GLOBAL WATER CYCLE: EXTENSION ACROSS THE EARTH SCIENCES Progress Report on the Work Plan NASA Earth Observing System NAGW-2686 November, 1993 The Pennsylvania State University EARTH SYSTEM SCIENCE CENTER and MARSHALL SPACEFLIGHT CENTER INTRODUCTION This document describes the progress of our research under NASA Grant NAGW- 2686, "Global Water Cycle : Extension Across the Earth Sciences" during the period December, 1992 - November, 1993. The "Near Term Objectives" of the original Work Plan (p. 67-87) and the key tasks listed in the "Near Term Research Priorities, Supplement to the Work Plan for Confirmation" (February, 1991) describe FY91, FY92, and FY93 research plans. We have keyed the sections in this document to the detailed research objectives listed below. The Susquehanna River Basin Experiment focuses four of our original objectives in a regional experiment. Progress on this element of our research program is reported in a combined section. OBJECTIVES Objective 1. Evaluation of GCM capability to simulate hydrologic cycle components Objective 2. Development of standardized model products Objective 3. Evaluate GCM sensitivity to lower boundary forcing Objective 4. Generation of global data sets for documentation of global change, specifically for hydrologic variables, and for global model validation and verification Objective 5. Development and testing of components of an atmospheric model of the regional hydrologic cycle Objective 6. Testing the adaptability of hydrologic models for sensitivity to spatial and topographic scales Objective 7. Test methodologies for indirect measurement of soil water content and integrate results into mesoscale model simulations Objective 8. Development of a comprehensive database and GIS for the Susquehanna River Basin to facilitate the development and evaluation of the nested model approach to regional prediction Objective 9. Development of a comprehensive database and GIS for the Cape Region in E. Central Florida in order to better understand process coupling between the land and atmosphere in a subtropical summer environment. Objective 10. Determination of ice sheet mass balance Objective 11. Development of a landscape evolution model Objective 12. Climate - Agriculture Modeling RESEARCH ELEMENTS Global Circulation Modeling (Objectives 1, 2, 3, 4) (E. Barron, R. Crane, F. Robertson, J. Christy, and D. Fitzjarrald) GENESIS The archived results of a ten-year integration with an R15 version of GENESIS have served as a very interesting database in evaluating GENESIS capabilities. Results were presented at the Annual Meetings of the American Meteorological Society in January (Robertson, et al., 1993) and at the first GENESIS Workshop at NCAR in September. At R15 GENESIS has some difficulty in reproducing tropical precipitation structure. For some purposes (zonal averages) this is not a problem. However, with regard to interannual variability in the Indian Monsoon, or the configuration of mid-latitude planetary waves which are to some extent forced by tropical divergence, this resolution is important. This is in part the motivation for porting the GENESIS model to the MSFC XMP system so that additional resources can be used in making higher resolution runs (see below). We are anticipating moving to T31 resolution as a standard mesh size. Our analysis has also indicated that high cloudiness in the model is somewhat deficient over tropical convective areas. Although one could adjust the cloud diagnostic parameters in the model, much more understanding could be gained by formulating a predictive equation for condensate that makes the physics behind the water mass and radiative energy balance more realistic. We have obtained the convective code from GENESIS and are beginning to look at how this might be done. Preparations have been made to run the GENESIS model on the MSFC SSL Cray. With diligent work by Bill Peterson at ESSC and Jayanthi Srikishen at MSFC, all the files and the makefiles are in place and procedures for archiving and analyzing the data are being developed. We are currently waiting for the MSFC system administrator to reconfigure the XMP file structure to enable it to use long Unix file names, so that the file and directory structure matches the ESSC Cray YMP. As soon as the reconfiguraton is accomplished the model can be integrated. Data transfer over slow networks has been a significant constraint in working on the ESSC Cray from MSFC. The ability to run the model and analyze the results in-house will enable more efficient studies of interannual variability of moisture processes, since these studies will entail the generation, transfer, and analysis of large quantities of data by the model. (Note: Please refer to Susquehanna River Basin Experiment section for a complete listing of experiments completed) GENESIS Sensitivity to SST Anomalies In tandem with the 10-year baseline integrations of GENESIS using a slab ocean model, integrations were made using fixed annual cycle sea surface temperatures (SST's) and Atmospheric Modeling Intercomparison Project (AMIP) CAC observed SST's as lower boundary forcing. An analysis of the precipitation variance amplitude shows that the model forced with the observed SST's has considerably more variability than does the climatological varying SST's. Tropical response to this forcing is fairly robust. 1983 minus 1982 precipitation in the GENESIS model and from MSU show quite good agreement (evidence of the strong ENSO forcing). In middle latitudes over North America and over Asia, the model response in terms of precipitation anomalies, hydrology and tropospheric mean temperatures matches poorly with observations. Nevertheless, some events seem to be reproduced qualitatively, most notable among these being the 1988 drought over North America which coincided with anomalously cool equatorial eastern Pacific SST's. Motivated by questions of intermodel variability and signal to noise ratio, we have begun experiments with the GSFC/GLA 17-layer, 4 by 5 degree gridpoint climate model. This model has evolved over many years from the UCLA gridpoint model and incorporates a number of improvements in the surface, moisture, and cumulus parameterizations that are appropriate to this EOS study. As a start, we have begun to repeat the AMIP prescribed SST integration (years 1979 to 1988) using an improved parameterization of rainfall evaporation. With the AMIP decade-long run as control, we have commenced a series of experiments to investigate the influence of end-of-season snow cover on the subsequent development of spring and summer precipitation patterns. For example, were the midwest floods of 1993 due to a greater than normal snowpack in the west that preserved late spring soil moisture? The first few years of this integration suggest that indeed the resulting changes in the surface energy balance due to altered Bowen ratio can impact equivalent barotropic atmospheric thermal structure and, hence, storm track position. It is important in this study to do enough cases to get an idea of the inherent chaotic variability of the model simulation and the sensitivity of this variation to changes in snow cover. To this end, we will use the 10 spring seasons of the AMIP control run as different realizations to apply the enhanced snow cover, then compare the subsequent precipitation patterns to each unaltered control run. Initial results indicate that the snow cover forcing is adequate to significantly alter subsequent precipitation, and that the soil moisture field must have equilibrated before the experiment is started. The first 2 years of the AMIP control run appear to retain too much of the arbitrary soil moisture initial conditions. Global Data Sets for Earth System Change Documentation Having completed the initial study in early FY93 with Special Sensor for Microwave/Imaging (SSM/I) F8 retrievals of integrated liquid water and water vapor, we have devoted efforts this year to increasing the data base which now extends through mid- calendar year 1993. It was felt that more than the original three years was necessary in order to produce accurate statistics, even for intraseasonal variability (20-90 days). In FY94 we expect to complete statistical analysis on a 5-yr climatology which will document synoptic through interannual time scales. We have undertaken a pilot study to assess the feasibility of forcing moisture conservations with SSM/I data (F. Robertson). The objective of this work is to produce a climatology which is based heavily on remotely-sensed water vapor (SSM/I, High Resolution Infrared Radiation Sounder (HIRS-2), SSMT/2) and which uses additional constraints from global gridded analyses. A unique aspect of this study is a semi- prognostic approach whereby remotely-sensed moisture data are assimilated into an evolving analysis. Wind and temperature fields from global gridded analyses (e.g. GSFC, National Meteorological Center (NMC) or European Centre for Medium Range Forecasting (ECMWF) reanalyses) are used to drive predictive equations for water vapor, condensate, and precipitation. In addition to transport processes, parameterized bulk microphysics and moist convection affect the distribution of water substance. The incorporation of remotely- sensed water vapor is accomplished by a nudging procedure which updates the evolving water vapor field and constrains it to observations. This semi-prognostic approach differs from four dimensional data assimilation (FDDA) in several ways: (1) only these moisture fields are prognosed, (2) it is much less computationally expensive, so many experiments with differing moist physics can be done. In fact it can serve as a testbed for future FDDA model convective and cloud parameterizations and (3) current reanalysis efforts do not include explicit simulation of cloud water or ice. This year we have focused on testing of convective parameterizations that will allow us to diagnose the effect that closure assumptions have on estimates of mass flux. This strongly affects the production of upper-tropospheric condensate. We have been diagnosing the TOGA-COARE period of Nov. 1992-Jan. 1993 (Robertson and McCaul, 1993) and have been able to reproduce the observed upper-level cloudiness field which has such a strong effect on cloud radiative forcing in the western Pacific. In 1994 we will be folding in MSU precipitation estimates, Tim Liu's oceanic evaporation estimates, and assessing the differences in analysis results arising from differences in the ECMWF and GSFC analyses. We expect the results of this effort ( a global analysis of vapor, precipitation and cloud condensate) to be very useful in two ways: First, it will help in interpreting moisture relationships to kinematic, bulk microphysical and convective processes. This will facilitate comparison of the International Satellite Cloud Climatology Program (ISCCP) cloud climatology and Earth Radiation Budget (ERB) cloud radiative forcing estimates to global distributions of water vapor. Second, it will assist in improving parameterizations for convection and cloudiness that could be used in the GSFC EROS assimilation model. Work on the analysis of heating rates from global gridded analyses, partitioned into latent heat release, radiative flux convergence, sensible heat flux and eddy heat transports has focused, in the last half of FY93, on gearing up codes to use the GSFC reanalysis products, now estimated to be available in Dec. 1993. We have coordinated with Sig Schubert and Richy Rood at GSFC regarding current strengths and issues with the analysis system and anticipate diagnostic studies to resume in 1994. In addition we have explored with Prashant Sardeshmukh, NOAA ERL CRD, the possibility of comparing his divergent circulation estimates, derived from vorticity balance considerations, to the divergent circulation in the GSFC assimilated products. Because these two methods are largely independent of each other, an intercomparison will strengthen understanding of the uncertainties in diabatic heating estimates and the moisture diagnostics noted in this task. Research in 1993 by Christy continued to focus on the determination of lower stratospheric temperature. A paper (Spencer and Christy, 1993) documents the high precision of channel 4 in the Microwave Sounding Unit (MSU) data. In another paper, Christy and Drouilhet (1993) we have added more information on the precision issue and have studied the daily variability of the zonal mean anomalies. We found daily zonal mean anomalies of lower stratospheric temperature are known to within 0.01 deg C near the equator and within 0.07 near the poles. Signal to noise ratios are generally above 100 and approach 1000 near the equator and poles. By calculating the meridional curvature (second derivative) near the equator, an index for the QBO was determined. However, the typical relationships reported by others for QBO versus polar anomalies was not found in the MSU data. Ozone concentrations during October over the South Polar region were will correlated with MSU 4 prior to the effects of Mt. Pinatubo in 1991. The global lower tropospheric temperature trend since 1979 shows the atmosphere cooling at a 0.03 deg C per decade rate. Much of this can be attributed to the cooling due to Pinatubo. The effects of Pinatubo, El Chichon and El Ninos have been calculated in the lower troposphere. Once removed they show a decadal trend of +0.09 deg C. This may be the warming, however slight, due to enhanced greenhouse gases. We continued to work in 1993 on the creation of an optimal climatological precipitation and streamflow data set. The Navy Cal/Val operational SSM/I rainfall algorithm was submitted for the WETNET PIP-1 precipitation intercomparison project to assess a possible baseline global rainfall scheme. It was learned that the surface calibration data set used to develop this algorithm was inadequate. Comparison with the GPCP rainfall climatology, ECMWF model, and other independent "truth" sets show that the operational algorithm underestimates instantaneous rainrate by a factor of two. The ECMWF 12-h forecast is calibrated with GPCP raingage data, and therefore agreed best with the GPCP climatology. Further studies are planned as part of the PIP-2 case study experiments. Synoptic-Scale Validation The methodologies have been developed for synoptic-scale validation through translation across scales, from the synoptic-scale circulation to the regional climate. The focus for this work in 1993 has been on precipitation: tropical rainfall in southern Mexico and, in contrast, snowfall in the Colorado Rockies. The approach used is to apply neural nets to develop empirical relationships between the large-scale circulation and the regional precipitation. A simple feed-foward net architecture has proved very effective in both locations (Colorado and Mexico). Colorado Basin Snowfall The results of the analysis of snowfall in the Colorado Basin show that the neural net demonstrates that the synoptic circulation (as represented by the 700 mb height field) exerts a strong control on snowfall. The relationship between circulation and snowfall varies as a function of geographic region which reflects largely individual sub-drainage basins. The circulation accounts for approximately 70% of the day-to-day variability in snowfall. This relationship is highly non-linear. Traditional linear regression methods explain only 34% of the variance between circulation and snowfall (McGinnis, 1993). Mexican Precipitation Previous work in this part of the EOS research project found that neural nets could be used to relate the surface and 500 mb circulations to precipitation over Chiapas, southern Mexico. The neural net was able to predict the onset of the summer rainy season, as well as the phase of individual rain events. The present work extends these earlier results by de- composing the net and establishing the sensitivity of the rainfall to individual features of the 500 mb and surface pressure fields. Different features of the circulation are important for explaining rainfall variability during different parts of the season. The Inter-Trpocial Convergence Zone (ITCZ) controls rainfall during the onset of the rains and during their dissipation at the end of the season. In the middle of the season the rainfall variability is controlled in part by upper level divergence (easterly waves), and in part by the extension of the Bermuda High, which affects moisture supply to the region (Hewitson and Crane, 1993). Mesoscale Modeling (Objective 5) (T. Ackerman, B. Albrecht, D. Lamb, T. Warner) Improved Treatment of Cloud Processes in a Mesoscale Model Cirrus clouds are an important but poorly understood component of the climate system. Due to their large spatial extent and cold temperatures, cirrus clouds have a large impact on the planetary radiation budget by reducing outgoing longwave radiation and increasing the downward longwave flux at the surface. In addition, radiative heating in the cloud deck can be a significant factor. Despite the importance of cirrus, however, current cloud parameterization schemes used in general circulation models are clearly inadequate in their prediction of cirrus occurrence and diagnosis of cirrus radiative processes, while most mesoscale models contain no cirrus parameterizations at all. The objective of our current activity is to improve the treatment of cirrus formation and life cycle, and radiative impacts in the MM4 and MM5 versions of the Penn State University/National Center for Atmospheric Research mesoscale model. Our approach may be outlined briefly as follows: 1) Collect a data set of cirrus cloud characteristics (such as frequency of occurrence, height, ice water content, and optical properties), associated radiative fluxes, and environmental variables (such as temperature, humidity, and wind); 2) Use the data to produce diagnostic 4-D fields of the environmental variables; 3) Use a mesoscale model to simulate these same 4-D fields and compare them to simulated and observed fields; 4) Incorporate a cloud parameterization scheme into the mesoscale model and use it to predict the occurrence of cirrus cloud and associated radiative properties; 5) Compare the predicted and observed cirrus characteristics and computed and observed radiative fluxes. The data which we are using was collected during the First ISCCP Regional Experiment (FIRE) Cirrus II campaign in Coffeyville, Kansas, in 1991. The hub site of this experiment was located within the area covered by the National Weather Service (NWS) wind profiler demonstration array. The wind profilers measure hourly-averaged profiles of the horizontal wind speed and direction from about 1 km to the tropopause. In principle, the complete 4-D wind field can be diagnosed from these profilers by interpolating the observations to a uniform 3-D grid, calculating the horizontal divergence, and then applying continuity to obtain the vertical velocity. In practice, due to the spatial scale of the profilers and instrument errors, it proved to be difficult to obtain realistic results. However, during the past year, we have completed a diagnostic package that allows us to produce 4-D wind fields from the profilers, as well as the divergence and vorticity (Mace et al., 1993). In order to test the method for consistency, we have applied it to several case studies (Mace et al., 1993a,b). Producing 4-D thermodynamic fields (temperature and absolute humidity) for the experiment period is much more challenging because the only available data is from radiosonde releases. We have developed a technique to use the radiosonde-measured temperatures and humidities in conjunction with the 4-D wind field to produce 4-D thermodynamic fields. During the past year, we have also completed a month-long series of overlapping model runs for the FIRE II campaign using MM5, the non-hydrostatic version of the PSU/NCAR mesoscale model at 1 km resolution. The model was run from initial conditions for 48 hours with a 12 hour overlap at the beginning of each run and the end of the previous run. By averaging the two runs during the 12 hour overlap, a virtually seamless mesoscale model simulation for the entire experiment period has been produced. During the past year, we have also completed a study of the effect of non-spherical ice crystals on solar and infrared radiative fluxes. This work defines the needed precision of optical property measurements in ice clouds in order to achieve computed accuracies of 5 W/m2 or better in surface fluxes. The purpose is to determine how well cirrus cloud parameterizations need to specify cloud optical properties in order to achieve the requisite accuracy in flux computations. Over the course of the next year, we intend to publish the results of these diagnostic studies. (Both studies are nearing completion as Ph. D. dissertations.) The next step is to intercompare the diagnostic fields, the MM5 results, and an FDDA product from the NWS. From this intercomparison, we will be able to ascertain whether the model captures the thermodynamics and dynamics necessary for cirrus formation. If the answer is positive, we will incorporate a cloud parameterization scheme and begin evaluation and modification of that scheme. Regional Modeling of Boundary Layer Clouds Collaboration with Shouping Wang at NASA Huntsville on the development and improvement of parameterizations of boundary layer clouds has continued during this funding period. This collaboration resulted in the development of a regional boundary layer model that simulates conditions associated with both fair-weather cumulus and stratocumulus and predicts inversion height and the thermodynamic structure below the inversion. The regional model has been tested and evaluated using ECMWF analyses to define boundary conditions for the stratocumulus regime off the coast of California (Wang et al., 1993). This regional model is now being used to simulate cloud and boundary layer structure over the central Atlantic during the Atlantic Stratocumulus Transition Experiment 1993 (ASTEX). Further evaluation of the effects of drizzle on the thermodynamic structure of the trade wind boundary layer was made using a one-dimensional model (Albrecht, 1993). The Atlantic Stratocumulus Transition -- ASTEX A unique data set was collected during the summer of 1992 as part of ASTEX. The study area for this experiment was over the eastern Atlantic in the vicinity of the Azores and the Madeiras. During this experiment Penn State deployed an extensive suite of remote sensing systems to monitor cloud characteristics continuously from the island of Santa Maria. A 94 GHz cloud radar was used to define cloud-top variations and the internal structure of the stratocumulus decks (Ackerman, et al., 1993). A microwave radiometer provided estimates of cloud liquid water path and integrated water vapor. Cloud base height was defined from a laser ceilometer and cloud base temperature was estimated from an IR radiometer. Broadband radiometers provided fluxes of shortwave and longwave radiation at the surface. Upper-air soundings (4-8 per day) from Santa Maria, Porto Santo, and four ships were placed on the Global Telecommunications System (GTS) and were assimilated into ECMWF and NMC analyses. These analyses provide the thermodynamic, wind, and divergence fields needed to drive the regional model described previously. Initial simulations of the boundary layer and cloud characteristics of the ASTEX study area have been made by Shouping Wang using the regional model. In addition, the detailed measurements from the surface-based remote sensors and the aircraft data that were collected during ASTEX have been used to develop data sets that will be used to test and further develop the parameterizations used in the model. Patrick Minnis, of NASA Langley, is comparing cloud statistics from the Santa Maria measurements with those he retrieves from satellites. Once this verification is completed, the satellite cloud characteristics from Minnis's analyses for the entire ASTEX study area will be compared with characteristics simulated by the regional boundary layer model. Future tasks include: 1) Refine ASTEX regional boundary layer simulations and compare with satellite and surface-based observations. 2) Develop a method for using the regional model with precipitable water estimates from satellite observations with the Scanning Multichannel Microwave Radiometer (SMMR) to improve specification of the water vapor content above the inversion in the subtropics. We will use ASTEX surface-based radiometers as ground-truth for SSMR retrievals. 3) Test the sensitivity of the regional boundary layer model to the large-scale forcing obtained from the output of several GCM's. This will first be done using different versions of NCAR GCM's. The large-scale parameters (vertical velocity, above-inversion moisture, etc.) most critical to the prediction of low cloud will be identified. Cloud distributions will be compared with those from ISSCP. 4) Define how EOS observations will be used to further refine and develop the regional model and define how EOS observations can be used to test the parameterizations used in the regional model. Evaluate the merit of using adjoint techniques to assist with this evaluation. Cloud Microphysics Process Studies The primary goal of the cloud microphysical studies has been to find ways to improve the representation of clouds in atmospheric models over various scales. While the dynamic and thermodynamic setting for cloud formation is established via mesoscale forcing of the atmosphere, the utilization of the excess water is dictated by the phase transformation processes that occur on the microscale. The scales interact in complicated ways, thus necessitating the use of detailed numerical modeling. The approach taken to investigate the role of microphyscial processes in cloud formation, precipitation evolution and cloud persistence has been to develop a process- oriented model from which physically-based parameterizations can be devised and implemented in three-dimensional mesoscale models. Phase 1 of the investigation, completed early in the year, involved the formulation and testing of computer code for the microphysical model, a one-dimensional kinematic framework in which the cloud particles are categorized simultaneously by water mass and solute content (Chen, 1992; Chen and Lamb, 1993a). Ice particles are additionally categorized according to their shape, as indicated by the ratio of lengths along the principal crystallographic axes. Such a multicomponent framework permits the study of a variety of aerosol-clod interaction phenomena not otherwise amenable to investigation. Phase 2 of the investigation has been using the microphysical model to perform sensitivity studies to gain a deeper understanding of the processes involved in cloud formation. One study (Zahn, 1993) looked at the interactive roles played by the atmospheric aerosol during the formation and dissipation of boundary-layer stratocumulus clouds. Even though the boundary layer itself was idealized, it became apparent that the large end of the aerosol size distribution plays an important role in determining the cloud microstructure. Giant cloud nuclei significantly help to destabilize the cloud by initiating the collision-coalescence process that leads to rain formation. At the same time, the coalescence of cloud drops serves as a mechanism to generate additional giant nuclei and to a perpetuation of the rainout process. This mechanism also enhances the atmospheric removal of trace substances and so impacts the chemistry of the atmosphere (Zahn et al., 1993). Such nonlinear processes can only be studied with a model of suitable complexity, in particular one that can track more than one component at a time. The ice phase can be an important initiator of precipitation from cold clouds, as well as the dominant particle type cirriform clouds. A traditionally used parameterization of the crystal habit (shape or form), derived originally from observation, was shown this year to be explainable in terms of fundamental surface-kinetic and vapor-transport mechanisms (Chen and Lamb, 1993b). From this new understanding of the physics of habit formation, the parameterization has been extended to permit calculations of shape changes with time and in repsonse to varying environmental conditions. Our new "adaptive" parameterization scheme should eventually help improve the treatment of ice and precipitation formation in mesoscale models with prognostic water capabilities. Susquehanna River Basin Experiment (Objectives 5,6,7,8) (E. Barron, T. Carlson, T. Gardner, L. Kump, A. Miller, G. Petersen, D. Peuquet, T. Warner, B.Yarnal) Introduction The Susquehanna River Basin (SRB) Experiment (SRBEX) continued to evolve during 1993. SRBEX is the first of several planned regional experiments which are to be undertaken in our EOS research program. The SRB is a 62,419 km2 watershed covering portions of New York, Pennsylvania, and Maryland (Figure 1). Our project seeks to integrate a wide range of observations on climate, hydrology, topography and land resources, with a suite of numerical simulation models for the purpose of studying the elements of the global water cycle (Figures 2 and 3). These models include: the GENESIS global circulation model (GCM), the Penn State/National Center for Atmospheric Research (NCAR) mesoscale model (MM), the Soil Hydrology Model (SHM), the Penn State University Biosphere-Atmosphere Model Scheme (PSUBAMS), the Terrestrial Hydrologic Model (THM), and the Water Quality Model (WQM). Ultimately our results will provide valuable information for evaluating the human dimensions of global change in the SRB. In 1993 the SRBEX research group, which consists of approximately 15 faculty, staff and graduate students, met on a regular basis (approximately every 2 weeks) and continued to build the model linkages and support infrastructure required to carry our research effort through the EOS launch date. As a group, we focused on establishing a sequence of simulation scenarios that will allow us to test, refine, and validate the model linkages. We have established a timeline to guide our interactions through mid-1995 (Figure 4). This evolutionary process has opened new areas of interaction and synergism within our team. The following discussion of the SRBEX research elements elaborates on the relationships shown in Figures 3 and 4. Global Circulation Model The primary objective of the GCM research is to evaluate the capability of the global models to simulate hydrologic cycle components and to execute the simulations which can be nested with the MM to produce high resolution regional predictions. Last year we adopted a new GCM (GENESIS, developed by Starley Thompson and David Pollard at NCAR) and we have now completed extensive testing of this model and we have produced a series of simulations designed for the nesting experiments and SRBEX. These experiments include: 1) R15 control run with a mixed layer ocean (typical of controls for doubled CO2 experiments) 2) R15 run similar to (1) with increased ocean heat transport (designed to evaluate a new capability) 3) R15 run similar to (1) with zero ocean heat transport (designed to evaluate a new model capability) 4) R15 run similar to (1) with uniform ocean heat transport (designed to evaluate cases where the actual distribution of ocean currents is not certain) 5) R15 control run to evaluate lupanov exponents 6) R15 run with 10 years of fixed annually varying sea surface temperatures derived from observations (the best possible case for knowledge of the ocean surface and also a contribution to the AMIP). 7) R15 run with fixed climatological sea surface temperatures (annually varying sea surface temperatures but with the climatologic average rather than interannual variations). 8) R15 control run repeated with doubled atmospheric carbon dioxide 9) T42 control run with a mixed layer ocean (experiment 1 repeated for 8x higher horizontal resolution) 10) T42 run similar to (9) but with doubled atmospheric carbon dioxide 11) T42 run with 10 years of fixed annually varying sea surface temperature (in progress - high resolution version of (6)). These experiments provide important tests of model capability, evaluate the importance of spatial resolution of the GCM in driving the MM, provide the best possible cases of specified conditions for the GCM in driving the mesoscale model, allow direct comparison of the nested model simulations with observations, and plan for future CO2 nested model predictions at a regional scale. The first case studies of nesting the GCM with MM have also been completed (Guerrero, 1992). This work has established the procedures for model nesting. We have also continued our efforts at validation of the GCM climates and exploring the capability of GENESIS to simulate components of the hydrologic cycle. For example, we have been processing the history tapes to derive daily data from the GENESIS present day run using the ten years of observed sea surface temperatures, and the 2xCO2 simulation. Once the daily data are extracted, the plan will be to examine the synoptic-scale circulation in the GENESIS climate. This will be a validation exercise similar to our previous work using the Goddard Institute for Space Studies (GISS) model. We will also look at circulation and temperature/precipitation relationships over the SRBEX region using the neural net approach, and conduct a similar analysis using the GENESIS data. The objective will be to determine whether the same large-scale versus local control is present in the GCM climate as is observed in the NMC data. Synoptic Climatology of the SRB Work through 1992 focused on developing general theory and methods in synoptic climatology (Yarnal, 1993a). This year that research shifted to the specific needs of SRBEX and synoptic climatologyÕs role in that experiment. ManabeÕs (1991) ÒtriadÓ strategy for climate prediction clarifies the place of synoptic climatology in SRBEX (Figure 5). The first activity involves development and application of individual and linked models. The second incorporates in-depth comparison of model output with observations. The third concerns monitoring of the climate system by in-situ and remote observations. Together, the three activities make up the climate-prediction triad, which is informed by process studies and contributes to the human dimensions. Figure 5. The triad strategy for improving climate prediction (modified from Manabe, 1991). Synoptic climatology in SRBEX fits two of these boxes. First, process studies are underway that relate the atmospheric circulation on daily and longer time scales to precipitation, evapotranspiration, runoff, soil moisture and water quality. Of these variables, precipitation is critical because it is a principal output of the GCM and the MM and a necessary input to the remaining SRBEX models (Figure 3). Present process studies are directed to improving the initial synoptic classifications (Yarnal, 1993b; Yarnal et al., 1993) and to relating the atmospheric circulation to varying scales of surface climate. The latter involves conducting synoptic climatologies on watersheds of various sizes, ranging from WE-38 (the intensive study area in the Mahantango Creek watershed) to the SRB itself. Preliminary findings for the SRBÕs Leading Ridge One watershed show how wintertime migratory cyclonic systems control precipitation, runoff and high-sulfate events, while stationary fronts dominate summer hydrology (Yarnal and Draves, 1993). The second box of Figure 5 to which synoptic climatology relates is model validation through diagnostic study. By definition, synoptic climatology is the study of relationships between the atmospheric circulation and the surface environment. Thus, the synoptic climatologies of precipitation, evapotranspiration, runoff, soil moisture and water quality provide diagnostics of model linkages between the large-scale atmospheric models (GENESIS and MM) and the smaller-scale models of the surface environment (THM, PSUBAMS, SHM and WQM). These model-linkage validations Ð when compared to conventional validations that relate model output to simple variables Ð are a unique feature of SRBEX. Initial model-linkage validations by synoptic climatology will follow each linked-model experiment. Through an iterative process, these diagnostics will improve the models, while the models will suggest the most appropriate future remote sensing and diagnostic activities (Figure 5). Penn State/NCAR Mesoscale Model The Penn State/NCAR mesoscale model (MM) is a versatile three-dimensional, limited-area meteorological numerical model. The MM uses a high-resolution planetary boundary layer (PBL) model developed by Blackadar, and described by Zhang and Anthes (1982). This PBL model uses a modified version of BATS, the Biosphere-Atmosphere- Transfer Scheme, (Lakhtakia and Warner, 1993) as a sophisticated surface-physics/soil- hydrology parameterization module. During the last year, a number of model-development efforts have been underway that will enable the MM to be used as a more effective tool for studies of regional climate change. The major effort involved the evaluation of the 5th generation of the PSU/NCAR mesoscale model (MM5) for a variety of meteorological situations. This model differs significantly from the previous version (MM4), which has been utilized for about the last eight years, in that it is nonhydrostatic and it has a more sophisticated atmospheric hydrologic cycle, an improved atmospheric radiation scheme, and more efficient numerics to make longer-range simulations more tractable. MM5 has been tested on a number of meteorological cases over a wide range of scales, where the emphasis has been on the hydrologic cycle and surface forcing. For example, the surface energy budget has been evaluated through comparison of the predicted surface-layer temperatures and special field program surface temperatures in the San Joaquin Valley; the tests showed that the surface energy-budget equation in MM4/5 produced good quality simulations of the diurnal variation of surface temperature, provided that the surface conditions were specified accurately. As soon as MM5 Version 1 is released for public use (before the end of 1993) the Biosphere-Atmosphere Transfer Scheme (BATS) will be incorporated into MM5, and that model will then be used as part of the SRBEX suite of models (Figure 3). In the interim, testing of BATS in a non-hydrostatic version of MM4 continues. The domain to be used in the regional simulations for the Susquehanna River Basin is shown in Figure 6. A coarse-grid mesh of 61 x 61 grid points in the horizontal and 36 km grid spacing is used, with 2 nested grids of 12 and 4 km grid spacing. All three grids are centered over the Mahantango Creek watershed. Detailed land-surface-process parameterization schemes like BATS require accurate information on land-surface and soil properties. To address these requirements, the USGS- EROS Data Center (EDC) Land-Surface Characteristics Database has been acquired for SRBEX studies. This database includes 159 land-cover classes at a 1 km resolution for the entire 48 conterminous United States. The land-cover classes in this database were reduced to the 18 vegetation/surface-cover types used in BATS. The data were transformed to meet the cartographic requirements of the mesoscale-model grid, and were then aggregated (using a modal aggregation technique) from the 1 km original resolution to all three model grids (36, 12 and 4 km) (Smith, 1993). The USDA-SCS STATSGO soil database provides the most useful resource for characterizing the nature of the soils found at scales represented by mesoscale model domains. Through a cooperative agreement with the USDA-SCS National Soil Survey Center we have recently obtained STATSGO soils information for the eastern two-thirds of the conterminous United States. The linkage of the MM with the THM and the SHM is a central element within SRBEX (Figure 3). This linkage consists of the SHM providing initial values of soil-water content to the MM and the THM. In turn, the MM provides the simulated precipitation data to the THM. This process is depicted in Figures 7 and 8 (Lakhtakia et al., 1993). In Step 1, the GIS is used to register the data obtained from a particular database (e.g., the USGS- EDC land-surface-characteristics, the USDA-SCS soils and the USGS 3 arc-second digital elevation model (DEM) data) to a specific model grid. In this particular case there are two different types of horizontal grids involved (i.e., the SHM-MM grids and the THM grids). The GIS provides the means for model grid registration. Step 2 represents the execution of the SHM for the particular experiment. The output from the SHM is regridded in the GIS (Step 3) in order to provide part of the initial conditions for the MM and the THM. The MM is executed in Step 4. The precipitation fields simulated by the MM are piped into the GIS environment, where they are regridded (Step 5) for use by the THM as a forcing term in the surface hydrology balance. The THM is executed in Step 6 and its output is placed in the GIS (Step 7). A series of experiments with these models is planned, as shown in the timeline diagram (Figure 4). The first of these simulations, a storm event, has been chosen and the initial conditions for the MM created. Soil-Vegetation-Atmosphere (SVAT) Model/Remote Sensing Soil Hydrology Model (SHM) link with Mesoscale Model (MM) The use of a sophisticated scheme like BATS within the MM requires the initialization, among other variables, of the soil-water-content profile at each of the MM grid-point locations. Chris Smith demonstrated in his M.S. thesis (Smith, 1993) that a soil hydrology model could be used to provide the MM with a field of initial soil-water content. The soil hydrology model used in that work is the SHM, a one-dimensional, diffusion- gravitation soil-hydrology model driven by conventional meteorological, soil and vegetation information (Capehart and Carlson, 1993a). The data used in Chris Smith's work was obtained from the NCAR archives and the USGS-EDC Land-Surface-Characteristics data. The SHM was executed over the MM grid domains (Figure 6) for a four-month period in 1990 (March-July) to produce the soil- water-content profile for each of the grid-point locations. This information was then used as part of the initial conditions for a 12 hour simulation with the MM at the end of the four- month period (mid-July 1990 - coincident with the NASA MAC-HYDRO mission). In the process of linking the SHM to the MM, it was found that only two months of input meteorological data (rather than four) are required to achieve a stable initial soil-water- content field for the MM. A paper, co-authored by Smith, Lakhtakia, Capehart and Carlson (Smith et al., 1993b), on this subject has been submitted to the Bulletin of the American Meteorological Society. We have also prepared -- with the aid of William Jester, an engineering student -- an animated film showing the time evolution of the soil-water- content fields as generated by the SHM for the aforementioned four months of 1990. This film will soon be available for demonstration (copy available for loan). Steve DiRienzo, an M.S. candidate in meteorology, plans to conduct another pilot study with the SHM. He will execute the SHM for different land-cover conditions using operational weather data available in the Penn State Department of Meteorology rather than relying on archived data. One goal in this work is to allow the SHM to be used as a pedagogical tool and another goal is to investigate the possibility of using more than one land-cover category at a given grid point in the MM grid. Eventually, however, we plan to execute the SHM for an arbitrary grid of points over the entire United States, Canada and Mexico using the Penn State weather data. Penn State University Biosphere Atmosphere Model Scheme -- PSUBAMS The Penn State University Biosphere Atmosphere Model Scheme (PSUBAMS) is our principal tool for interpretation of satellite (visible and infrared temperature) imagery. For the past two years we have been working on a new method that allows determination of the surface soil-water content, the surface energy balance and the fractional vegetation cover over partially vegetated surfaces. To date, one paper (coauthored by Carlson, Gillies and Perry) has been accepted for publication in Remote Sensing Reviews (Carlson et al., 1993). A second paper, based in part on the Ph.D. thesis of Robert Gillies (Gillies, 1993) is about to be submitted to the Journal of Applied Meteorology (Gillies and Carlson, 1993a). One of the unique aspects of the technique we describe in these two papers is the recovery of the surface soil-water content, as opposed to a bulk value, which neglects the presence of vegetation and so, has far less physical meaning. We are working on an additional paper in which the surface soil-water-content measurements reduced from C-130 NS001 radiometric measurements from the 1990 MAC- HYDRO mission are to be collated with surface soil-water-content estimates obtained from the Push Broom Microwave Radiometer measurements made at the same time. This paper, by Carlson, Gillies and Schmugge will be submitted to Agricultural and Forest Meteorology as part of a dedicated volume of papers associated with the recent workshop on Thermal Remote Sensing of the Energy and Water Balance over Vegetation, held in La Londe, France, in September 1993 (Gillies, et al., 1993). This workshop constitutes an ESSC initiative and was organized by Toby Carlson. Robert Gillies is organizing the proceedings for that workshop. Although we have not yet subjected the method to rigorous analysis, we anticipate making some verification using the datasets available from International Satellite Land Surface Climatology's (ISLSCP) First Field Experiment (FIFE). Several case studies will be made using coincident NOAA-AVHRR and C-130 NS001 radiometer data from the intensive field phases of FIFE in 1987 and 1989. A way of linking PSUBAMS to the SHM (Figure 3) constitutes the focus of Bill Capehart's Ph.D research. Capehart is adapting the method of systematically inserting remotely determined estimates of surface soil-water content and fractional vegetation cover in the SHM. He plans to develop a technique for nudging the continual series of estimates provided by the SHM with intermittent estimates as determined from satellite observations (Capehart and Carlson, 1993b). Two other related projects involving PSUBAMS are planned. First, we are contemplating a study of deforestation/urbanization. We will use archived AVHRR data for a couple of rapidly growing urban areas and two forests, one in Mexico and one in the United States or Canada. We propose to develop a new type of deforestation/urbanization index based on two parameters -- surface soil-water content and fractional vegetation cover (Gillies and Carlson, 1993b). These parameters are highly descriptive of land use and are also required as input to climate models. Consequently, the deforestation/urbanization index can also be used to study climatological effects of land-use changes with the aid of climate models. A second aspect of PSUBAMS will be, in part, pedagogical. We plan to develop a version of PSUBAMS with a sophisticated user interface and a "book" (or training manual) that will allow the model to be used to teach and assist persons not trained in boundary layer meteorology (such as architects, agronomists, engineers, etc.) to address problems in their own discipline. To foster this use of the model, we plan to conduct workshops through ESSC starting sometime in 1995. Watershed Modeling Studies Terrestrial Hydrologic Model (THM) The THM is a multi-layer, multi-level, raster-based model for distributed-parameter modeling of the terrestrial hydrologic system. During 1993 we focused on continued model component development, improvement, and validation. Figure 9 illustrates the basic components of the current model. Initially, the model requires a DEM. The drainage network is delineated and all cells are initialized. Cell accumulation, flow direction, overland flow-routed, and channel flow routed files are generated by the program. Because the model was written without arrays, dataset size is, effectively, unlimited. Many other models of a similar nature have limited study area capabilities due to the sizing of the arrays and hardware memory restrictions. Figure 9. Terrestrial Hydrologic Model components Current hydrologic-abstraction options include a phi index, a constant loss regardless of item and soil conditions and a curve number, linked to time only through rainfall patterns and almost entirely dependent on the soil characteristics. Efforts to utilize the Green-Ampt infiltration scheme employed by the SHM have proven very fruitful and it is intended that the full integration of the Green-Ampt routine and the THM will be completed by the end of the year. To date the model has received limited testing on a few sample DEM's. Simulated data has been used for testing of many of the model routines. One of the goals for the immediate future is test the model with the data represented in Table 1. Calibration of the model with actual storm events and comparison of the model with other hydrologic prediction models is planned for early 1994. Other research objectives include the investigation of data scale on the hydrologic parameterization and the eventual effect on the outfall hydrograph and overall mass balance. Data Description 30m Resolution 100m Resolution 1 km Resolution DEM Mahantango Mahantango & Susquehanna Mahantango & Susquehanna Land Use / Land Cover Mahantango Mahantango & Susquehanna Mahantango & Susquehanna Sat. Hyd. Conductivity (STATSGO) Mahantango Mahantango & Susquehanna Mahantango & Susquehanna Matrix Potential (STATSGO) Mahantango Mahantango & Susquehanna Mahantango & Susquehanna Field Capacity (STATSGO) Mahantango Mahantango & Susquehanna Mahantango & Susquehanna Hydrologic Soil Grouping A,B,C, etc (STATSGO) Mahantango Mahantango & Susquehanna Mahantango & Susquehanna SSURGO Mahantango Mahantango Mahantango Table 1. Data for terrestrial hydrologic model testing and calibration In 1993 our research was presented and published in the Proceedings of the American Society of Civil Engineers - 1993 Intenational Symposium on Engineering Hydrology in San Francisco (Miller and Johnson, 1993). An abstract has been accepted for oral presentation and publication at the 1994 American Water Resources Association, Annual Summer Symposiium (Johnson and Miller, 1993). Initialization of the SCS TR-55 Hydrologic Model with Spatial Information Egide Nizeyimana, under the supervision of Thomas Gardner, has just begun an M.S. program in Geosciences. His research will focus on the use of DEM's, soils information and remotely sensed data to generate input parameters for the SCS TR-55 hydrologic model for 3 watersheds of different sizes and cell resolutions. Comparisons will be performed between the USDA-SCS State Soil Geographic (SSURGO) data and the more generalized USDA-SCS STATSGO soils data, aerial photographs and Landsat Thematic Mapper (TM) images, as well as Landsat TM and Advanced Very High Resolution Radiometer (AVHRR) data. The TR-55 model, which is written in Basic, will be translated into the C language in order to interface to the GRASS GIS. Precipitation input data will be a series of rainfall events of 24-hour duration with a return period of 10 years estimated from rainfall-frequency distribution maps. The streamflow discharge will be estimated for type II storm rainfall distribution. The degree of accuracy of runoff curve numbers from the above sources in predicting runoff depth and peak discharge will be determined by comparing predicted to observed data for the same rainfall events. Statistical analyses will be performed using linear regression. Synthetic Aperature Radar Studies at Mahantango Creek Watershed The NASA-sponsored MAC-HYDRO mission in July 1990 acquired a range of multi-temporal SAR data, including like and cross polarized data in three frequencies, over the Mahantango Creek watershed. During the mission a range of soil-moisture conditions existed, providing an opportunity to test the sensitivity of the SAR data to soil-moisture variations. Eric Warner, a Ph.D. candidate in Soil Science under the supervision of Gary Petersen, is evaluating the MAC-HYDRO SAR data to determine the extent of the relative impacts of soil moisture, crop cover and terrain on SAR backscatter, to normalize crop and terrain influences on SAR backscatter, to predict soil-moisture content, and to examine the effectiveness of slant to ground range conversion and reflector-based calibration on the sensitivity of SAR backscatter. Terrain influences are examined as the product of multi-scale interactions. Previous research has noted these interactions as occurring at pixel and supra-pixel levels. However, a limited range of terrain variables were examined, resulting in an incomplete understanding of terrain influences on SAR backscatter. The use of DEM's will allow for the determination of a more complete set of terrain variables, improving insight concerning the relationship between SAR backscatter and terrain. Approaching SAR image studies through ground and instrument variables allows for a more complete exploration of the limitations of the instrument. The results obtained from this study will provide other researchers with increased insight about SAR as a data- collection tool. For those specifically involved in modeling, this insight will include an understanding of which SAR frequency and polarization combination allow for collection of desired data and the impact of calibration and terrain on the data-collection process. The increased understanding will also be valuable in initializing models and conducting sensitivity and error analyses. Water Quality Modeling Introduction Our goals in 1993 were to compile an extensive chemistry and hydrological dataset with which to test the SRBEX hydrology and water quality model parameterizations, to use this dataset to assess the environmental factors (rainfall, temperature, hydrological flow path) which control rates of chemical weathering, to determine how the effect of these factors vary as a function of catchment size, and to use this information to develop empirical relationships for water quality which can be used within the SHM and THM to estimate chemical fluxes. In addition, we have been analyzing a number of rock cores recovered from the Mahantango Creek watershed for mineralogy, bulk chemistry, and fracture density and orientation. The latter analysis is proving quite useful in our modeling of flowpaths within the watershed, and indicating that the bulk of flow is confined to the upper 5 m of the fractured bedrock layer. The following describes our achievements over the last project year, grouped in terms of the types of models being developed. Watershed Mass Balance Model We now have nearly a year of observations of weekly surface-water and monthly well-water chemistries for watershed WE-38 (a Mahantango Creek sub-basin) (Figure 10) and for two subwatersheds within WE-38, WD-38 and W.2 (Figure 10). Watershed W.2 is located at the top of the watershed on forested colluvium, and receives little agricultural impact. WD-38 is located in a hilly region at the bottom of the watershed, and is extensively impacted by agricultural activity. WE-38 itself includes a variety of surficial environments, with forested ridges, farmed and wooded valleys; it is thought to be representative of typical, non-carbonate Ridge and Valley watersheds within the Susquehanna Drainage Basin. Surface-water, well-water, and rain-water samples have been analyzed for a variety of dissolved constituents, including both major and trace elements, and the nutrient elements (Figure 11). We will soon complete the year-long sampling campaign, and will then be carrying out individual mass-balance calculations for each element in each subwatershed. Part of these calculations requires realistic estimates of the mineralogy and thickness of soils throughout each watershed. Figure 11. Major cation concentration and discharge versus time for weir WE-38. We are using the USDA-SCS SSURGO soils data in a GIS format to aid in this process. Figure 12 shows one of these layers: the distribution of soil types and sampling locations within the watershed. Recharge Model One of our research goals is to utilize the output from the (SHM) (Figure 3) to determine the rates of chemical weathering as a function of depth within a soil column. As a step in this direction, we have developed a recharge model to estimate recharge fluxes below the soil zone, based primarily on observed water table fluctuations. Yearly recharge cycles calculated with the model are comparable to those calculated in the standard way, using a water budget. A cumulative recharge plot (Figure 13) demonstrates that most of the recharge of groundwaters in this watershed occurs during the 10 largest recharge events. We are working with Brent Yarnal to relate these events to patterns in the watershed's climatology using synpotic climatologic techniques. The implication for us is that the contribution of dissolved ions from the soil zone to the groundwaters occurs during infrequent, large events, and should lead to temporal variability in groundwater quality. Our well-water chemistry observations substantiate this prediction. Our next steps include a comparison of our recharge estimates to those from the SHM, and a field test of the models on a small hillslope in WE-38. Figure 13. Cumulative recharge versus number of recharge events for 1983 - 1987. Chemical Weathering Model In conjunction with the recharge modeling effort, we have developed a leaching model which calculates the rates of mineral dissolution as a function of soil-moisture content, soil- water chemistry, temperature, reactive mineral surface area, and recharge frequency. We have run the model for the ideal case of quartz dissolution, and have found that both temperature and rainfall recurrence interval have large effects on the rate of mineral dissolution (Figure 14). These are factors that have not been well considered in discussions concerning the discrepancy between laboratory and field determinations of the rates of chemical weathering. Given that previous watershed mass balance calculations have typically assumed uniform temperatures and water-saturated soil conditions, we may now explain these discrepancies quite readily, given the orders of magnitude sensitivity to these factors we calculate. Figure 14. Effective leaching rate as a function of rainfall recurrence interval and temperature. Calculation is for the dissolution of quartz in a silt loam. Human Dimensions of Environmental Change in the SRB Research on the human dimensions of environmental change has been added to SRBEX. Physical studies are not sufficient to understand the SRBÕs hydrology; hydrologic change can only be understood as a complex function of physical and social forces. Preliminary investigation (Yarnal, 1993b) has shown that several sectors of human activity Ð agriculture, forestry, mining, industry and energy, urbanization, and transportation Ð have had direct impacts on basin hydrology. Furthermore, these activities are driven by several interacting social forces, i.e., population change, economic growth, technological innovation, socio-political institutions, and culture and beliefs (National Research Council, 1992). To predict future hydrologic change, these social driving forces must be understood. Therefore, the scientific objectives of the human-dimensions research are: 1) To distinguish the human-induced changes in basin hydrology over time; 2) To identify the human activities that caused those changes; 3) To understand the social forces driving those activities. The ultimate goal is to understand how human systems interact with physical processes to produce the observed hydrology of the SRB. Understanding this interaction will guide monitoring efforts and suggest policy for mitigating and adapting to environmental change. The human dimensions program of SRBEX just got underway in 1993. Initial activities included surveys of human-induced hydrologic change in the SRB (Yarnal, 1992/93, Kasper et al., 1993) and GIS-based studies of land use in one sub-watershed (Chamberlain, 1993; Kasper, 1993). Work in progress spotlights the human dimensions of land-use/land-cover change. SRBEX GIS and Data Management System Activities as part of the SRBEX GIS and Data Management System have focused on obtaining additional geographic data needed to support SRBEX modeling activities, augmenting existing software for preprocessing the geographic data and remotely sensed imagery, and enhancing facilities for cataloging and archiving these data. New geographic data acquired during the year include 25 meter resolution land-use data for the entire Chesapeake Basin, derived from Landsat Thematic Mapper imagery as part of the EPA EMAP program; soils information (USDA-SCS STATSGO) for the eastern two-thirds of the U.S.; additional USGS 3-arc second digital elevation data (DEM) to complete coverage for the Cheseapeake Basin; and 90-meter DEM data for a number of 7.5 minute map quads in the SRB. Because these data are in a variety of formats, generated by diverse image processing and GIS packages, we have continued to acquire and, when necessary, develop additional software for preprocessing and format conversion. Newly acquired software includes the SUN workstation versions of the ARC/Info GIS software, and the LAS and ERDAS Imagine raster image processing packages. These packages are being installed both on workstations attached to the ESSC network and on two new SUN SPARCstation Model 10/30 computers, acquired with non-NASA funding, located at the Office for Remote Sensing of Earth Resources (ORSER) at Penn State. Software written by SRBEX team members during the past year includes ARC/Info Arc Macro Language (AML) routines for extracting properties from the STATSGO and converting LAS images between VAX and SUN SPARCstation data formats. Work in progress includes additional software modules for regridding and rescaling rasterized geographic data into formats and map projections needed for model ingest. Data structures have been defined to facilitate catalog searches and retrievals of data at all scales of SRBEX analysis. These data structures are being implemented to facilitate access from the ESSC workstation network with automatic migration of data to and from a large tape archive using the Cray Data Migration software on the Cray supercomputer. We have begun the process of transferring existing GIS data from offline magnetic tape and cartridge disk archives located at ORSER to the ESSC archival system. Because of the importance of complete documentation of all data, (e.g. remotely sensed imagery and ancillary spatial data, as well as model outputs) we have started development of an interactive system for entering, editing, and retrieving data documentation and other metadata. We will attempt to implement the draft "Content Standards for Spatial Metadata" being developed by the Federal Geographic Data Committee, although the initial version may support only a subset of these standards. We are attempting to keep informed about similar efforts elsewhere, in the hope that we will be able to incorporate features and modules from other systems into our own design, thereby minimizing duplication of effort. Advanced Techniques for the Representation of Geographic Information Temporal representation in geographic information systems has long been conceptually and structurally deficient. In the past year the temporally-based component of a database design has been successfully implemented in a demonstration prototype called TGIS (Temporal Geographic Information System). This data model, in its current form, incorporates locational information with a temporal context. Presently, TGIS also includes the following elements as operational capabilities: a specialized graphics interface, a facility for converting digital data from several other formats for input into TGIS, and an initial suite of manipulation and display algorithms. The initial testing phase of TGIS has been completed, showing very favorable results in terms of both storage compactness of the data model and efficiency of the initial manipulation and display algorithms. Cape Experiment (Objective 9) (S. Goodman) Research activities for this period have been directed toward modeling surface energy and hydrologic processes utilizing data collected during the Convection and Precipitation/Electrification Experiment (CaPE) held in east central Florida in 1991. The objectives of this project are to establish and apply methodologies for the diagnosis of land and atmospheric water budget components for the CaPE region (approximately 25000 km2). The underlying philosophy guiding this study is that these techniques can be applied on scales consistent with Global Energy and Water Cycle (GEWEX) Contiental International Project (GCIP) activities such as the CART ARM experiment and ultimately the Mississippi basin. The surface energy and water flux component of this investigation is being carried out using a land surface model based on BATS, in conjunction with data from a wide array of measurement systems. The model has been tested using data from the two Florida State University flux sites. These stations were used because of the availability of model input variables - wind, temperature, humidity, pressure, precipitation and solar and longwave downwelling radiation, as well as flux measurements necessary for model validation. Results from these simulations indicate that the model, using soil and vegetation parameter values appropriate for the local conditions, is capable of accurately estimating surface energy and moisture fluxes. Model simulations are currently being performed for each of the 38 PAM sites within our study area, with the aim of producing an initial estimate of areal heat and moisture fluxes. Thirteen of the PAM stations measured incident shortwave radiation; 4 of these also collected reflected shortwave, emitted longwave and net radiation, and soil temperatures. Model sensitivity to radiation input will be tested using a variety of methods for specifying solar and longwave fluxes using the point measurements. For example, what is the impact on model-diagnosed fluxes of using spatially uniform radiative input, as compared with values measured at each site? A more sophisticated modeling scheme for estimating areal fluxes for the CaPE domain has been designed. This method incorporates BATS, geographic information (landcover classes and soil properties), and statistical distributions of surface properties (such as leaf area index, albedo and fractional vegetation cover) based on high-resolution remotely sensed data. Distributions of normalized difference vegetation index (NDVI) and spectral albedo have been derived from 20 m resolution SPOT imagery for each of the 18 land cover classes in the study area. The BATS model will be run at grid points for the CaPE domain; each grid cell will be treated as a mixture of landcover types. To add further realism to the model, the statistical distributions of surface properties within each landcover 'patch' will be represented via a discrete probability density function inferred from the observed distributions of NDVI and albedo. Scale issues will be addressed with a series of model runs in which the resolution of remotely sensed data, used to establish the nature of surface variability, is degraded. Preliminary analyses have shown that degradation of SPOT data from 20 mup to 1 km resolution (simulating AVHRR footprints) results in large changes in both mean and variance of surface properties. Ice Sheet Mass Balance (Objective 10) (R. Alley) Our research related to the determination of ice sheet mass balance is proceeding smoothly. One of the main assumptions in projections of future sea level is that much of the water released by melting of low-latitude glaciers and of low-altitude regions of polar ice caps will be transferred to high-altitude regions of ice caps because of increased precipitation in the warmed world. We have tested this relation against data from the Greenland Ice Sheet II (GISP2) deep ice core from the central region of the Greenland ice sheet, and we find that it is not well-supported in Greenland. The usual expectation (precipitation proportional to the temperature gradient of the saturation vapor pressure) is not realized in the recent (Holocene) or in the the Younger Dryas cold event or the main ice age (Oldest Dryas). This relation predicts an increase of about 11% in precipitation per degree warming, but the ice core records less than 1% increase in accumulation per degree warming. Transitions between glacial and interglacial conditions (warming at the end of the ice age, cooling into the Younger Dryas, warming from the Younger Dryas) show slightly more than 11% change in accumulation per degree change in temperature, but this almost certainly is caused by changes in strom tracks (which were far south of Greenland in cold periods but near Greenland in warm periods). These data suggest that temperature has little to do with precipitation in central Greenland, and that atmospheric dynamics dominate. This forms the M.S. research for Wanda Kapsner and will be presented to the Union Session at the Fall, 1993 meeting of the American Geophysical Union and subsequently prepared for publication (Kapsner, et al., 1993). Chris Shuman has advanced his demonstration that changes in near-surface snow conditions on ice sheets over days to weeks can be detected using Special Sensor for Microwave Imaging (SSM/I) polarization data. He is conducting spatial and temporal mapping of events that affect the paleoclimatic record archived in ice cores, and that affect sublimational moisture fluxes from the ice sheet to the atmosphere. A report of this work was just published (Shuman, 1993) and a second manuscript is in press in Geophysical Research Letters. Landscape Evolution Modeling (Objective 11) (R. Slingerland, G. Tucker) As presented at the December 1992 NASA EOS site visit, our research is focused on the general question, "How do erosion, transportation, and deposition of earth's surface materials reflect climate change? We are posing two more specific questions: 1) How do catchment morphometrics and sediment yields evolve in response to changing climate? 2) What are the interactions among climate, tectonics, and topography at the regional scale? To answer these questions we have developed a Geophysical Landscape Evolution Model (GOLEM), using various geomorphic parameters such as drainage density and relief from a variety of settings. This has been accomplished and the results presented in a paper at the Third International Geomorphology Conference in August 1993 (Tucker and Slingerland, 1993). A paper has also been accepted for publication in the Journal of Geophysical Research (Tucker and Slingerland, 1993) REFERENCES Ackerman, T.P., B.A. Albrecht, M.A. Miller, E. Clothiaux, R. Peters, and W. Syrett. 1993. Remote sensing of cloud properties using a 94 GHz radar. In: Topical Symp. on Combined Optical-Microwave Earth and Atmosphere Sensing. March 22-25. Albuquerque, NM. pp. 211-214. Albrecht, B.A. 1993. Effects of precipitation on the thermodynamic structure of the trrade wind boundary layer. J. Geophys. Res. 98:7327-7337. Bluth, G.J.S. and L.R. Kump. 1993. Lithologic and climatologic controls of river chemistry. Geochim. et. Cosmochim. Acta. (In Press) Capehart, W.J and T.N. Carlson. 1993a. Estimating near-surface soil moisture availability using a meteorologically driven soil water profile model. J. Hydrology. (Accepted for Publication) Capehart, W.J. and T.N. Carlson. 1993b. Estimation of surface moisture availability using a hydrology budget model aided by surface satellite observations and a soil- vegetation atmosphere transfer scheme (SVAT). 21st Con. on Ag. and Forest Met. San Diego, CA. (Abstract Submitted) Carlson, T.N., R.R. Gillies, and E.M. Perry. 1993. A method to make use of thermal infrared temperature and NDVI measurements to infer surface soil water content and fractional vegetation cover. Remote Sensing Reviews -- Special Issue on Recent Advances in Remote Sensing Science. (Accepted for Publication) Chamberlain, R. 1993. Fragipans and land use in the Mahantango Creek Intensive Study Area. Unpublished working paper. Dept. of Geography. The Pennsylvania State University. Chen, J.P. 1992. Numerical simulations of the redistribution of atmospheric trace chemicals through cloud processes. Unpublished Ph.D. dissertation. Dept. of Meteorology, The Pennsylvania State University. 342 p. Chen, J.P. and D. Lamb. 1993a. Simulation of cloud microphysical and chemical processes using a multicomponent framework. Part I: Description of the microphysical model. J. Atmos. Sci. (In Press) Chen, J.P. and D. Lamb. 1993b. The theoretical basis for the parameterization of ice crystal habits: Growth by vapor deposition. J. Atmos. Sci. (In Press) Christy, J.R. and S.J. Drouilhet. 1993. Variability in daily, lower zonal mean lower stratospheric temperatures. J. Climate. (In press) Christy, J.R. 1992. Monitoring global temperature changes from satellites (Chapter 11). Global Climate Change: Implications, Challenges and Mitigation Measures. S.K. Majumdar et al. eds. The Pennsylvania Academy of Science. 566 pp. Dutton, E.G. and J.R. Christy, 1992. Solar radiative forcing at selected locations and evidence for global lower tropospheric cooling following the eruptions of El Chichon and Pinatubo. Geophys. Res. Lett. 19:2313-2316. Gillies, R.R. 1993. A physically based land-use classification scheme using remote solar and thermal infrared measurements suitable for describing urbanization. Unpublished Ph.D. dissertation. Dept. of Architecture. University of Newcastle, UK. Gillies, R.R and T.N. Carlson. 1993. Thermal remote sensing of surface soil water content with partial vegetation cover for incorporation into mesoscale prediction models. (Submitted to J. Appl. Met.) Gillies, R.R and T.N. Carlson. 1993b. A physically based modeling approach for including remotely derived measurements in the study of land use change. Effects of Human-Induced Changes on Hydrological Systems. American Water Resources Association (AWRA) Summer Symposium. (Abstract Submitted) Gillies, R.R., A. Olioso, and K. Humes. 1993. (Ed's). Proc. Workshop on Thermal Remote Sensing of the Energy and Water Balance over Vegetation in Conjunction with Other Sensors. (In Preparation) Guerrero, T. 1992. Linking a general circulation model and a mesoscale model to examine the effects of model resolution on a simulated last glacial maximum storm in the western North Atlantic Ocean. Unpublished M.S. thesis. Department of Geosciences, The Pennsylvania State University. 233p. Hewitson, B.C. and R.G. Crane. 1993. Precipitation controls in southern Mexico. In: Hewitson, B.C. and R.G. Crane (Eds.). Neural Nets: Applications for Geography. Kluwer Academic Pub. Dordrecht, Holland. (In Press) Johnson, D.L. and A.C. Miller. 1993. A hydrologic model for use with raster data. Effects of Human-Induced Changes on Hydrological Systems. American Water Resources Association (AWRA) Summer Symposium. (Abstract Submitted) Kapsner, W.R. 1993. Response of snow accumulation to temperature variations in Central Greenland. Trans. Amer. Geophys. Union, San Fransisco, CA. (Abstract Submitted) Kasper, W.J. 1993. Modeling Spatio-Temporal Patterns of Nitrous Oxide and Methane Soil-Gas Fluxes on a Farmstead. Unpublished M.S. thesis. Dept. of Geography. The Pennsylvania State University. Kasper, W.J., E.P. Arabas, C.J. Rosin, L.L. Diamond, R.A. Heidl, and A.S. Hertz. 1993. The human dimensions of water-quality changes in the Susquehanna River Basin. Unpublished working paper. Dept. of Geography. The Pennsylvania State University. Lakhtakia, M.N. and T.T. Warner. 1993. A comparison of simple and complex treatments of surface hydrology and thermodynamics suitable for mesoscale atmospheric models. Mon. Wea. Rev. (Accepted for Publication) Lakhtakia, M.N., D.A. Miller, R.A. White, and C.B. Smith. 1993. GIS as an integrative tool in climatologic and hydrologic modeling. Second Int. Con./Workshop on Integrating Geographic Information Systems and Environmental Modeling, September 26-30. Breckenridge, CO. Mace, G.G., T.P. Ackerman, and E. Clothiaux. 1993. Mesoscale diagnostic quantities from arrays of doppler wind profilers and radiosondes. In: Preprints of the Fourth Symposium on Global Change Studies. January 17-22. Anaheim, CA. pp. 232- 234. Mace, G.G. and T.P. Ackerman. 1993a. Cirrus cloud development in a mobile upper tropospheric trough: The November 26th FIRE cirrus case study. Proc. of the FIRE Cirrus Science Con. June 14-17, 1993. Breckenridge, CO. Mace, G.G. and T.P. Ackerman. 1993b. Examination of the observed synoptic scale Cirrus cloud environment: The December 4-6 FIRE Cirrus case study. Proc. of the FIRE Cirrus Science Con. June 14-17, 1993. Breckenridge, CO. Manabe, S. 1991. Triad strategy for improving climate prediction. In: Confronting Climate Change: Risks, Implications and Responses, edited by Irving M. Mintzer, New York: Cambridge University Press. p. 51. McGinnis, D. 1993. Predicting snowfall from synoptic circulation: A comparison of linear regression and neural network methodologies. In: Hewitson, B.C. and R.G. Crane (Eds.). Neural Nets: Applications for Geography. Kluwer Academic Pub., Dordrecht, Holland. (In Press) Miller, A.C. and D.L. Johnson. 1993. Formulation of a hydrologic model for use with remotely sensed data. Proc. Amer. Soc. Civ. Eng. - 1993 Int. Symp. on Eng. Hydro., San Francisco, CA. National Research Council, 1992: Global Environmental Change: Understanding the Human Dimensions. Washington, DC: National Academy Press. Robertson, F.R. and E.W. McCaul. 1994. Large scale structure of water vapor and condensate over the TOGA COARE Region. 6th Amer. Met. Soc. Con. on Climate Variations. January 23-28, 1994 Nashville, TN. (Abstract Submitted) Robertson, F.R., E.J. Barron, S. Goodman, D. Fitzgerald, B. Bishop, J. Christy, S. Thompson, and D. Pollard. 1993. GENESIS climate model: Intercomparsions with multiple climate data bases. In: Preprints Fourth Symp. on Global Change Stud. Jan. 17-22, 1993 Anaheim, CA. Amer. Met. Soc. pp. 3-8. Shuman, R.B. Alley, and S. Anadakrishnan. 1993. Characterization of a hoar- development episode using SSM/I brightness temperatures in the vicinity of the GISP2 site, Greenland. Ann. Glaciol. 17:183-188. Shuman, C.A. and R.B. Alley. 1993. Spatial and temporal characterization of hoar formation in Central Greenland using SSM/I brightness temperatures. Geophys. Res. Letters. (In Press) Smith, C.B. 1993. Initialization of soil-water-content for regional-scale atmospheric prediction models. Unpublished M.S. thesis. Dept. of Meteorology. The Pennsylvania State University. 87 p. Smith, C.B., M.N. Lakhtakia, W.J. Capehart, and T.N. Carlson. 1993b. Initialization of soil-water content in regional-scale atmospheric prediction models. Bull. of Amer. Met. Soc. (Paper Submitted) Spencer, R.W. and J.R. Christy. 1993. Precision lower stratopsheric temperature monitoring with the MSU: Validation and results 1979-91. J. Climate 6: 1194- 1204. Tucker, G.E. and R.L. Slingerland. 1993. Erosional dynamics, flexural isostasy and long- lived escarpments: A numerical modeling study. J. Geophys Res. (Paper Submitted) Tucker, G.E. and R. Slingerland. 1993. A model of regional-scale denudation: Results from verification studies. In: Third Int. Geomorphology Con. Program with Abstracts. Aug. 23-28, 1993. Hamilton, Ontario. p. 262. Wang, S., B.A. Albrecht, and P. Minnis. 1993. A regional simulation of marine boundary-layer clouds. J. Atmos. Sci. 50:4022-4043. Yarnal, B. 1993a. Synoptic Climatology in Environmental Analysis. London: Belhaven Press. Yarnal, B. 1993b. Human dimensions of global environmental changes in the Susquehanna River Basin: A call for research. Pennsylvania Geographer XXX (2):19-34. Yarnal, B. and J.D. Draves. 1993 A synoptic climatology of stream flow and acidity. Climate Research 2:193-202. Yarnal, B., A.C. Comrie, M. Dilley, J.D. Draves, B.C. Hewitson,and K.B.Yelsey. 1993. On choosing the ÒbestÓ classification procedure in synoptic climatology. International Journal of Climatology. (Paper Submitted) Zahn, S.G. 1993. An investigation of warm-wloud microphysics using a multi-component cloud model: Interactive effects of the aerosol spectrum. Unpublished M.S. thesis. Dept. of Meteorology. The Pennsylvania State University. 107 p. Zahn, S.G., D. Lamb, and J.P. Chen. 1993. A modeling study of cloud drop coalescence and its impact on in-cloud chemistry and deposition. Paper A248, 87th Annual Mtg. Air and Waste Mgmt. Assoc. 19-24 June, 1994. Cincinnati, OH. (In Preparation) Zhang, D. and R.A. Anthes. 1982. A high resolution model of the planetary boundary layer - sensitivity tests and comparisons with SESAME-79 data. J. Appl. Met. 21:1594-1609. EOS CONTRIBUTIONS 1992-1993 Ackerman, T.P., B.A. Albrecht, M.A. Miller, E. Clothiaux, R. Peters, and W. Syrett. 1993. Remote sensing of cloud properties using a 94 GHz radar. In: Topical Symp. on Combined Optical-Microwave Earth and Atmosphere Sensing. March 22-25. Albuquerque, NM. pp. 211-214. Albrecht, B.A. 1993. Effects of precipitation on the thermodynamic structure of the trrade wind boundary layer. J. Geophys. Res. 98:7327-7337. Barron, E.J. 1993. Water and global change. Presentation at Ann. Meet. Amer. Met. Soc., Anaheim , CA. Barron, E.J. 1993. Addressing critical issues of climate and hydrology from space. Trans. Am. Geophys. Union, San Francisco, CA. (Abstract Submitted) Bluth, G.J.S. and L.R. Kump. 1993. Lithologic and climatologic controls of river chemistry. Geochim. et. Cosmochim. Acta. (In Press) Capehart, W.J and T.N. Carlson. 1993a. Estimating near-surface soil moisture availability using a meteorologically driven soil water profile model. J. Hydrology. (Accepted for Publication) Capehart, W.J. and T.N. Carlson. 1993b. Estimation of surface moisture availability using a hydrology budget model aided by surface satellite observations and a soil- vegetation atmosphere transfer scheme (SVAT). 21st Con. on Ag. and Forest Met. San Diego, CA. (Abstract Submitted) Carlson, T.N. and R.R. Gillies. 1993. A physical approach for inverting vegetation index with surface radiometric temperature to estimate surface soil water content. Proc. Workshop on Thermal Remote Sensing of the Energy and Water Balance over Vegetation in Conjunction with Other Sensors. (In Preparation) Carlson, T.N., R.R. Gillies, and E.M. Perry. 1993. A method to make use of thermal infrared temperature and NDVI measurements to infer surface soil water content and fractional vegetation cover. Remote Sensing Reviews -- Special Issue on Recent Advances in Remote Sensing Science. (Accepted for Publication) Carlson, T.N. and A.Olioso.1993. Radiometric temperature response of soybean and corn leaves to water stress. Proc. Workshop on Thermal Remote Sensing of the Energy and Water Balance over Vegetation in Conjunction with Other Sensors. (In Preparation) Chamberlain, R. 1993. Fragipans and land use in the Mahantango Creek Intensive Study Area. Unpublished working paper. Dept. of Geography. The Pennsylvania State University. Chen, J.P. 1992. Numerical simulations of the redistribution of atmospheric trace chemicals through cloud processes. Unpublished Ph.D. dissertation. Dept. of Meteorology, The Pennsylvania State University. 342 p. Chen, J.P. and D. Lamb. 1993. Simulation of cloud microphysical and chemical processes using a multicomponent framework. Part I: Description of the microphysical model. J. Atmos. Sci. (In Press) Chen, J.P. and D. Lamb. 1993. The theoretical basis for the parameterization of ice crystal habits: Growth by vapor deposition. J. Atmos. Sci. (In Press) Christy, J.R. and S.J. Drouilhet. 1993. Variability in daily, lower zonal mean lower stratospheric temperatures. J. Climate. (In press) Christy, J.R. and R.W. Spencer. 1993. Monitoring global temperature from satellites. Storm 1: 18-24. Christy, J.R. 1992. Monitoring global temperature changes from satellites (Chapter 11). Global Climate Change: Implications, Challenges and Mitigation Measures. S.K. Majumdar et al. eds. The Pennsylvania Academy of Science. 566 pp. Dutton, E.G. and J.R. Christy. 1992. Solar radiative forcing at selected locations and evidence for global lower tropospheric cooling following the eruptions of El Chichon and Pinatubo. Geophys. Res. Lett. 19:2313-2316. Fitzjarrald, D., F. Robertson, E. Barron, S. Thompson, and D. Pollard. 1994. Simulated interannual variability in the hydrologic cycle over North America. Ann. Meet. Am. Met. Soc., Nashville, TN. (Abstract Submitted). Gillies, R.R. 1993. A physically based land-use classification scheme using remote solar and thermal infrared measurements suitable for describing urbanization. Unpublished Ph.D. dissertation. Dept. of Architecture. University of Newcastle, UK. Gillies, R.R. and T.N. Carlson. 1993a. Thermal remote sensing of surface soil water content with partial vegetation cover for incorporation into mesoscale prediction models. (Submitted to J. Appl. Met.) Gillies, R.R. and T.N. Carlson. 1993b. Translation of satellite measurements to land surface parameters. Proc. ERDAS Northern Regional User's Group Meeting. Gillies, R.R. and T.N. Carlson. 1993c. A physically based modeling approach for including remotely derived measurements in the study of land use change. Effects of Human-Induced Changes on Hydrological Systems. American Water Resources Association (AWRA) Summer Symposium. (Abstract Submitted) Gillies, R.R., A. Olioso, and K. Humes. 1993. (Ed's). Proc. Workshop on Thermal Remote Sensing of the Energy and Water Balance over Vegetation in Conjunction with Other Sensors. (In Preparation) Guerrero, T. 1992. Linking a general circulation model and a mesoscale model to examine the effects of model resolution on a simulated last glacial maximum storm in the western North Atlantic Ocean. Unpublished M.S. thesis. Department of Geosciences, The Pennsylvania State University. 233p. Hewitson, B.C. and R.G. Crane. 1993. Precipitation controls in southern Mexico. In: Hewitson, B.C. and R.G. Crane (Eds.). Neural Nets: Applications for Geography. Kluwer Academic Pub., Dordrecht, Holland. (In Press). Johnson, D.L. and A.C. Miller. 1993. A hydrologic model for use with raster data. Effects of Human-Induced Changes on Hydrological Systems. American Water Resources Association (AWRA) Summer Symposium. (Abstract Submitted) Kapsner, W.R. 1993. Response of snow accumulation to temperature variations in Central Greenland. Trans. Amer. Geophys. Union, San Fransisco, CA. (Abstract Submitted) Kasper, W.J. 1993. Modeling Spatio-Temporal Patterns of Nitrous Oxide and Methane Soil-Gas Fluxes on a Farmstead. Unpublished M.S. thesis. Dept. of Geography. The Pennsylvania State University. Kasper, W.J., E.P. Arabas, C.J. Rosin, L.L. Diamond, R.A. Heidl, and A.S. Hertz. 1993. The human dimensions of water-quality changes in the Susquehanna River Basin. Unpublished working paper. Dept. of Geography. The Pennsylvania State University. Lakhtakia, M.N. and T.T. Warner. 1993. A comparison of simple and complex treatments of surface hydrology and thermodynamics suitable for mesoscale atmospheric models. Mon. Wea. Rev. (Accepted for Publication) Lakhtakia, M.N. 1993. The use of a complex treatment of surface hydrology and thermodynamics within a mesoscale model and some related issues. Colloquium and Workshop on Multiscale Coupled Modeling. NASA. February 22-25. Calverton, MD. Lakhtakia, M.N., D.A. Miller, R.A. White, and C.B. Smith. 1993. GIS as an integrative tool in climatologic and hydrologic modeling. Second Int. Con./Workshop on Integrating Geographic Information Systems and Environmental Modeling, September 26-30. Breckenridge, CO. Lakhtakia, M.N.1992. BATS in the MM4 model and a procedure to obtain initial soil- water content for the model when BATS is used. The Second Penn State / NCAR Mesoscale Model User's Workshop. October 20-22. Boulder, CO. Mace, G.G., T.P. Ackerman, and E. Clothiaux. 1993. Mesoscale diagnostic quantities from arrays of doppler wind profilers and radiosondes. In: Preprints of the Fourth Symposium on Global Change Studies. January 17-22. Anaheim, CA. pp. 232- 234. Mace, G.G. and T.P. Ackerman. 1993a. Cirrus cloud development in a mobile upper tropospheric trough: The November 26th FIRE cirrus case study. Proc. of the FIRE Cirrus Science Con. June 14-17, 1993. Breckenridge, CO. Mace, G.G. and T.P. Ackerman. 1993b. Examination of the observed synoptic scale Cirrus cloud environment: The December 4-6 FIRE Cirrus case study. Proc. of the FIRE Cirrus Science Con. June 14-17, 1993. Breckenridge, CO. McGinnis, D. 1993. Predicting snowfall from synoptic circulation: A comparison of linear regression and neural network methodologies. In: Hewitson, B.C. and R.G. Crane (Eds.). Neural Nets: Applications for Geography. Kluwer Academic Pub., Dordrecht, Holland. (In Press) Miller, A.C. and D.L. Johnson. 1993. Formulation of a hydrologic model for use with remotely sensed data. Proc. Amer. Soc. Civ. Eng. - 1993 Int. Symp. on Eng. Hydro., San Francisco, CA. Robertson, F.R. and E.W. McCaul. 1993. Large scale structure of water vapor and condensate over the TOGA COARE Region. 6th Amer. Met. Soc. Con. on Climate Variations. January 23-28, 1994 Nashville, TN. (Abstract Submitted) Robertson, F.R., E.J. Barron, S. Goodman, D. Fitzgerald, B. Bishop, J. Christy, S. Thompson, and D. Pollard. 1993. GENESIS climate model: Intercomparsions with multiple climate data bases. In: Preprints Fourth Symp. on Global Change Stud. Jan. 17-22, 1993 Anaheim, CA. Amer. Met. Soc. pp. 3-8. Shuman, R.B. Alley, and S. Anadakrishnan. 1993. Characterization of a hoar- development episode using SSM/I brightness temperatures in the vicinity of the GISP2 site, Greenland. Ann. Glaciol. 17:183-188. Shuman, C.A. and R.B. Alley. 1993. Spatial and temporal characterization of hoar formation in Central Greenland using SSM/I brightness temperatures. Geophys. Res. Letters. (In Press) Smith, C.B. 1993. Initialization of soil-water-content for regional-scale atmospheric prediction models. Unpublished M.S. thesis. Dept. of Meteorology. The Pennsylvania State University. 87 p. Smith, C.B., M.N. Lakhtakia, W.J. Capehart , and T.N. Carlson. 1993a. Initialization of soil-water content for regional-scale atmospheric prediction models. Con. on Hydroclimatology: Land-Surface/Atmosphere Intereactions on Global and Regional Scales. Amer.Met. Soc. January 17-22. Anaheim, CA. pp.24 - 27. Smith, C.B., M.N. Lakhtakia, W.J. Capehart, and T.N. Carlson. 1993b. Initialization of soil-water content in regional-scale atmospheric prediction models. Bull. of Amer. Met. Soc. (Paper Submitted) Spencer, R.W. and J.R. Christy. 1993. Precision lower stratopsheric temperature monitoring with the MSU: Validation and results 1979-91. J. Climate 6: 1194- 1204. Tucker, G.E. and R.L. Slingerland. 1993. Erosional dynamics, flexural isostasy and long- lived escarpments: A numerical modeling study. J. Geophys Res. (Paper Submitted) Tucker, G.E. and R. Slingerland. 1993. A model of regional-scale denudation: Results from verification studies. In: Third Int. Geomorphology Con. Program with Abstracts. Aug. 23-28, 1993. Hamilton, Ontario. p. 262. Uttal, T., S.M. Shaver, E.E. Clothiaux, and T.P. Ackerman. 1993. Cloud boundaries during FIRE II. In: Proc. 26th Int. Con. on Radar Met. Amer. Met. Soc. May 24- 28. Norman, OK. pp. 573-575. Wang, S., B.A. Albrecht, and P. Minnis. 1993. A regional simulation of marine boundary-layer clouds. J. Atmos. Sci. 50:4022-4043. Wang, S. and Q. Wang. 1993. Roles of drizzle in a third-order turbulence closure model of the nocturnal stratus-topped marine boundary layer. J. Atmos. Sci. (In Press) Wang, S. 1993. Regional simulations of marine boundary layer clouds during ASTEX. Ann. Mtg. Amer. Geophys. Union. Baltimore, MD. Wang, S. 1993. Comparison of clouds derived from a regional boundary layer model, satellite data and ECMWF analysis for ASTEX period. Eighth Con. on Atmospheric Radiation, 23-28 January 1994. Nashville, TN. Warner, T.T., Y.-H. Kuo, J.D. Doyle, J. Dudhia, D.R. Stauffer, and N.L. Seaman. 1992. Nonhydrostatic, mesobeta-scale, real data simulations with the Penn State University/National Center for Atmospheric Research Mesoscale Model. Met. and Atmos. Phys. 49:209-227. Warner, T.T. and M.N. Lakhtakia. 1993. Simulation of mesoclimate with regional meteorological models. Directory of CRAY Sponsored University Research & Development Grants. CRAY Research, Inc. Yarnal, B., 1993a. Synoptic Climatology in Environmental Analysis. London: Belhaven Press. Yarnal, B. 1993b. Human dimensions of global environmental changes in the Susquehanna River Basin: A call for research. Pennsylvania Geographer XXX (2):19-34. Yarnal, B. and J.D. Draves. 1993 A synoptic climatology of stream flow and acidity. Climate Research 2:193-202. Yarnal, B., A.C. Comrie, M. Dilley, J.D. Draves, B.C. Hewitson,and K.B.Yelsey. 1993. On choosing the ÒbestÓ classification procedure in synoptic climatology. International Journal of Climatology. (Paper Submitted) Zahn, S.G. 1993. An investigation of warm-wloud microphysics using a multi-component cloud model: Interactive effects of the aerosol spectrum. Unpublished M.S. thesis. Dept. of Meteorology. The Pennsylvania State University. 107 p. Zahn, S.G., D. Lamb, and J.P. Chen. 1993. A modeling study of cloud drop coalescence and its impact on in-cloud chemistry and deposition. Paper A248, 87th Annual Mtg. Air and Waste Mgmt. Assoc. 19-24 June, 1994. Cincinnati, OH. (In Preparation)