Noah Diffenbaugh | Stanford University (original) (raw)

Papers by Noah Diffenbaugh

Environmental Research Letters, 2016

Global change biology, Jan 28, 2016

Understanding tropical rainforest carbon exchange and its response to heat and drought is critica... more Understanding tropical rainforest carbon exchange and its response to heat and drought is critical for quantifying the effects of climate change on tropical ecosystems, including global climate-carbon feedbacks. Of particular importance for the global carbon budget is net biome exchange of CO2 with the atmosphere (NBE), which represents nonfire carbon fluxes into and out of biomass and soils. Subannual and sub-Basin Amazon NBE estimates have relied heavily on process-based biosphere models, despite lack of model agreement with plot-scale observations. We present a new analysis of airborne measurements that reveals monthly, regional-scale (~1-8 × 10(6) km(2) ) NBE variations. We develop a regional atmospheric CO2 inversion that provides the first analysis of geographic and temporal variability in Amazon biosphere-atmosphere carbon exchange and that is minimally influenced by biosphere model-based first guesses of seasonal and annual mean fluxes. We find little evidence for a clear s...

The 4th IPCC report concludes that climate change is now unequivocal and projected increases in e... more The 4th IPCC report concludes that climate change is now unequivocal and projected increases in evaporation and atmospheric water content could intensify the hydrological cycle. However, the coarse spatial resolution and biases in global climate model simulations limit their usefulness in climate impact assessment. In order to improve on these limitations, we used high-resolution regional climate model (RegCM3) simulations, covering

Agu Fall Meeting Abstracts, Dec 1, 2002

Anthropogenic increases in atmospheric carbon dioxide (CO2) concentrations may affect vegetation ... more Anthropogenic increases in atmospheric carbon dioxide (CO2) concentrations may affect vegetation distribution both directly through changes in photosynthesis and water-use efficiency, and indirectly through CO_{2} induced climate change. Additionally, changes in vegetation distribution due to these direct and indirect effects may induce land surface-atmosphere feedbacks that create further change in both regional climate and regional vegetation distribution. Using a regional climate model (RegCM2.5) coupled to an equilibrium vegetation model (BIOME4), we quantitatively tested the sensitivity of climate and vegetation in the western United States to both the direct and indirect effects of doubled pre-Industrial atmospheric CO2 concentrations and to land surface-atmospheric feedbacks induced by the initial vegetation sensitivities. In assessing regional vegetation responses to the initial effects of elevated CO_{2} levels, vegetation in the western United States was sensitive to changes in photosynthesis and water use efficiency caused by increased CO2 availability, with woody biome types replacing less woody types throughout the domain. Vegetation was also sensitive to the initial climatic effects of increased CO_{2} concentrations, particularly at high elevations, both due to warming throughout the domain and to decreased precipitation in key mountain regions such as the Sierra Nevada and the Cascade and Blue Mountains of Oregon. Additionally, these patterns changed when the initial climatic and non-climatic effects of CO2 on vegetation were tested in combination, creating sensitivities not seen in either of the individual cases and indicating that climatic and non-climatic effects must be considered in tandem when assessing the potential impacts of elevated CO_{2} levels. Finally, asynchronous coupling of RegCM2.5 and BIOME4 tested the role of land surface-atmosphere feedbacks in shaping the regional response to elevated global atmospheric CO2 concentrations. The initial regional vegetation responses to elevated atmospheric CO_{2} levels resulted in regional land surface-atmosphere feedbacks that further altered regional climate and vegetation distribution, particularly along ecotones. Because the asynchronous coupling of RegCM2.5 and BIOME4 represents a new regional climate model application, further work is required to constrain the model sensitivity to choice in methodological variables such as length and number of iterations and domain size and placement.

ABSTRACT In recent years, accelerated glacier recession trends have been reported for the Himalay... more ABSTRACT In recent years, accelerated glacier recession trends have been reported for the Himalayan region, based largely on the studies in the Eastern Himalayas. However, recent studies carried out in the Karakoram Himalayas, suggest an expansion of glaciers and a reduction in summer streamflow due to a significant decrease in summer temperatures. Snow and ice melt from these glaciers is the primary input to the Upper Indus River, upstream of Tarbela Reservoir, a water resource for irrigation and hydroelectric power in Pakistan. Given the complexity of variations in the Himalayan glaciers and their socio-economic significance, it is important to understand the mechanisms that governed these changes in the historical period. More importantly, it is critical to accurately project expected future changes in the extent of Himalayan glaciers due to anthropogenic variations in temperature and precipitation patterns and impacts of such changes on water supply and agriculture. Projecting future changes to perennial water supply or flood risk in the Upper Indus River requires a modeling tool that can represent the effect of glacier and snow cover fluctuations. In this study, the Variable Infiltration Capacity (VIC) hydrology model is modified to better represent ice accumulation, ablation and transport in alpine glacier systems. Ice transport from the higher elevations to the lower elevations within a new glacial ice layer is represented via both ice deformation and basal sliding based on the Glen’s flow law. The transformation of snow to ice during the snow metamorphism process is implemented based on snow density changes within snowpack. The revised VIC model forced with daily precipitation and temperature data from the high resolution FVGCM-RegCM3 climate model for reference time period of 1961-1990, is evaluated at watershed scale using observed historical river discharge, glacier velocities and point observations of snow accumulation. Changes in streamflow seasonality in the Upper Indus Basin (UIB) are determined for the future climate period 2071-2100 relative to the reference period using daily precipitation and temperature data from RegCM3 climate model. The results from this analysis provide an enhanced understanding of the influence of glacier and snow cover variations on the magnitude and timing of total annual discharge of major rivers in the UIB.

Enhanced radiative forcing due to elevated greenhouse gas concentrations could cause an increase ... more Enhanced radiative forcing due to elevated greenhouse gas concentrations could cause an increase in evaporation and atmospheric water content, and consequently, a more intense global hydrological cycle. High resolution climate change simulations have projected more extreme surface temperatures throughout the United States and more extreme precipitation along the Gulf Coast, in the Pacific Northwest and east of the Mississippi. Such

ABSTRACT The effects of future climate change on the meteorological conditions associated with lo... more ABSTRACT The effects of future climate change on the meteorological conditions associated with low air quality events are poorly understood. Stagnant air masses are susceptible to air quality degradation due to ozone-producing photochemical alteration, particulate matter accumulation and recirculation, and temperature induced water vapor increases. In this study we apply the National Climatic Data Center's (NCDC) Air Stagnation Index to both National Center for Atmospheric Research/National Centers for Environmental Prediction reanalysis data and Coupled Model Intercomparison Project (CMIP3) ensemble climate predictions. Utilizing this air stagnation metric we gauge the ability of CMIP3 simulations to capture meteorological conditions observed in the 20th century reanalysis data and then assess the impact of increased greenhouse gas concentrations on future air stagnation event frequency. Our analysis deconstructs the NCDC stagnation index into its three principal components, 10 m winds, 500 mb winds, and precipitation events, in an effort to determine which meteorological factors play the prominent role in air stagnation event frequency within the recent past and into the future. We examine the applicability of the NCDC stagnation index to the global domain and analyze seasonal shifts in stagnation event frequency due to increased atmospheric greenhouse gas concentrations and associated feedbacks. A preliminary examination of reanalysis data suggests that the frequency of stagnation days is highest in southwestern North America, southeastern Asia, southern Africa, and within the tropical latitudes. These spatial patterns are the result of different regional NCDC stagnation index parameter dependencies; mid-latitude and subtropical areas with relatively high stagnation event frequencies tend to correspond to regions that rarely receive precipitation and often experience light low level winds, whereas tropical areas with relatively high stagnation event frequencies tend to occur in regions that often experience light upper and lower level winds, regardless of precipitation frequency.

Geophysical Research Letters, 2016

Late 20th century rising surface temperature trends and changing precipitation patterns have caus... more Late 20th century rising surface temperature trends and changing precipitation patterns have caused significant hydrological changes in the Western United States. Because as much as 75 percent of the water supply in the Western United States comes from snow melt, understanding how climatic changes may impact the snow cover and snow-driven streamflows in the next several decades is critical to the development of near-term mitigation policies. Using bias corrected daily-scale maximum temperature, minimum temperature and precipitation, and original surface winds from high-resolution multi-member ensemble simulations of a regional climate model (RegCM3), we drive a hydrological model (VIC) at 1/8th degree spatial and sub-daily temporal resolution to investigate near-term hydrological changes in the Western United States. The RegCM3 integration includes five ensemble members, which are driven by the NCAR Community Climate System Model (CCSM3) and cover the continental United States at 25 km horizontal spacing. Bias correction is applied to the monthly times-series of RegCM3 simulated data, and correction from individual months is subsequently translated to the RegCM3 simulated daily time-series. The simulations cover 1960-1999 in the historic period and 2000-2039 (A1B) in the future period. Analyses of these multi-member ensemble hydrological simulations are focused on the projection of the range of potential hydrological changes, particularly those associated with snow cover and streamflow, at multi-decadal time scales. In addition, using high spatial and temporal resolutions, we investigate the response of fine-scale hydrological feedbacks to net precipitation and temperature changes, as well as their daily extremes.

The early Holocene drying of the North American mid-continent is an important case study for unde... more The early Holocene drying of the North American mid-continent is an important case study for understanding regional aridity, its drivers, and local-scale responses. This paper synthesizes eolian, fossil pollen, δ13C, lake-level, and other paleohydrological proxies to document the spatial and temporal patterns in the onset and rate of drying. Although most sites show first signs of drying between 10 and 8 ka, some sites begin to dry as early at 14ka and others as late as 6ka. There is a significant time-transgressive trend, with western sites beginning to dry before eastern sites. However, the mean squared error for this regression is 1.58ka, indicating that the timing of local responses can depart considerably from the regional trend. Reconstructions of the Holocene dynamics of the eastern prairie-forest ecotone agree with prior maps, showing eastward prairie advance between 10,000 and 8,000 calendar years before present (ka), a maximum eastward position of the ecotone from 7 to 6ka, and westward prairie retreat after 6ka. However, the reconstructed rate of early Holocene deforestation is more rapid than previous estimates. Sites vary in the rate of response, with approximately one-third of sites showing rapid (transitional periods lasting less than 300 years) responses to early Holocene drying. A cluster of rapid responses at ca. 8ka may be caused by accelerated rates of regional drying, in turn forced by the collapse of the Laurentide Ice Sheet and drainage of Lake Agassiz. Other rapid responses likely represent non-linear site responses to progressive drying. The 21st-century trajectory for the Great Plains is uncertain, because climate models differ over the direction of regional precipitation trends, but future drying likely would trigger threshold-type shifts in ecotone position. Local responses to future regional drying trends may vary widely in timing and rate, challenging detailed impact assessments.

The influence of land surface changes upon global and regional climate has been shown both for an... more The influence of land surface changes upon global and regional climate has been shown both for anthropogenic and non-anthropogenic changes in land surface distribution. Because validation of global climate models (GCMs) is dependent upon the use of accurate boundary conditions, and because changes in land surface distribution have been shown to have effects on climate in areas remote from those changes, we have tested the sensitivity of a GCM to a global Mid Holocene vegetation distribution reconstructed from the fossil record, a first for a 6 ka GCM run. Large areas of the globe exhibit statistically significant seasonal warming of 2 to 4 ° C, with peak warming of 10 ° C over the Middle East in June-July-August (JJA). The patterns of maximum warming over both Northern Asia and the Middle East strongly coincide with the patterns of maximum decrease in albedo in all seasons. Likewise, cooling of up to 4 ° C over Northern Africa associated with the expansion of savanna and broadleaf evergreen forest also coincides with increases in surface heat flux of up to 35 W/m2 in March-April-May (MAM) and 60 W/m2 in JJA. At both the regional and global scale, the magnitude of vegetation forcing is equal to that of 6 ka orbital forcing, emphasizing the importance of accurate land surface distribution for both model validation and future climate prediction.

Agu Fall Meeting Abstracts, Dec 1, 2010

The simulation of observed 20th Century decadal variations in precipitation over West Africa are ... more The simulation of observed 20th Century decadal variations in precipitation over West Africa are improved with the use of observed sea surface temperatures as prescribed boundary conditions to atmospheric general circulation models, however inadequacies still remain in the simulation of the basic components of the West African Monsoon (WAM), particularly at seasonal and interannual time-scales. A single-model ensemble approach is applied with a regional climate model RegCM3 to quantify the sensitivity of the West African Monsoon (WAM) to initial and boundary conditions, as well as the treatment of physical processes including convective and land surface transfer schemes over the time period of 1980-2008. The three convective schemes used for comparison include MIT-Emanuel, Kuo, and Grell with both the Arakawa and Schubert, and Fritsch and Chappell closure assumptions. RegCM3 is coupled to both the BATS and CLM3 land surface models. Initial and boundary conditions are derived from both NCEP-NCAR Reanalysis Project 2 data, and a five-member ensemble from the NCAR CCSM3 general circulation model. Variability of West African precipitation and surface air temperature distribution at interannual timescales, as well as the fundamental circulation features of the WAM, including the Africa Easterly Jet (AEJ), Tropical Easterly Jet (TEJ), low-level monsoon westerlies, and easterly waves shaping those distributions are examined. Initial results indicate the interannual variability of precipitation and surface temperature over the West African domain is predominantly sensitive to the representation of convective processes within the regional model, which have a substantial influence on the atmospheric dynamics associated with the monsoon. In particular, the influence of convective and land-atmosphere transfer scheme on the modulation of easterly waves is explored. The results of the study help to refine our understanding of the complex interactions between land-atmosphere fluxes and convective initiation driving precipitation variability over West Africa at seasonal and interannual scales, a necessary element for improving short term and decadal climate prediction.

The mid-continent of North America has long been recognized to have experienced a period of prono... more The mid-continent of North America has long been recognized to have experienced a period of pronounced aridity during the mid-Holocene. A mechanistic explanation for these drier-than-modern conditions, however, has been elusive. Past GCM simulations of 6000 yrs B.P. have failed to produce the strong moisture anomalies inferred from pollen, lake level, diatom, and isotope data, perhaps because important ocean- and land-surface feedbacks were not incorporated into model experiments and because of the coarse resolution of GCMs. Here, we present an updated synthesis of paleoclimatic data from the upper Midwest for comparison with new regional climate model simulations of 6000 yrs B.P. using RegCM2. We focus our analysis on changes in the seasonality of precipitation that may inform our understanding of the climate processes underlying the arid conditions. We use the modern analog technique to infer seasonal moisture levels from fossil pollen data and a simple hydrologic box model to evaluate the seasonal inputs to past lake levels. Patterns in pollen, lake-level, and isotope data support the idea that the dry conditions arose from lower than modern winter precipitation at 6000 yrs B.P.; but existing hypotheses about the dry conditions have often emphasized the contribution of greater than modern summer insolation for producing either high summer evaporative losses or severely reduced summer precipitation as also occurred during the 1930s Dust Bowl drought. Our new RegCM2 simulations appear consistent with our data synthesis. They show widespread negative moisture-balance anomalies in winter and spring, as well as large positive moisture-balance anomalies in summer. The consistency between the data and model imply a need to develop a new framework for understanding the mid-Holocene in the upper Midwest.

Agu Fall Meeting Abstracts, Dec 1, 2010

Despite the relevance for human decision-making, our ability to accurately predict regional clima... more Despite the relevance for human decision-making, our ability to accurately predict regional climate on decadal time scales is limited. Using a high-resolution, multi-decadal, multi-member ensemble climate model experiment, we seek to better understand the processes shaping the variations in the pattern and magnitude of warm-season climate anomalies projected for the continental U.S. over the near-term period, and to explore whether

A new continental-scale comparison of pollen-based climate reconstructions, lake-level data, and ... more A new continental-scale comparison of pollen-based climate reconstructions, lake-level data, and regional climate model output reveals significant changes in North American moisture gradients during the past 6000 years. New seasonal precipitation reconstructions from fossil pollen data span the entire continent and show seasonally-varying mid-Holocene minus Modern anomaly patterns. Long-known features, such as the intensification of the southwestern U.S. monsoon during the mid-Holocene, appear evident, as do additional seasonal shifts, such as the occurrence of severe winter drought in the northeast U.S. and Rocky Mountains. The classic eastward shift of the prairie in the north-central U.S. during the mid-Holocene appears to coincide with little change in summer precipitation, but with a 25% decrease in spring and fall precipitation relative to today. Lake-level data confirm the dry conditions in the northeast U.S., mid-continent, and Rocky Mountains. Regional climate-model simulations for a domain centered on the U.S. using RegCM3, show strong similarities to the observed hydroclimatic anomalies. In a series of experiments focusing on the effects of insolation and oceanic conditions at 6000 years BP, the model consistently simulates low total soil moisture in the Rockies and mid-continent, and high total soil moisture in southwest U.S. Overall, the effects of insolation change explain many of the observed anomalies. Only in the northeast U.S. and western Great Plains does data- model agreement depend strongly on additional ocean variability. The inclusion of interannual modern-like ocean variability with the effects of insolation produces the best data-model comparison, and indicates that annual-to-decadal climatic variability may help shape orbitally-forced climatic changes.

Agu Fall Meeting Abstracts, Dec 1, 2009

The persistence of extended drought events throughout West Africa during the 20th Century has mot... more The persistence of extended drought events throughout West Africa during the 20th Century has motivated a significant effort to understand the mechanisms driving African climate variability. This study uses the Community Climate System Model version 3 (CCSM3) to examine the relative roles of future changes in GHG forcing and the associated SST feedback in shaping potential future climate change over West Africa. We are particularly interested in determining the cause of simulated changes in summer precipitation. To analyze the dynamical response to GHG forcing and SST feedbacks, we conduct twenty-year time-slice experiments (1980-1999 and 2080-2099) using CAM3 (the atmospheric component of CCSM3) and the CCSM3-generated SSTs. The atmospheric model realistically simulates both observed precipitation distribution and atmospheric circulation patterns. Changes in future atmospheric radiative forcing are found to be the primary driver of precipitation increases projected throughout the Sahel region (poleward of 12°N). In turn, the direct atmospheric forcing has a cooling effect throughout portions of the central Sahel. Dynamically, these changes in atmospheric forcing generate a slight northward displacement and weakening of the African easterly jet (AEJ) and a strengthening of the tropical easterly jet (TEJ). Alternatively, the future SST feedback is primarily responsible for the 21st century drying projected over much of the Guinean Coast (equatorward of 12°N). This decrease in precipitation is associated with both a weakening of the monsoon westerlies and the TEJ as well as a decrease in rising motion throughout the mid-levels of the troposphere. These findings confirm previous suggestions of a potential increased role of GHG forcing on future rainfall variability in West Africa, particularly in the Sahel, and a continuation of the highly influential role of SST variations seen throughout the 20th century.

Extreme climate events have been increasing over much of the world, and dynamical models predict ... more Extreme climate events have been increasing over much of the world, and dynamical models predict further increases in response to enhanced greenhouse forcing. We examine the ability of a high-resolution nested climate model, RegCM3, to capture extreme temperature and precipitation events over the conterminous United States, using high-quality observational and reanalysis data for comparison. Our analyses reveal that RegCM3 captures the pattern of mean, interannual variability, and trend in extreme temperature and precipitation events. However, consistent biases do exist, including wet biases in the topographically-complex regions of the western United States and hot biases in the southern and central United States. The biases in extreme precipitation in the western United States are associated with excessively strong surface and low- level winds. The biases in extreme temperature and precipitation in the southcentral United States are at least partially driven by biases in circulation and moisture fields in RegCM3. Our evaluation should enable more informed application and improvement of high-resolution climate models for the study of future changes in socially- and economically-relevant extreme events.

Our multi-ensemble member experiments using the NCAR CCSM3 over the period 1951-2099 indicate tha... more Our multi-ensemble member experiments using the NCAR CCSM3 over the period 1951-2099 indicate that global warming results in weakening of South Asian summer monsoon by weakening of meridional tropospheric temperature gradient, easterly-shear and upper-and-lower level monsoon circulation. Although similar anomalies in summer monsoon dynamics suppress summer monsoon precipitation at present, simulated weakening in South Asian summer monsoon dynamics in response to increases in greenhouse forcing does not lead to suppression of summer monsoon precipitation. In order to identify the mechanisms and/or biases responsible for such opposite responses of summer monsoon dynamics and summer monsoon precipitation, we have conducted two sets of multi-ensemble sensitivity experiments forced with anthropogenic increases in GHG+SST, SST-only and GHG-only. We find that SST feedbacks dominate the response of precipitation in the late 21st century, with changes in GHG forcing having little effect when applied without changes in SST. However, we find that CCSM SST biases create substantial biases in simulated summer precipitation. Thus, we are conducting a third set of experiments in which we prescribe bias-correct SSTs in the late 20th and 21st centuries, using a quantile-mapping technique. The SST-bias- correction experiments will help to understand both the relative role of changes in GHG and SST in dictating the response of South Asian summer monsoon to global warming, and the influence of SST biases on the simulated future climate change.

Environmental Research Letters, 2016

Global change biology, Jan 28, 2016

Understanding tropical rainforest carbon exchange and its response to heat and drought is critica... more Understanding tropical rainforest carbon exchange and its response to heat and drought is critical for quantifying the effects of climate change on tropical ecosystems, including global climate-carbon feedbacks. Of particular importance for the global carbon budget is net biome exchange of CO2 with the atmosphere (NBE), which represents nonfire carbon fluxes into and out of biomass and soils. Subannual and sub-Basin Amazon NBE estimates have relied heavily on process-based biosphere models, despite lack of model agreement with plot-scale observations. We present a new analysis of airborne measurements that reveals monthly, regional-scale (~1-8 × 10(6) km(2) ) NBE variations. We develop a regional atmospheric CO2 inversion that provides the first analysis of geographic and temporal variability in Amazon biosphere-atmosphere carbon exchange and that is minimally influenced by biosphere model-based first guesses of seasonal and annual mean fluxes. We find little evidence for a clear s...

The 4th IPCC report concludes that climate change is now unequivocal and projected increases in e... more The 4th IPCC report concludes that climate change is now unequivocal and projected increases in evaporation and atmospheric water content could intensify the hydrological cycle. However, the coarse spatial resolution and biases in global climate model simulations limit their usefulness in climate impact assessment. In order to improve on these limitations, we used high-resolution regional climate model (RegCM3) simulations, covering

Agu Fall Meeting Abstracts, Dec 1, 2002

Anthropogenic increases in atmospheric carbon dioxide (CO2) concentrations may affect vegetation ... more Anthropogenic increases in atmospheric carbon dioxide (CO2) concentrations may affect vegetation distribution both directly through changes in photosynthesis and water-use efficiency, and indirectly through CO_{2} induced climate change. Additionally, changes in vegetation distribution due to these direct and indirect effects may induce land surface-atmosphere feedbacks that create further change in both regional climate and regional vegetation distribution. Using a regional climate model (RegCM2.5) coupled to an equilibrium vegetation model (BIOME4), we quantitatively tested the sensitivity of climate and vegetation in the western United States to both the direct and indirect effects of doubled pre-Industrial atmospheric CO2 concentrations and to land surface-atmospheric feedbacks induced by the initial vegetation sensitivities. In assessing regional vegetation responses to the initial effects of elevated CO_{2} levels, vegetation in the western United States was sensitive to changes in photosynthesis and water use efficiency caused by increased CO2 availability, with woody biome types replacing less woody types throughout the domain. Vegetation was also sensitive to the initial climatic effects of increased CO_{2} concentrations, particularly at high elevations, both due to warming throughout the domain and to decreased precipitation in key mountain regions such as the Sierra Nevada and the Cascade and Blue Mountains of Oregon. Additionally, these patterns changed when the initial climatic and non-climatic effects of CO2 on vegetation were tested in combination, creating sensitivities not seen in either of the individual cases and indicating that climatic and non-climatic effects must be considered in tandem when assessing the potential impacts of elevated CO_{2} levels. Finally, asynchronous coupling of RegCM2.5 and BIOME4 tested the role of land surface-atmosphere feedbacks in shaping the regional response to elevated global atmospheric CO2 concentrations. The initial regional vegetation responses to elevated atmospheric CO_{2} levels resulted in regional land surface-atmosphere feedbacks that further altered regional climate and vegetation distribution, particularly along ecotones. Because the asynchronous coupling of RegCM2.5 and BIOME4 represents a new regional climate model application, further work is required to constrain the model sensitivity to choice in methodological variables such as length and number of iterations and domain size and placement.

ABSTRACT In recent years, accelerated glacier recession trends have been reported for the Himalay... more ABSTRACT In recent years, accelerated glacier recession trends have been reported for the Himalayan region, based largely on the studies in the Eastern Himalayas. However, recent studies carried out in the Karakoram Himalayas, suggest an expansion of glaciers and a reduction in summer streamflow due to a significant decrease in summer temperatures. Snow and ice melt from these glaciers is the primary input to the Upper Indus River, upstream of Tarbela Reservoir, a water resource for irrigation and hydroelectric power in Pakistan. Given the complexity of variations in the Himalayan glaciers and their socio-economic significance, it is important to understand the mechanisms that governed these changes in the historical period. More importantly, it is critical to accurately project expected future changes in the extent of Himalayan glaciers due to anthropogenic variations in temperature and precipitation patterns and impacts of such changes on water supply and agriculture. Projecting future changes to perennial water supply or flood risk in the Upper Indus River requires a modeling tool that can represent the effect of glacier and snow cover fluctuations. In this study, the Variable Infiltration Capacity (VIC) hydrology model is modified to better represent ice accumulation, ablation and transport in alpine glacier systems. Ice transport from the higher elevations to the lower elevations within a new glacial ice layer is represented via both ice deformation and basal sliding based on the Glen’s flow law. The transformation of snow to ice during the snow metamorphism process is implemented based on snow density changes within snowpack. The revised VIC model forced with daily precipitation and temperature data from the high resolution FVGCM-RegCM3 climate model for reference time period of 1961-1990, is evaluated at watershed scale using observed historical river discharge, glacier velocities and point observations of snow accumulation. Changes in streamflow seasonality in the Upper Indus Basin (UIB) are determined for the future climate period 2071-2100 relative to the reference period using daily precipitation and temperature data from RegCM3 climate model. The results from this analysis provide an enhanced understanding of the influence of glacier and snow cover variations on the magnitude and timing of total annual discharge of major rivers in the UIB.

Enhanced radiative forcing due to elevated greenhouse gas concentrations could cause an increase ... more Enhanced radiative forcing due to elevated greenhouse gas concentrations could cause an increase in evaporation and atmospheric water content, and consequently, a more intense global hydrological cycle. High resolution climate change simulations have projected more extreme surface temperatures throughout the United States and more extreme precipitation along the Gulf Coast, in the Pacific Northwest and east of the Mississippi. Such

ABSTRACT The effects of future climate change on the meteorological conditions associated with lo... more ABSTRACT The effects of future climate change on the meteorological conditions associated with low air quality events are poorly understood. Stagnant air masses are susceptible to air quality degradation due to ozone-producing photochemical alteration, particulate matter accumulation and recirculation, and temperature induced water vapor increases. In this study we apply the National Climatic Data Center's (NCDC) Air Stagnation Index to both National Center for Atmospheric Research/National Centers for Environmental Prediction reanalysis data and Coupled Model Intercomparison Project (CMIP3) ensemble climate predictions. Utilizing this air stagnation metric we gauge the ability of CMIP3 simulations to capture meteorological conditions observed in the 20th century reanalysis data and then assess the impact of increased greenhouse gas concentrations on future air stagnation event frequency. Our analysis deconstructs the NCDC stagnation index into its three principal components, 10 m winds, 500 mb winds, and precipitation events, in an effort to determine which meteorological factors play the prominent role in air stagnation event frequency within the recent past and into the future. We examine the applicability of the NCDC stagnation index to the global domain and analyze seasonal shifts in stagnation event frequency due to increased atmospheric greenhouse gas concentrations and associated feedbacks. A preliminary examination of reanalysis data suggests that the frequency of stagnation days is highest in southwestern North America, southeastern Asia, southern Africa, and within the tropical latitudes. These spatial patterns are the result of different regional NCDC stagnation index parameter dependencies; mid-latitude and subtropical areas with relatively high stagnation event frequencies tend to correspond to regions that rarely receive precipitation and often experience light low level winds, whereas tropical areas with relatively high stagnation event frequencies tend to occur in regions that often experience light upper and lower level winds, regardless of precipitation frequency.

Geophysical Research Letters, 2016

Late 20th century rising surface temperature trends and changing precipitation patterns have caus... more Late 20th century rising surface temperature trends and changing precipitation patterns have caused significant hydrological changes in the Western United States. Because as much as 75 percent of the water supply in the Western United States comes from snow melt, understanding how climatic changes may impact the snow cover and snow-driven streamflows in the next several decades is critical to the development of near-term mitigation policies. Using bias corrected daily-scale maximum temperature, minimum temperature and precipitation, and original surface winds from high-resolution multi-member ensemble simulations of a regional climate model (RegCM3), we drive a hydrological model (VIC) at 1/8th degree spatial and sub-daily temporal resolution to investigate near-term hydrological changes in the Western United States. The RegCM3 integration includes five ensemble members, which are driven by the NCAR Community Climate System Model (CCSM3) and cover the continental United States at 25 km horizontal spacing. Bias correction is applied to the monthly times-series of RegCM3 simulated data, and correction from individual months is subsequently translated to the RegCM3 simulated daily time-series. The simulations cover 1960-1999 in the historic period and 2000-2039 (A1B) in the future period. Analyses of these multi-member ensemble hydrological simulations are focused on the projection of the range of potential hydrological changes, particularly those associated with snow cover and streamflow, at multi-decadal time scales. In addition, using high spatial and temporal resolutions, we investigate the response of fine-scale hydrological feedbacks to net precipitation and temperature changes, as well as their daily extremes.

The early Holocene drying of the North American mid-continent is an important case study for unde... more The early Holocene drying of the North American mid-continent is an important case study for understanding regional aridity, its drivers, and local-scale responses. This paper synthesizes eolian, fossil pollen, δ13C, lake-level, and other paleohydrological proxies to document the spatial and temporal patterns in the onset and rate of drying. Although most sites show first signs of drying between 10 and 8 ka, some sites begin to dry as early at 14ka and others as late as 6ka. There is a significant time-transgressive trend, with western sites beginning to dry before eastern sites. However, the mean squared error for this regression is 1.58ka, indicating that the timing of local responses can depart considerably from the regional trend. Reconstructions of the Holocene dynamics of the eastern prairie-forest ecotone agree with prior maps, showing eastward prairie advance between 10,000 and 8,000 calendar years before present (ka), a maximum eastward position of the ecotone from 7 to 6ka, and westward prairie retreat after 6ka. However, the reconstructed rate of early Holocene deforestation is more rapid than previous estimates. Sites vary in the rate of response, with approximately one-third of sites showing rapid (transitional periods lasting less than 300 years) responses to early Holocene drying. A cluster of rapid responses at ca. 8ka may be caused by accelerated rates of regional drying, in turn forced by the collapse of the Laurentide Ice Sheet and drainage of Lake Agassiz. Other rapid responses likely represent non-linear site responses to progressive drying. The 21st-century trajectory for the Great Plains is uncertain, because climate models differ over the direction of regional precipitation trends, but future drying likely would trigger threshold-type shifts in ecotone position. Local responses to future regional drying trends may vary widely in timing and rate, challenging detailed impact assessments.

The influence of land surface changes upon global and regional climate has been shown both for an... more The influence of land surface changes upon global and regional climate has been shown both for anthropogenic and non-anthropogenic changes in land surface distribution. Because validation of global climate models (GCMs) is dependent upon the use of accurate boundary conditions, and because changes in land surface distribution have been shown to have effects on climate in areas remote from those changes, we have tested the sensitivity of a GCM to a global Mid Holocene vegetation distribution reconstructed from the fossil record, a first for a 6 ka GCM run. Large areas of the globe exhibit statistically significant seasonal warming of 2 to 4 ° C, with peak warming of 10 ° C over the Middle East in June-July-August (JJA). The patterns of maximum warming over both Northern Asia and the Middle East strongly coincide with the patterns of maximum decrease in albedo in all seasons. Likewise, cooling of up to 4 ° C over Northern Africa associated with the expansion of savanna and broadleaf evergreen forest also coincides with increases in surface heat flux of up to 35 W/m2 in March-April-May (MAM) and 60 W/m2 in JJA. At both the regional and global scale, the magnitude of vegetation forcing is equal to that of 6 ka orbital forcing, emphasizing the importance of accurate land surface distribution for both model validation and future climate prediction.

Agu Fall Meeting Abstracts, Dec 1, 2010

The simulation of observed 20th Century decadal variations in precipitation over West Africa are ... more The simulation of observed 20th Century decadal variations in precipitation over West Africa are improved with the use of observed sea surface temperatures as prescribed boundary conditions to atmospheric general circulation models, however inadequacies still remain in the simulation of the basic components of the West African Monsoon (WAM), particularly at seasonal and interannual time-scales. A single-model ensemble approach is applied with a regional climate model RegCM3 to quantify the sensitivity of the West African Monsoon (WAM) to initial and boundary conditions, as well as the treatment of physical processes including convective and land surface transfer schemes over the time period of 1980-2008. The three convective schemes used for comparison include MIT-Emanuel, Kuo, and Grell with both the Arakawa and Schubert, and Fritsch and Chappell closure assumptions. RegCM3 is coupled to both the BATS and CLM3 land surface models. Initial and boundary conditions are derived from both NCEP-NCAR Reanalysis Project 2 data, and a five-member ensemble from the NCAR CCSM3 general circulation model. Variability of West African precipitation and surface air temperature distribution at interannual timescales, as well as the fundamental circulation features of the WAM, including the Africa Easterly Jet (AEJ), Tropical Easterly Jet (TEJ), low-level monsoon westerlies, and easterly waves shaping those distributions are examined. Initial results indicate the interannual variability of precipitation and surface temperature over the West African domain is predominantly sensitive to the representation of convective processes within the regional model, which have a substantial influence on the atmospheric dynamics associated with the monsoon. In particular, the influence of convective and land-atmosphere transfer scheme on the modulation of easterly waves is explored. The results of the study help to refine our understanding of the complex interactions between land-atmosphere fluxes and convective initiation driving precipitation variability over West Africa at seasonal and interannual scales, a necessary element for improving short term and decadal climate prediction.

The mid-continent of North America has long been recognized to have experienced a period of prono... more The mid-continent of North America has long been recognized to have experienced a period of pronounced aridity during the mid-Holocene. A mechanistic explanation for these drier-than-modern conditions, however, has been elusive. Past GCM simulations of 6000 yrs B.P. have failed to produce the strong moisture anomalies inferred from pollen, lake level, diatom, and isotope data, perhaps because important ocean- and land-surface feedbacks were not incorporated into model experiments and because of the coarse resolution of GCMs. Here, we present an updated synthesis of paleoclimatic data from the upper Midwest for comparison with new regional climate model simulations of 6000 yrs B.P. using RegCM2. We focus our analysis on changes in the seasonality of precipitation that may inform our understanding of the climate processes underlying the arid conditions. We use the modern analog technique to infer seasonal moisture levels from fossil pollen data and a simple hydrologic box model to evaluate the seasonal inputs to past lake levels. Patterns in pollen, lake-level, and isotope data support the idea that the dry conditions arose from lower than modern winter precipitation at 6000 yrs B.P.; but existing hypotheses about the dry conditions have often emphasized the contribution of greater than modern summer insolation for producing either high summer evaporative losses or severely reduced summer precipitation as also occurred during the 1930s Dust Bowl drought. Our new RegCM2 simulations appear consistent with our data synthesis. They show widespread negative moisture-balance anomalies in winter and spring, as well as large positive moisture-balance anomalies in summer. The consistency between the data and model imply a need to develop a new framework for understanding the mid-Holocene in the upper Midwest.

Agu Fall Meeting Abstracts, Dec 1, 2010

Despite the relevance for human decision-making, our ability to accurately predict regional clima... more Despite the relevance for human decision-making, our ability to accurately predict regional climate on decadal time scales is limited. Using a high-resolution, multi-decadal, multi-member ensemble climate model experiment, we seek to better understand the processes shaping the variations in the pattern and magnitude of warm-season climate anomalies projected for the continental U.S. over the near-term period, and to explore whether

A new continental-scale comparison of pollen-based climate reconstructions, lake-level data, and ... more A new continental-scale comparison of pollen-based climate reconstructions, lake-level data, and regional climate model output reveals significant changes in North American moisture gradients during the past 6000 years. New seasonal precipitation reconstructions from fossil pollen data span the entire continent and show seasonally-varying mid-Holocene minus Modern anomaly patterns. Long-known features, such as the intensification of the southwestern U.S. monsoon during the mid-Holocene, appear evident, as do additional seasonal shifts, such as the occurrence of severe winter drought in the northeast U.S. and Rocky Mountains. The classic eastward shift of the prairie in the north-central U.S. during the mid-Holocene appears to coincide with little change in summer precipitation, but with a 25% decrease in spring and fall precipitation relative to today. Lake-level data confirm the dry conditions in the northeast U.S., mid-continent, and Rocky Mountains. Regional climate-model simulations for a domain centered on the U.S. using RegCM3, show strong similarities to the observed hydroclimatic anomalies. In a series of experiments focusing on the effects of insolation and oceanic conditions at 6000 years BP, the model consistently simulates low total soil moisture in the Rockies and mid-continent, and high total soil moisture in southwest U.S. Overall, the effects of insolation change explain many of the observed anomalies. Only in the northeast U.S. and western Great Plains does data- model agreement depend strongly on additional ocean variability. The inclusion of interannual modern-like ocean variability with the effects of insolation produces the best data-model comparison, and indicates that annual-to-decadal climatic variability may help shape orbitally-forced climatic changes.

Agu Fall Meeting Abstracts, Dec 1, 2009

The persistence of extended drought events throughout West Africa during the 20th Century has mot... more The persistence of extended drought events throughout West Africa during the 20th Century has motivated a significant effort to understand the mechanisms driving African climate variability. This study uses the Community Climate System Model version 3 (CCSM3) to examine the relative roles of future changes in GHG forcing and the associated SST feedback in shaping potential future climate change over West Africa. We are particularly interested in determining the cause of simulated changes in summer precipitation. To analyze the dynamical response to GHG forcing and SST feedbacks, we conduct twenty-year time-slice experiments (1980-1999 and 2080-2099) using CAM3 (the atmospheric component of CCSM3) and the CCSM3-generated SSTs. The atmospheric model realistically simulates both observed precipitation distribution and atmospheric circulation patterns. Changes in future atmospheric radiative forcing are found to be the primary driver of precipitation increases projected throughout the Sahel region (poleward of 12°N). In turn, the direct atmospheric forcing has a cooling effect throughout portions of the central Sahel. Dynamically, these changes in atmospheric forcing generate a slight northward displacement and weakening of the African easterly jet (AEJ) and a strengthening of the tropical easterly jet (TEJ). Alternatively, the future SST feedback is primarily responsible for the 21st century drying projected over much of the Guinean Coast (equatorward of 12°N). This decrease in precipitation is associated with both a weakening of the monsoon westerlies and the TEJ as well as a decrease in rising motion throughout the mid-levels of the troposphere. These findings confirm previous suggestions of a potential increased role of GHG forcing on future rainfall variability in West Africa, particularly in the Sahel, and a continuation of the highly influential role of SST variations seen throughout the 20th century.

Extreme climate events have been increasing over much of the world, and dynamical models predict ... more Extreme climate events have been increasing over much of the world, and dynamical models predict further increases in response to enhanced greenhouse forcing. We examine the ability of a high-resolution nested climate model, RegCM3, to capture extreme temperature and precipitation events over the conterminous United States, using high-quality observational and reanalysis data for comparison. Our analyses reveal that RegCM3 captures the pattern of mean, interannual variability, and trend in extreme temperature and precipitation events. However, consistent biases do exist, including wet biases in the topographically-complex regions of the western United States and hot biases in the southern and central United States. The biases in extreme precipitation in the western United States are associated with excessively strong surface and low- level winds. The biases in extreme temperature and precipitation in the southcentral United States are at least partially driven by biases in circulation and moisture fields in RegCM3. Our evaluation should enable more informed application and improvement of high-resolution climate models for the study of future changes in socially- and economically-relevant extreme events.

Our multi-ensemble member experiments using the NCAR CCSM3 over the period 1951-2099 indicate tha... more Our multi-ensemble member experiments using the NCAR CCSM3 over the period 1951-2099 indicate that global warming results in weakening of South Asian summer monsoon by weakening of meridional tropospheric temperature gradient, easterly-shear and upper-and-lower level monsoon circulation. Although similar anomalies in summer monsoon dynamics suppress summer monsoon precipitation at present, simulated weakening in South Asian summer monsoon dynamics in response to increases in greenhouse forcing does not lead to suppression of summer monsoon precipitation. In order to identify the mechanisms and/or biases responsible for such opposite responses of summer monsoon dynamics and summer monsoon precipitation, we have conducted two sets of multi-ensemble sensitivity experiments forced with anthropogenic increases in GHG+SST, SST-only and GHG-only. We find that SST feedbacks dominate the response of precipitation in the late 21st century, with changes in GHG forcing having little effect when applied without changes in SST. However, we find that CCSM SST biases create substantial biases in simulated summer precipitation. Thus, we are conducting a third set of experiments in which we prescribe bias-correct SSTs in the late 20th and 21st centuries, using a quantile-mapping technique. The SST-bias- correction experiments will help to understand both the relative role of changes in GHG and SST in dictating the response of South Asian summer monsoon to global warming, and the influence of SST biases on the simulated future climate change.