in several ways. Of particular significance is the circulation’s effect on the dispersion of pollutants such as ozone and aerosols produced in and around the city. To investigate these effects, the fifth-generation Pennsylvania State University–National Center for Atmospheric Research Mesoscale Model (MM5) was used to perform numerical simulations for two lake-breeze events—one in July 1999 and another in July 2002. The model runs were verified with data from several locations around the Chicago area. The simulated breeze circulation decreased the rate of increase in air temperature while penetrating roughly 12 km inland and lasting about 8 h, in reasonable agreement with observations. Furthermore, the inland penetration distance was related to the strength of the maximum vertical velocity within the front. Calculations of trajectories and transport of particles showed that the breeze tended to transport particles trapped within it to the north when release occurred before the circulation came ashore, whereas particles released at the time of the breeze’s landfall or afterward moved more northeasterly, in the direction of the prevailing wind. Thirty-four percent of all released particles were trapped by the circulation and raised to a height of at least 300 m, and 20% of the particles remained in the lowest 100 m above the surface. In addition, sensitivity tests showed little change in the modeled breeze when measured surface temperatures for Lake Michigan were used as initial conditions and boundary conditions in the place of surface skin temperature (as derived by the National Centers for Environmental Prediction). Raising the lake temperatures significantly in the simulation yielded a more elongated vertical circulation and a briefer lake-breeze event that did not reach as far inland.

Natural radionuclides have been proposed as a means of assessing the transport of ozone (O3) and aerosols in the troposphere. Beryllium-7 (7Be) is produced in the upper troposphere and lower stratosphere by the interaction of cosmogenic particles with atmospheric nitrogen and oxygen. 7Be has a 53.29-day half-life (478 keV gamma) and is known to attach to fine particles in the atmosphere once it is formed. It has been suggested that O3 from aloft can be transported into rural and urban regions during stratospheric-tropospheric folding events leading to increased background levels of O3 at the surface. 7Be can be used as a tracer of upper atmospheric air parcels and the O3 associated with them. Aerosol samples with a 2.5-microm cutoff were collected during 12-hr cycles (day/night) for a 30-day period at Deer Park, TX, near Houston, in August-September of 2000, and at Waddell, AZ, near Phoenix, in June-July of 2001. A comparison of 7Be levels with 12-hr O3 averages and maxima shows little correlation. Comparison of nighttime and daytime O3 levels indicate that during the day, when mixing is anticipated to be higher, the correlation of 7Be with O3 in Houston is approximately twice that observed at night. This is consistent with mixing and with the anticipated loss of O3 by reaction with nitric oxide (NO) and dry deposition. At best, 30% of the O3 variance can be explained by the correlation with 7Be for Houston, less than that for Phoenix where no significant correlation was seen. This result is consistent with the intercept values obtained for 7Be correlations with either O3 24-hr averages or O3 12-hr maxima and is also in the range of the low O3 levels (25 ppb) observed at Deer Park during a tropical storm event where the O3 is attributable primarily to background air masses. That is, maximum background O3 level contributions from stratospheric sources aloft are estimated to be in the range of 15-30 ppb in the Houston, TX, and Phoenix, AZ, area, and levels above these are because of local tropospheric photochemical production.

In the May and June of 1998, field measurements were taken at a site near the Usery Pass Recreation Area, ∼27 miles from the downtown Phoenix area, overlooking Phoenix and Mesa, Arizona. This site was selected to examine the impacts of the Phoenix urban plume on the Usery Pass Recreation Area and surrounding regions. Data were obtained for ultraviolet-B (UVB) radiation, nitrogen dioxide (NO 2 ), peroxyacetyl nitrate (PAN), ozone (O 3 ), and carbon monoxide (CO). Nocturnal plumes of NO 2 (in tens of ppb), observed near midnight, were correlated with CO and anti-correlated with O 3 . This behavior was consistent with the titration of locally generated NO by boundary layer O 3 to form the nighttime NO 2 plumes that were subsequently transported into the Usery Pass Recreation area. Nitrate radical (NO 3 ) production rates were calculated to be very high on the edges of these nocturnal plumes. Examination of O 3 and PAN data also indicates that Phoenix is being affected by long-range transport of pollutants from the Los Angeles to San Diego areas. A regional smoke episode was observed in May, accompanied by a decrease in UVB of factor of two and a decrease in O 3 and an increase in methyl chloride. Low level back trajectories and chemical evidence confirm that the smoke event originated in northern Mexico and that the reduced O 3 levels observed at Usery Pass could be partially due to reduced photolysis rates caused by carbonaceous soot aerosols transported in the smoke plume. The results are discussed with regard to potential effects of local pollution transport from the Phoenix air basin as well as an assessment of the contributions from long-range transport of pollutants to the background levels in the Phoenix-Usery Pass area.

The processed excess 14C data are used to calibrate the cross-tropopause transport in a two-dimensional model. The model results are used to diagnose the mechanisms responsible for cross-tropopause transport in the model and the sensitivity of the transport rate to the changes in the transport parameter. Although the total flux across the tropopause is dominated by the eddy diffusion flux along the isentropic surface due to the large values assigned to the diffusion coefficient Kyy at the boundary, the cross-tropopause transport is more sensitive to changes in the circulation. Nitrogen oxides emitted by engines of high-speed civil transport could cause ozone depletion in the lower stratosphere. The expected depletion is proportional to the amount of oxides of nitrogen retained in the stratosphere. For a typical fleet, lowering the tropopause height by 1.2 km can cause a 14–22% increase in the amount of reactive nitrogen retained in the stratosphere. Thus it is necessary to have a fine enough vertical resolution near the tropopause in the two-dimensional model for determining the stratospheric residence time of the engine emissions deposited near the tropopause.

[1] Direct measurements of the dry deposition velocity of peroxyacetyl nitrate (PAN) were made during the daytime between the months of July and October above a grassland surface in northern Illinois by a modified Bowen ratio technique. Differences in the air temperature, water vapor content, and PAN concentration were measured between the heights of 3.0 m and 0.92 m. Although the measurement uncertainties were large, the cumulative data indicate a slight downward flux of PAN, with an average and standard error of 0.13 ± 0.13 cm s−1 for the dry deposition velocity. Theoretical calculations showed that thermochemical decomposition of PAN on leaf and soil surfaces heated to temperatures above the ambient air levels would contribute less than 15% of the total PAN flux at the elevations of the PAN measurements. A theoretical evaluation of the transfer of PAN through leaf stomata and the plant cuticular membrane indicated that uptake of PAN by vegetation during the daytime is controlled by transfer through the leaf stomata rather than the cuticular membrane. The stomatal resistance for PAN is greater by a factor of 1.6 than the value for O3. The mesophyll resistance for O3 is also expected to be less than the value for PAN, because O3 has more reaction sites within plant cells and reacts faster than PAN with protein thiols of the cell membranes. Measurements from other studies indicate that the dry deposition velocity for PAN above a vegetated surface during the daytime is lower by a factor of 0.5–0.3 than for O3. Our measurements of the PAN deposition velocity agree with the results of previous studies and with theoretical calculations based on the physicochemical properties of PAN and the grassland surface. These measurements imply that removal of PAN from the daytime atmospheric boundary layer by thermochemical decomposition is more rapid than dry deposition to a grassland surface.