^Meteorology and Physical Oceanography^Avissar, Roni^Avissar, Roni^DE-FG02-92ER61453d the southern hemisphere extratropic213^1^Zeng,Fanrong^1994^6^A Large-Eddy Simulation Study of the Convective Boundary Layer in Complex Terrain: Implications for Subgrid-Scale Parameterizations in GCMs^M.S.^Rutgers University^31^Thesis^^^Computer Hardware, Advanced Mathematics, and Model Physics (CHAMMP)^^^^^^^389^^^^^^^^^^^^Meteorology and Physical Oceanography^Avissar, Roni^Avissar, Roni^DE-FG02-9S.^Rutgers University^39^Thesis^^^Computer Hardware, Advanced Mathematics, and Model Physics (CHAMMP)^^^^^^^387^^^^^^^^^^^matics, and Model Physics (CHAMMP)^^^^^^^383^^^^^^^^^^^^Meteorology and Physical Oceanography^Avissar, Roni^Avissar, Roni^DE-FG02-92ER61453rsity^61^Thesis^^^Carbon, Climate, and Vegetation^^^^^^^4^^^^^^^^^^^^Agricultural Sciences^Biswas, P.K.^212^1^Li,Bin^1993^6^The Impact of the Spatial Variability of Land-Surface Characteristics on Land-Surface Energy Fluxes^M.rs Are of Greatest Importance for Atmospheric Models^M.S.^Rutgers University^78^Thesis^^^Computer Hardware, Advanced Mathe y^M.S.^Colorado State University^98^Thesis^^^Atmospheric Radiation Measurement (ARM)^^^^^^^153^^^^^^^^^^^^Atmospheric Science^Stephens, Graeme L.^Stephens, Graeme L.^DE-FG03-94ER61748^rbon, Climate, and Vegetation^^^^^^^6^^^^^^^^^^^^Botany^Str 210^1^Collins,Dan C^1992^6^An Evaluation with the Fourier Amplitude Sensitivity Test (FAST) of Which Land-Surface Paramete 93^1^Schneider,Timothy L^1994^6^Backscattering by Nonspherical Ice Particles at Millimeter Wavelengths: A Theoretical Studs^Ph.D.^Colorado State University^202^Dissertation^^^National Institute for Global Environmental Change (NIGEC)^^^^^^^151^^^^^^^^^^^^Atmospheric Science^Stephens, Graeme L.^Stephens, Graeme L.^DE-FC03-90ER61010^535^ation^^^Carbon, Climate, anA^152^The role of ice clouds in the hydrological cycle is uncertain. As a direct consequence, we do not fully understand t=92^1^Duda,David P^1994^6^Macrophysical and Microphysical Influences on Radiative Transfer in Two Dimensional Marine Stratu@mation (IPA) calculations of the same cross section. This matches the findings from Cahalan et al. (1994) for monochromati'c calculations in a simple cloud model, and suggests that accurate computations of domain averaged albedo in unbroken mari*ne stratus can be made using IPA calculations with one dimensional radiative transfer models.SєyZǑjK[AS6^1^Zhou,Chuan^1994^6^Spectral Correction of the Multi-Filter Rotating Shadowband Radiometer Shortwave Sensor^M.S.^State UVniversity of New York at Albany^58+^Thesis^^^Atmospheric Radiation Measurement (ARM)^^^^^^^12^^^^^^^^^^^^Atmospheric Scien^ces^Michalsky, Joseph, and Lee Harrison^Michalsky, Joseph, and Lee Harrison^DE-FG02-90ER61072<Is`F&8aA^11^The multifilter rotating shadowband radiometer (MFRSR) measures total shortwave irradiance with an unfiltered silicono photodiode. The largest error terms in pyranometer measurements, in general, are caused by angular response and temperaturre response imperfections. In the MFRSR the temperature is held at a constant value, and the angular response deviation from ideal is measured before deployment and measurements are corrected accordingly. The main error in silicon photodiode radiometry arises because of the mismatch of spectral responses between the ideal wavelength-independent response sensor and the wavelength- dependent response of the silicon cell. To evaluate this error, total horizontal, diffuse horizontal, and direct normal irradiance measurements made with an absolute cavity radiometer-calibrated thermopile pyrheliometer and tra cking disk diffuse pyranometer are compared to the silicon response measurements made with an MFRSR. The root-mean-square !errors for the MFRSR without spectral corrections are significantly improved compared with standard silicon cell pyranomet"er with a 60% improvement in direct normal and a 55% improvement in diffuse horizontal irradiance. The spectral correction#s provide a relatively small additional improvement of 15 to 20% in the diffuse and total horizontal irradiances. Further,$ it is found that for five-minute measurements 40%, 30%, and 20% of the remaining difference in direct normal, total horiz%ontal, and diffuse horizontal irradiance measurements, respectively, arises from the difference in the time response of th&e silicon cell sensors (seconds) and thermopile sensors (seconds). Consequently, silicon cell pyranometry with root-mean-square errors of less that 10 W/m{+2} in each component are possible.뗌ڣŝ>wi?*EݩL]E%?}|y,(10^1^Benkovitz,Carmen M^1994^6^Development and Testing of a High-Resolution Model for Tropospheric Sulfate Driven by Obser)vation-Derived Meteorology^Ph.D.^New York University^265^Dissertation^^^Atmospheric Sciences^^^^^^^20^^^^^^^^^^^^Earth Sys/tem Science^Hoffert, Martin I.^Schwartz, Stephen E.^DE-AC02-76CH00016CH00016ion Measurement (ARM)n7 Aq|<2+A^19^A high-resolution three-dimensional Eulerian transport and transformation model has been developed to simulate concen,trations of tropospheric sulfate for specific times and locations; it was applied over the North Atlantic and adjacent con-tinental regions during October and November, 1986. The model represents emissions of anthropogenic SO{-2} and sulfate and. of biogenic sulfur species, horizontal and vertical transport, gas-phase oxidation of SO{-2} and dimethylsulfide, aqueous/-phase oxidation of SO{-2}, and wet and dry deposition of SO{-2}, sulfate, and methanesulfonic acid (MSA). The meteorologiS0cal driver is the 6-hour output from the forecast model of the European Centre for Medium-Range Weather Forecasts. CalculaU1ted sulfate concentrations and column burdens, examined in detail for October 15 and October 22 at 6Z, are related to exisX2ting weather patterns. These results exhibit rich temporal and spatial structure; the characteristic (1/e) temporal autocoi3rrelation time for the sulfate column burdens over the central North Atlantic averages 10 hours; 95% of the values were 25l4 hours or less. The characteristic distance of spatial autocorrelation over this region depends on direction and averages {51600 km; with 10th percentile value of 400 km and 90th percentile value of 1700 km. Daily average model sulfate concentrat~6ions at the lowest vertical level accurately represent the spatial variability, temporal episodicity, and absolute magnitu7des of surface concentrations measured by monitoring stations in Europe, Canada and Barbados; over 50% of model concentrat8ions are within a factor of 3 of the observations. Over 50% of weekly model sulfate wet deposition amounts are within a fa9ctor of 3 of observations from US monitoring stations. Over the 34-day period modeled, contributions from anthropogenic em:issions to the sulfate over the mid North Atlantic Ocean ranged from 44 to 66%, contributions from biogenic emissions were; between 6 and 12%. Calculated average yields for sulfate (47 to 72%) and MSA (13%), and turnover times for SO{-2} (2 to 3< days) and sulfate and MSA (4 to 8 days) are comparable to previous estimates; however, these quantities depend on meteorological conditions and on the geographic and vertical distributions.#c4jM 2[{5[D0U  @5S\8.^9KF$‰>9^1^Angevine,Wayne^1993^6^Heat and Momentum Flux Measurements in the Planetary Boundary Layer with a Wind Profiling Radar/?Radio Acoustic Sounding System^Ph.D.^University of Colorado^85^Dissertation^^^Atmospheric Radiation Measurement (ARM)^^^^^^^18^^^^^^^^^^^^Electrical and Computer Engineering^Avery, Susan^Gage, Ken^DE-AI06-91RL12090, DE-AI06-91RL12092RL12092W;AA^17^The planetary or atmospheric boundary layer is the lowest 100-2000 m of the atmosphere, and contains the sources of mBost energy and pollutants that affect the entire atmosphere. Boundary layer structure and dynamics are key to understandinCg, modeling, and predicting climate, weather and pollution. New and improved techniques for measuring the dynamics of the Dboundary layer are needed. Existing methods involving towers, surface measurements, or aircraft have limited flexibility oEr are expensive. This dissertation describes methods for measuring heat flux (virtual temperature flux) and momentum flux Fin convective boundary layers. The instrument used is the 915 MHz boundary layer radar wind profiler Radio Acoustic SoundiGng System (profiler/RASS). As the name suggests, the profiler was developed to make wind measurements, and this work is anH expansion of its capabilities. The radio acoustic sounding system uses the profiler and attachments to measure virtual teImperature. Before the profiler/RASS could be used to make flux measurements, the ability to make simultaneous measurementsJ of wind velocity and temperature had to be developed. A feasibility study was conducted in Platteville, CO in June 1991 tKo determine if flux measurements were a practical application of the profiler, with encouraging results. The Rural OxidantLs in the Southern Environment II (ROSE II) experiment in Alabama in June 1992 provided the opportunity to compare flux meaiMsurements from the profiler/RASS to those made by a surface sonic anemometer and an aircraft, The National Center for AtmokNspheric Research (NCAR) King Air. The results indicate that the profiler/RASS is capable of making heat flux measurements zOthat compare well with aircraft and surface measurements. The primary limitation on the precision of the measurements is t}Phe sampling uncertainty of the turbulence, a limitation that is the same for any fixed instrument. The feasibility of makiQng momentum flux measurements needs to be explored further in more favorable conditions. In the course of analyzing the flRux measurements, methods of determining the mixing depth (boundary layer height) and improved data quality control techniques were developed.IRTYN\kzLIfN\΋' k'jܓ#=Q6rUԜ2D6Q]m >zY8ԤmL 7LA@o"$ p13^1^Pulwarty,Roger S^1994^6^Annual and Interannual Variability of Convection over Tropical South America: 1980-1989^Ph.D.q^University of Colorado^220^Dissertation^^^Carbon, Climate, and Vegetation^^^^^^^26^^^^^^^^^^^^Geography^Barry, Roger G.^Bradley, Raymond S., and Henry F. Diaz^DE-FG02-ER69017^R69017^dᥖێ2V mBmLjNqX4L'3 &tsA^25^Analyses of temporal (and spatial) variability in a wide range of ocean and atmospheric indices suggest that the annutal cycle plays a central role in virtually all of the recognized principal modes of large-scale climatic variability (intruaseasonal variability, tropospheric quasi-biennial oscillation, El Nio-Southern Oscillation (ENSO) events). The major resvearch objective of this thesis is a study of the nature and controls of processes, associated with variability in convectiwon on these timescales, over the region of the South American 'monsoon'. Variability on multiple timescales are examined uxsing empirical orthogonal functions, lagged correlations and singular spectrum analysis. Unique aspects of the annual cyclye of convection over tropical South America are documented in detail. The study shows that the controls of convection and zcirculation differ from the Northern Hemisphere summer maximum, over Panama and the eastern tropical Pacific, to that in t{he Amazon Basin (Southern Hemisphere maximum). Changes in the onset, longevity, and trajectory of key convective centers, |associated with the annual cycle over different regions of tropical South America, appear to be modulated by sea surface t}emperature variations and gradients in the eastern Pacific and tropical Atlantic, location of the intertropical convergenc~e zone, phase of the-quasi-biennial oscillation etc. In particular, the southward and westward displacement of these centers, during strong ENSO 'warm' events, occur with widespread dryness over much of continental South America. Conversely, during strong 'cold' events, convection is restricted to locations north and eastward of the mean, resulting in higher rainfall over northern South America and Panama. The seasonal dependence of these influences, on climate and circulation over the region, is documented in detail. Interannual variability and trends are documented for rainfall over the central Amazon Basin, and for runoff in the Rio Negro/Amazon and Rio Parana/Paraguay river basins. Interactions and their evolution can lead to observed anomalies in circulation and seasonal precipitation, not always associated with ENSO, suggesting that 'wet' or 'dry' periods result from enhancement or suppression of the annual cycle. Further work is expected to include a study of the influence of ENSO events, with different development characteristics, on the seasonal cycle.hwzd+BC 2 }Q14^1^Hu,Yong-Xiang^1994^6^A Study of the Link Between Cloud Microphysics and Climate Change^Ph.D.^University of Alaska - Fairbanks^121^Dissertation^^^Quantitative Links^^^^^^^28^^^^^^^^^^^^Atmospheric Sciences, Geophysical Institute^Stamnes, KnNut^Stamnes, Knut^DE-FG06-90ER61020|u&J,ԁGjݤק%fbKZ}xXY`͘+62ǒ›EF[&ɚvA^27^The climate related cloud radiative properties (fluxes and heating rates) are found to be determined by the second and the third moments of the cloud droplet size distribution. The detailed distribution of the cloud droplet size is unnecessary to obtain for climate purposes. An accurate parameterization of cloud optical properties suitable for climate models is developed. A new radiative-convective model has been developed and used for studying cloud-climate interactions. The energy balance at the Earth-atmosphere interface is treated in a self-consistent manner which avoids artificial tuning. The cloud radiative properties are accurately incorporated and are suitable for sensitivity studies of cloud-radiation-climate interactions. A sensitivity study of role of the cloud microphysical properties in the climate system is performed by studying the impact of cloud radiative forcing on the equilibrium state temperature. The cloud equivalent radius is found to be a very important variable in the climate system. A climate sensitivity study is performed to highlight the important role of the cloud absorption. An adjoint radiative transfer method is developed for use in the cloudy and aerosol-loaded atmospheres. The physical meaning of the adjoint radiative properties are discussed. The method is expected to be useful in climate modeling and remote sensing studies. A preliminary study of the atmospheric irreversibility is performed to elucidate the connection between cloud microphysical properties and the macrophysical direction of global climate. A variational principle which describes the macrophysical character of the climate system is established.1v}qVȬZx¼L6he role of ice clouds in the atmospheric energy balance. It is therefore necessary to quantify the existence of ice clouds chemical analysis. S^1985^6^The Use of Climate Classification Schemes to Assess the Performance of General Circulation Mlla^70^Thesis^^^Quantitative Links^^^^^^^148^^^^^^^^^^^^Mechanical Engineering^Alofs, Darryl John^Williams, Allen^DE-FG02-A^139^Relative to aboveground responses of crops to elevated atmospheric CO{-2}, data on belowground responses are scarce. Furthermore, little data under field conditions have been collected. Therefore, field studies were undertaken for two yea"rs to investigate the effect of CO{-2} concentration (360 and 550 mol mol{+-1} and soil water regime [wet, 100% of evapotA^150^Recent estimates of the effects of increasing amounts of anthropogenic sulfate aerosol on global climate have indicated that its impact on the radiative forcing of the atmosphere may be comparable in magnitude to the effect from increases in CO{-2}. Much of this impact is expected from the indirect effects of the aerosol on global cloud microphysics and the =subsequent impact on cloud albedo. However, internal horizontal variations in cloud optical properties are also known to a@ffect cloud albedo and reflectance, and thus affect the remote sensing of cloud microphysics. A broadband solar radiative Stransfer model (SHSG) capable of simulating the radiances and fluxes in a medium that varies both vertically and horizontaVlly was used to quantify the effects of both cloud microphysical changes and cloud inhomogeneity changes on the radiative properties of marine stratus. Two dimensional cross sections of cloud physics data taken from a set of three dimensional RAMS/LES simulations of marine stratus provided realistic optical property data for radiative transfer simulations. Along with a control run using typical marine CCN concentrations, two studies using enhanced concentrations of CCN were examined. The results of the radiative transfer calculations indicated that in unbroken marine stratus clouds the _net_ horizontal transport of photons over a domain of a few km was nearly zero, and the domain average broadband albedo computed in a two dimensional cross section was nearly identical to the domain average calculated from a series of independent pixel approxiA^111^The phosphorus cycle is simulated to determine the relative roles of dissolved and particulate organic matter in the "organic carbon pump". Experiments in which all of the new production is balanced by an export of fast-sinking particles xproduce the following features: (1) higher-than-observed phosphate concentrations below regions of high surface upwelling (nutrient trapping), (2) new production in excess of primary production in the Southern Ocean, (3) regions where vertically-integrated new production is equal to zero, and (4) lower-than-observed phosphate concentration in the main thermocline. Nutrient trapping and low thermocline phosphate are especially severe when a particle-flux length scale consistent with sediment trap observations is used. When 80% of the new production is balanced by an export of DOM, and when this DOM has a lifetime which increases with depth, all of the above unrealistic features are either eliminated or become less prominent. This DOM model also reproduces fairly well meridional nitrate fluxes in the North Atlantic computed with inverse methods and the distribution of "new" DOC in the West Pacific, assuming a constant Redfield ratio of C:N:P = 105:16:1. It is suggested therefore, that DOM plays a major role in the downward transport of organic matter and that there is an as-yet undiscovered pool of DOP in the ocean. The oxygen cycle is simulated to determine the Redfield ratio of oxygen to phosphorus (_r_{-O2:P}). A model with _r_{-O2:P} = 138 predicts more realistically the distributions of AOU and denitrification than a model with _r_{-O2:P} = 172, suggesting that the classical value is closer to observed than the recent revision.with the A^136^The objective of this research is to build a sampling system to collect cloud condensation nuclei (CCN) in sufficient amounts for chemical analysis. The sampling system consists of three virtual impactors and two cloud chambers which are connected in series to collect a CCN sample within a critical supersaturation (S{-_c_}) range of 0.16% to 1.0%. The first cloud chamber, the haze chamber, subjects the air sample to 100% relative humidity in order to grow haze drops. The second virtual impactor then uses inertial separation to remove all particles with an S{-_c_} of less than 0.16%. The second impactor showed an S{-_c_} cutoff of approximately 0.16% although the sharpness of the cutoff was less than expected. The second cloud chamber, the continuous flow diffusion chamber (CFD), subjects the sample to a 1.0% supersaturation such as found in clouds, and removes only particulates with an S{-_c_} of less than 1.0%. The CFD showed an S{-_c_} cutoff of close to 1.5%. The slightly higher cutoff is likely due to temperature fluctuations within the CFD, but did not greatly affect the overall system performance. In general the sampling system gave an S{-_c_} collection range close to the expected although the overall efficiency was lower than expected, which results in a longer sampling time to collect sufficient sample for17^1^Oh,Jai-Ho^1989^6^Physically-based General Circulation Model Parameterization of Clouds and their Radiative Interaction^Ph.D.^Oregon State University^315^Dissertation^^^Carbon, Climate, and Vegetation^^^^^^^34^^^^^^^^^^^^Atmospheric Sciences^Schlesinger, Michael^Schlesinger, Michael^DE-AI02-93ER61589oundation grant ATM-8619459Pe7ոi-qdC*F<wA^33^A physically-based parameterization of clouds and their radiative interactions has been developed for a multilayer atmospheric general circulation model. The main purpose of this development is to enable the evaluation of cloud optical depth feedback, a feedback of potential importance for greenhouse-gas-induced climate change. In the cloud-radiation parameterization, cloud water is a prognostic variable; fractional cloud amount is predicted semi-prognostically; stratiform and cumuloform clouds can coexist in a vertical atmospheric column; clouds can exist either as liquid water, ice or mixed liquid water-ice clouds; the solar spectrum is subdivided into three bands, two for Rayleigh scattering and ozone absorption, and one for water vapor absorption which is further subdivided into six intervals for the k-distribution method; multiple scattering is calculated by the two-stream, delta-Eddington method; the longwave spectrum is divided into four regions based on the absorbers; the radiative properties of clouds depend on their predicted cloud liquid water and ice amounts; and the vertical distribution of clouds is considered to have maximum overlap for contiguous cloud layers and random overlap for noncontiguous cloud layers. The parameters for the stratiform-cloud parameterization and the values chosen are: (1) the characteristic time for the evaporation of cloud water, = 50 minutes; (2) the autoconversion rate of cloud water into precipitation, a = (3600 s){+-1}; (3) the threshold relative humidity of the environment at which condensation can begin, U{-00} = 0; and (4) the threshold cloud water above which precipitation can begin, m{-c,w} = 0.5x10{+-4} g/g for water clouds; m{-c,i} = 0.4xI{-c}(T{-c}) for ice clouds, where I{-c}(T{-c}) are the values given by Heymsfield and Platt (1984) as a function of the cloud temperature T{-c}; m{-c,i} = 1.2xI{-c}(T{-c}) for ice clouds formed from an antecedent cumuloform cloud. The parameters for the cumuloform-cloud parameterization and the values chosen are: (1) the characteristic time for the evaporation of cloud water, = 50 minutes; (2) the parameter in the diagnostic equation for the cloudiness in terms of the mass flux at the base of the detrainment layer, = 10; and (3) the conversion rate of cloud water into precipitation, C{-0} = 0.002 m{+-1}. In addition, two radiation parameters were determined: (1) the mass absorption coefficient of cumuloform ice clouds, a{-0}{+up-arrow,down-arrow} = 0.096; and the multiple-scattering asymmetry factor for cumuloform clouds, g = 0.85. An evaluation of the performance of the model shows that the cloud-radiation parameterization is capable of successfully simulating many features of the observed cloud cover, the vertically integrated cloud water, precipitation, and the earth's radiation budget at the top of the atmosphere. The model is less successful in reproducing the cloud optical depths as given by the International Satellite Could Climatology Project observations. This apparent contradiction suggests that these optical depth observations may not be useful in validating a cloud-radiation parameterization. The comparison of the simulated and observed climates also shows that the cloud parameterization is deficient in its reproduction of the subtropical stratocumulus clouds. To rectify this deficiency it is necessary to develop a model for stratocumulus clouds 18^1^Jiang,Xingjian^1991^6^Carbon Dioxide Induced Ocean Climatic Change and Tracer Experiment with an Atmosphere-Ocean General Circulation Model^Ph.D.^University of Illinois at Urbana-Champaign^304^Dissertation^^^Carbon, Climate, and Vegetation^^^^^^^36^^^^^^^^^^^^Atmospheric Sciences^Schlesinger, Michael^Schlesinger, Michael^DE-AI02-93ER61589oundation grants ATMA^35^The characteristic response time of the climate system ranges 10-100 years for a CO{-2} doubling due to the heat penetration into the deep ocean. This delay is of fundamental importance in establishing when the greenhouse signal will emerge from the background noise due to the natural variability of the climate system. It is imperative that we validate a model's estimate of the response time of the climate system against observations. If the penetration of the greenhouse-gas-induced heating into the ocean is analogous to that of a passive tracer, then the latter may be used as a surrogate for the former and thereby establish the characteristic response time. The principal objective of this study is to determine whether or not the penetration of a passive tracer is analogous to the penetration of a greenhouse-gas-induced heating. The Atmosphere-Ocean General Circulation Model (A-O GCM) developed at Oregon State University has been used to study CO{-2}-induced climate change and the penetration of passive tracers into the world ocean. The present climate and a 2xCO{-2} climate have been simulated. In this study, the passive tracers tritium, CFC-11, CFC-12 and a "passive CO{-2}-induced heating" are simulated. A rudimentary carbon-cycle model is also included for the purpose of a preliminary determination of the amount of atmospheric CO{-2} taken up by the ocean. The CO{-2}-induced active and passive warmings are larger in the subtropics and high latitudes than in the tropics. The largest difference between the active and passive CO{-2}-induced heatings occur in the North Atlantic deep ocean, with maximum cooling about -1.5 C for the active case in layer four of the ocean (1150 m). There is no hemispherically asymmetric warming as that found by Manabe et al. (1990) and Stouffer et al. (1990). The convective overturning and large-scale sinking motion are responsible for the large penetration of CO{-2}-induced warming in high latitudes. The CO{-2}-induced circulation changes show that the North Atlantic thermohaline circulation is signifi cantly weakened due to the penetration of CO{-2}-induced heating. Associated with this change, the strength of North Atlantic conveyor belt is reduced, which results in a large warming in the upper ocean and cooling in the deep layers. The characteristic response time ranges from 40-50 years for the active CO{-2}-induced climate change, and 70-160 years for passive CO{-2}-induced climate change. The physical processes controlling the geochemical tracer penetration are very similar to those for the CO{-2}-induced heating. Detailed comparisons between the penetrations of the CO{-2}-induced heating and passive tracers show that there is no a single tracer, whether tritium, CFC-11, CFC-12 or the passive CO{-2}-induced heating, which penetrates into the ocean exactly like the active CO{-2}-induced heating in terms of distribution, transport or physical process. The overall results indicate that CFCs may be the best candidate as a surrogate for the CO{-2}-induced oceanic climate study. The carbon-cycle study shows that the model is capable of simulating the observed ocean-atmosphere pCO{-2} difference (delta-pCO{-2}) in most areas. However, the model cannot simulate the observed magnitude and detailed geographical distribution of delta-pCO{-2} because they are also determined by the organic and biological processes which are neglected in the present study. The model-simulated ocean uptake of atmospheric CO{-2} is about 2.61 GtC per year, which is" about 38.65.6% of the anthropogenic atmospheric CO{-2} released during the last decade.,.6t<;t u%19^1^Entwistle,Clarence Bruce^1992^6^Analysis of the Nature of Zero Feedback in the Climate System Using a Multilayer Radi2ative-Convective Model^M.S.^University of Illinois at Urbana-Champaign^201^Thesis^^^Carbon, Climate, and Vegetation^^^^^^^538^^^^^^^^^^^^Atmospheric Sciences^Schlesinger, Michael^Schlesinger, Michael^DE-AI02-93ER61589oundation grant ATM-9001310OA^37^The radiative-transfer-model (RT) feedback analysis method [Manabe and Wetherald (1980), Wetherald and Manabe (1980, S1988)] described by Schlesinger (1989) can provide an economical means of ranking the feedback processes in general circulh ation models (GCMs). To test the accuracy of this method a radiative-convective model (RCM), which can provide a true meask ure of feedback, is used. However, results of the RT method using a two-layer RCM developed from the radiative transfer pa rameterization of the Oregon State University two-layer GCM raise fundamental questions concerning the definition and natu re of zero feedback. To answer these questions a multi-layer RCM was developed based upon the radiative transfer parameter ization of the University of Illinois multi-layer GCM (Oh, 1989). Introduction of forward time marching to the multi-layer radiative transfer code revealed the need for modification. Accordingly, the radiative transfer code was modified and evaluated in comparison with 37 other such models that participated in the Intercomparison of Radiation Codes in Climate Models (ICRCCM) study. To answer the fundamental questions on the nature of zero feedback in the climate system, the MLRCM was run for two perturbation experiments, a one percent increase in the solar constant and a doubling of the CO{-2} concentration. Results from these experiments show that the non-vertically resolving, zero-dimensional planetary energy balance model prediction of the zero-feedback surface temperature change due to a change in external forcing gives the correct answer for a vertically resolving, one-dimensional model, provided that the change in planetary emissivity as the climate systemP responds to that forcing is included. Also, it is concluded that the zero-feedback climate system includes the processes of radiative transfer, surface convection and temperature-independent convective adjustment in the free atmosphere. TemperQature-dependent convective adjustment is not included, as it creates a feedback. Additionally, it is found that the vertical profile of the zero-feedback temperature change is not uniform with height, and depends upon the perturbation applied tRo the climate system.mH%^s5u:՞h/ |ԟ #WLn;hFlFpޯI  wVq3WCs6[#⏞.!|x20^1^Ramankutty,Navin^1994^6^An Empirical Estimate of Climate Sensitivity^M.S.^University of Illinois at Urbana-Champaign^I167^Thesis^^^Carbon, Climate, and Vegetation^^^^^^^40^^^^^^^^^^^^Atmospheric Sciences^Schlesinger, Michael^Schlesinger, Mi-chael^DE-AI02-93ER61589oundation grant ATM-9001310xgvCQ{ JUeTh떧SXup6kd(xے@7Wȵh⢟JA^39^The Intergovernmental Panel on Climate Change (IPCC) estimates the climate sensitivity to be in the range 1.5 to 4.5KC, with a best guess of 2.5C (Houghton et al., 1990, 1992), while Lindzen estimates delta-T{-2x} = 0.5C (Lindzen, 1990).K The IPCC estimate is based on equilibrium simulations using General Circulation Models (GCMs). This study presents an alt ernate method of estimating delta-T{-2x} using an empirical approach combining simple climate/ocean model simulations witho! observations. The same procedure is applied to narrow the range of uncertainty in the estimate of radiative forcing due tr"o some climate forcing agents. Our study yields an estimate of delta-T{-2x} = 2.9{-+2.3}{+-1.2}C, corresponding to estima#ted anthropogenic sulphate aerosol (ASA) forcing of delta-F{-SO4}(1978) = -0.6{--0.3}{++0.5} Wm{+-2}. This range of sensit$ivity is 17% larger than the IPCC range, but our estimated range of ASA radiative forcing is much smaller than the estimat%es from direct radiative-transfer calculations of -0.90.6 Wm{+-2}. Putative long-term solar-irradiance variations imply a& range of delta-T{-2x} = 1.6{-+3.6}{+-0.6}C, corresponding to a range of delta-F{-SO4}(1978) = -0.5{--0.6}{++0.4} Wm{+-2}'. A case is made for the plausibility of long-term solar-irradiance variations, but direct observational or physical evide(nce is still wanting. We use our simple model to detrend the observed temperature record and identify an oscillation of 65)-70-year timescale. The oscillation is confined to the North Atlantic and bounding Northern Hemisphere continental regions*. We believe that it is caused by internal ocean dynamics rather than external forcing. We have explored the implications +of volcanic aerosol forcing on global-mean surface temperatures. We estimate the cooling due to volcanic aerosols to be -1,.6{--1.5}{++0.8}C per unit aerosol optical depth. We also find evidence for cumulative long-term cooling due to frequent -occurrence of eruptions. However, this long-term cooling does not explain the warming-cooling-warming phases of the observ.ed temperatures from 1910-1970, as has been conjectured by Lamb (1970), Robock (1978, 1979) and Bryson and Goodman (1980). We find that climate sensitivity has a significant influence on the temperature response to volcanic forcing, contradicto021^1^Ackerman,Andrew S^1994^6^A Numerical Modeling Study of the Effects of Variations in Aerosol Concentrations on Stratif1orm Clouds in the Marine Boundary Layer^Ph.D.^University of Washington^230^Dissertation^^^Atmospheric Radiation Measurement (ARM)^^^^^^^42^^^^^^^^^^^^Atmospheric Sciences^Hobbs, Peter V.^Toon, Owen B., and Douglas L. Westphal^DE-AI03-94ER617433A^41^Marine stratiform clouds play an important role in the global radiative heat budget of the Earth because they overlie4 about a third of the oceans and they reflect much more sunlight than the ocean surface. The reflectivity of these clouds 5depends in part on the concentrations of cloud condensation nuclei (CCN) on which cloud droplets form. Here a numerical mo6del is used to investigate interactions between aerosol and cloud microphysics, radiative transfer, and turbulent mixing i7n the stratiform cloud-topped marine boundary layer. Results from model simulations are found to be in general agreement w8ith airborne measurements of marine stratocumulus clouds. However, the model underpredicts the concentrations of small clo+9ud droplets in the lower region of the cloud layer. This is consistent with the lack of a peak supersaturation near cloud -:base in the model results, which is attributable to horizontal averaging in the model. The model simulations indicate that; equilibrium CCN concentrations are sensitive to their formation rate. The times required to reach equilibrium were found tions than under the assumptions that cloud water is fixed and unactivated haze particles are ignored. Increased droplet cs?oncentrations generally (but not always) produce increased cloud water due to reduced drizzle. The number of haze particlev@s increases with droplet concentrations due to decreased peak supersaturations in the cloud. The model simulations show thAat when droplet collisions reduce droplet concentrations to extremely low values, a cloud layer can become so optically thBin that cloud-top radiative cooling is unable to drive vertical mixing. The stratocumulus-topped marine boundary layer canC then collapse to a shallow fog layer over the course of a day or more. The model was also used to investigate long-lived,D linear regions of enhanced cloud reflectivity that appear in satellite imagery downwind of ships. We have found that injeEctions of CCN, which are present in ship exhaust, can account for many of the observed properties of ship tracks. Ship traFcks are classified as Type 1, which are observed in visible satellite imagery, and Type 2, which are more common and are o~Gbserved in near-infrared imagery. The distinction between the two types is attributed to differences in ambient concentratHions of CCN that cause variations in turbulent mixing in the marine boundary layer, through the effect of cloud droplet concentrations on cloud-top longwave radiative cooling. C29c1U$C%`hԆRh<$!L !CN2`uDЃ`eq&d159^1^Li,Long^1994^6^The Global Atmospheric Response to Low Frequency Tropical Forcing^Ph.D.^University of California at D160^1^Ross,John Alan^1993^6^An Algebraic Stress Finite Element Model for Recirculating Turbulent Flows^Ph.D.^University of161^1^Nevison,Cynthia Dale^1994^6^A Model Analysis of the Spatial Distribution and Temporal Trends of N{-2}O Sources and Sgreater than 1.0 ppbv with maximum values of 1.7 ppbv. The PAN to O{-3} ratio relationship appears to be dependent on the ent, depending on the turnover time of the soil organic material. Using radiocarbon measurements and a simple model, I hav81^1^Medley,Steven D^1993^6^Collection of Cloud Condensation Nuclei for Chemical Analysis^M.S.^University of Missouri - Ro#es^^^^^^^307^^^^^^^^^^^^Chemistry^Zoller, William H.^Gaffney, Jeffrey S., and Nancy A. Marley^W-31-109-ENG-38, W-7405-ENG-'156^1^Ghan,Steven John^1988^6^Unstable Radiative-Dynamical Interactions^D.Sci.^Massachusetts Institute of Technology^180^T7157^1^Kinnison,Douglas Edward^1989^6^Effects of Trace Gases on Global Atmospheric Chemical and Physical Processes^Ph.D.^Un;158^1^Atherton,Cynthia Shaver^1994^6^Predicting Tropospheric Ozone and Hydroxyl Radical in a Global, Three-Dimensional, ChIT26^1^Daniels,RC^1991^6^An Internship with the Carbon Dioxide Information Analysis and Research Program at Oak Ridge NationLal Laboratory: The Effects of Greenhouse Gas Induced Sea Level Rise on the South Carolina Coast^M.A.^Miami University^174^^V27^1^Jastrow,Julie D^1994^6^Mechanisms of Aggregate Formation and Stabilization in Prairie Soils^Ph.D.^University of IllinaWois at Chicago^115^Dissertation^^^Ecosystems Function and Response^^^^^^^49^^^^^^^^^^^^Biological Sciences^Lussenhop, Johnl^Miller, R. Michael^W-31-109-ENG-38^7rJFUBfG9A %'pHܛaX0{A^=~ ?0'zZ/T/c:lFqTd X!pYA^48^Water-stable soil aggregates were studied in mollisols recovering from long-term cultivation. The general hypothesis Zinvestigated was that plant roots and fungal hyphae provide a mechanical framework for the formation of soil aggregates an[d, thereby, create the conditions necessary for the accrual of organic matter, which serves to further stabilize the aggre\gates. This hypothesis was investigated in three ways. First, the contributions of roots and the external hyphae of mycorr]hizal fungi to the formation and stabilization of macroaggregates (>0.25 mm in diameter) in restored tallgrass prairie wer^e shown to be greater than the effects of microbial biomass, water-soluble carbon, and total soil organic carbon by using _the statistical technique of path analysis. It was also demonstrated that the relative importance of different organic bin`ding mechanisms was generally a function of the physical size of the mechanism relative to the spatial scale of the aggregaate planes of weakness being bridged. Second, in a chronosequence of prairie restorations, the accrual of soil organic matbter was facilitated by, but lagged behind, the formation of macroaggregates. Inputs of particulate organic matter appearedc to be incorporated relatively rapidly into a mineral-associated fraction, which may be an important factor in aggregate sdtabilization. Third, carbon inputs and turnover rates in micro- and macroaggregates were compared by using the natural abuendance of stable carbon isotopes, which clearly demonstrated a much slower turnover of old carbon in microaggregates. Measfured increases in the net inputs of new carbon with increasing aggregate size confirmed the existence of an aggregate hiergarchy created by the binding of microaggregates into increasingly larger macroaggregates. Unexpectedly, the stable isotopeh data indicated that microaggregate formation and degradation may be more rapid than suggested by the apparent stability of microaggregates in cultivated soils or by the turnover rates for old carbon.*;DFՙxaXd &ꏻ?qbgp'pj28^1^Franson,Raymond L^1989^6^Mechanisms of Belowground Interactions among _Plantago lanceolata_, _Phleum pratense_ and _G klomus monosporum_^Ph.D.^University of Chicago^108^Dissertation^^^Ecosystems Function and Response^^^^^^^51^^^^^^^^^^^^Comm ittee on Evolutionary Biology^Miller, R. Michael^Miller, R. Michael^W-31-109-ENG-38^Px!J4Fv:}VIx> mA^50^A study of belowground interactions between two oldfield herbs was conducted to determine which of the many possible nmechanisms of interactions between plants occur and what is the relative importance and interrelationship of those that do 'o occur? Although mechanisms of plant interactions cannot be directly observed, the mechanisms by which plants are allowed pto interact can be manipulated and then the net result can be quantified. The experimental design is a complete factorial qcomposed of two plant species (_Plantago lanceolata_, _Phleum pratense_), two levels of fungal inoculation (addition of _Grlomus monosporum_, without addition of _G. monosporum_), four levels of belowground barrier between plants (no barrier, 20 *s micron mesh, 1 micron mesh, solid barrier), and three potential partners (_Plantago_, _Phleum_, no neighbor), and four re -tplicates. We found _Glomus_ partially alleviates the depressive effects of increasing density in _Plantago_ in monoculture ?u. _Phleum_ shows no response to _Glomus_ colonization. Without _Glomus_, _Plantago_ achieves a larger gain in biomass whil Cve _Phleum_ is suppressed only minimally. When _Glomus_ is added the net result of _Plantago-Phleum_ interactions changes f [wrom competitive to commensal. _Plantago_ achieves a higher growth rate while growing with _Phleum_ than it can when grown ^xalone while _Phleum_ achieves nearly the same biomass whether growing alone, in monoculture or with _Plantago_. This sugge mysts that _Phleum_ acts as a "root promoter" causing an increase in root mass, root length and specific root length of any qzplant growing with it. The use of meshes allows determination that all mechanisms of interaction in the experiment involved chemicals or microbes of less than 1 micron in diameter.RØC2s6]go69vNL6PQB`)T3ܕGX0O|29^1^Reinhardt,Debra R^1989^6^Growing Season Dynamics of Vesicular Arbuscular Mycorrhizal Fungi in Prairie Versus Pasture }Ecosystems^M.S.^University of Illinois at Chicago^42^Thesis^^^Ecosystems Function and Response^^^^^^^53^^^^^^^^^^^^Biologi cal Sciences^Lussenhop, John^Miller, R. Michael^W-31-109-ENG-38^5rz#@[]LmPsʯ:O;Pd2{Kÿ4 A^52^This study addresses the hypothesis that there is a community effect for levels of intraradical and extraradical vesi cular-arbuscular mycorrhizal fungal colonization of fibrous roots, based on general root morphology and phenology. Two gra ssland communities, a reconstructed prairie characterized by a predominance of warm-season grasses and an ungrazed pasture  composed of cool-season grasses, were chosen as the study sites because of their differences in root morphology as well as phenology. A community effect for the biomass of roots larger than 1.0 mm in diameter was found to occur. A significantl y greater amount of root biomass was in this size class in the prairie as compared to the pasture. However, the biomass of fibrous roots, i.e., roots less than 1.0 mm diameter, did not differ significantly between the two communities. Fibrous r oot length was significantly greater in the pasture than in the prairie due to greater lengths of medium (0.075-0.2 mm diapmeter) and fine (< 0.075 mm diameter) fibrous roots. Although a greater total length of total length of fibrous roots occu rred in the pasture community, the amount of VAM fungus as represented by total length of roots colonized was similar for the prairie and pasture communities (P = 0.34). However, the colonized lengths within the various size classes of fibrous roots differed between the two communities. Therefore, a pattern of fungal colonization which would explain the intraradical distribution among root size classes based on "susceptibility" of each size class to colonization was sought. A measure  of "susceptibility to colonization" by VAM fungi for each size class of root was determined by dividing the proportion of  total root length within a diameter size class by the proportion of total root length within that size class. These values demonstrated that the finer the fibrous root size class, the less susceptible it was to colonization. The susceptibility values for two of the three size classes were not significantly different for the two communities, indicating that the im portance of root diameter in determining level of colonization is similar in the two communities.|.ß[C p_Z'҄'Lѐd+B (:x!OwKn81KD,pWXpͬf$w}$T-1 sh1|30^1^Smith,William Scott^1994^6^A Study of the Cloud/Radiation Interaction Using a Second Order Turbulence Closure Radiative/Convective Model^Ph.D.^New Mexico State University at Las Cruces^283^Dissertation^^^Atmospheric Radiation Measurement ( ARM)^^^^^^^56^^^^^^^^^^^^Physics^Chen, T.-W., and C.-Y. J. Kao^Kao, C.-Y. Jim^W-7405-ENG-36^5-ENG-36^d$IZ+M$?8P A^55^A high resolution one-dimensional version of a second-order turbulence radiative/convective model, developed at Los A lamos National Laboratory, is used to simulate the diurnal cycle of marine stratocumulus clouds with respect to their inte raction with radiation and turbulent mixing. Particular attention is given to the role of shortwave heating in modulating the diurnal variation of the stratocumulus-capped boundary layer. The fidelity of the model to the underlying physics is a ssessed by comparing the model simulation against data taken at San Nicolas Island during the intensive field observation (IFO) of the First International Satellite Cloud Climatology Program (ISCCP) Regional Experiment (FIRE), conducted during June and July, 1987. The model is able to reproduce the observed diurnal cycle of the liquid water content, cloud base hei ght, radiative heating or cooling rates, and the mean and turbulence variables fairly well. The mechanisms which cause the diurnal variation and the decoupling of the boundary layer are examined.#dZiCנnmw_[)32>80q2 31^1^Crouzet,Yoan^1993^6^Equilibrium Vapor Pressure and Capillary Ring Formation of Adhering Spherical Aerosol Particles^M .S.^Texas A&M University^130^Thesis^^^Atmospheric Sciences^^^^^^^58^^^^^^^^^^^^Nuclear Engineering^Marlow, William H.^Marl ow, William H.^DE-FG02-91ER61207^*vB on0M[|WڬfJF~ `tZ ָEQ-牘8H FGt A^57^Liquid rings trapped by capillary condensation at the point of contact between two spherical particles present an int -eresting feature insofar as they often display negative curvatures, thereby inducing a lowering of the equilibrium vapor p 0ressures according to the Kelvin equation. This thesis work is divided up into two parts, and each part corresponds to a specific approach of the problem and different objectives. The first part of this work dealt with the calculations of the mean curvature of the meniscus for any two spherical particles in contact. Results were interpreted in terms of equilibrium vapor pressure and applied to condensation nuclei activation. Critical supersaturations were defined and compared to the predictions of the heterogeneous nucleation model for the case of water. A few applications were listed. The second part o Of this thesis work attempted a different approach by taking into consideration the van der Waals interaction forces acting R at a molecular level. The liquid was modeled as an assembly of one-nanometer liquid spheres, and a Monte Carlo code was d oeveloped in order to find the minimal energy configuration of the system. Energies were computed using Lifshitz theory bet rween macroscopic spheres. A simulation was performed for a specific case: liquid water and two hexadecane solid particles ~in contact. The final configuration of the assembly was first analyzed in order to estimate the contact angle; the interfa ce was then compared to the predictions of the previous approach. A very encouraging agreement was found on a qualitative level.AP鋪>УhA ^DюT2!$1vl%vlα&{wg; LIl笽>>CV=&k=|i3Siqa185^1^Kwon,Tae-Yong^1993^6^An Examination of the Time/Space Averaging Process Used in the Earth Radiation Budget Experiment^Ph.D.^State University of New York at Stony Brook^99^Dissertation^^^Carbon, Climate, and Vegetation^^^^^^^334^^^^^^^^^^^^Mechanical Engineering (Atmospheric Sciences)^Cess, Robert D.^Cess, Robert D.^DE-FG02-85ER60314K~طJF/\ l- 33^1^Steig,Eric Julien^1992^6^A Gas Extraction System for the Measurement of Carbon Dioxide and Carbon Isotopes in Polar I ce Cores^M.S.^University of Washington^52^Thesis^^^Graduate Fellowships for Global Change^^^^^^^61^^^^^^^^^^^^Geological S ciences^Stuiver, Minze^Wohlpart, A.^DE-AC05-76OR00033^76OR00033^\D굄S#L첬A.*Rʼjkou .4N+`F. A^60^Knowledge of the distribution of {+13}C in the glacial ocean, atmosphere, and biosphere is important to understandin g the causes of glacial/interglacial changes in atmospheric CO{-2} levels. Although, deep-ocean {+13}C values are well-co nstrained by ocean sediment studies, model-based estimates of changes in the carbon budget for the biosphere vary considerably. Measurement of atmospheric {+13}C in CO{-2} in ice cores will provide additional constraints on this budget and will also improve estimates of changes in the ocean surface layer {+13}C. Direct measurement of ancient atmospheric {+13}C can be accomplished through polar ice cores studies. I have designed and constructed a gas-extraction line for ice cores w ith particular attention to the specific difficulties of measuring {+13}C in CO{-2}. The ice is shaved, rather than crush ed, to minimize fractionation effects resulting from gas travel through long air-paths in the ice. To minimize the risk of  isotopic contamination and fractionation within the vacuum line, CO{-2} is separated immediately from the air; the CO{-2}  concentration is then measured by a simple pressure/volume comparison rather than by gas chromatography or spectroscopy. Measurements from Greenland ice core samples give an average value of 2802 ppm CO{-2} for pre-industrial samples, demonst -rating that the extraction system gives accurate, precise determinations of CO{-2} concentrations. Measurement of {+13}C 0from polar ice samples has not been achieved at this time. However, results on standard air samples demonstrate a precisio 9n for {+13}C of less than 0.2% at the 95% confidence level.w_03r/эJ_>)d7)8{ם@h%_yH =34^1^Shen,Karin Pauline^1992^6^A Microclimate Model to Investigate Greenhouse Warming of a Sub-Alpine Ecosystem^M.S.^Unive Hrsity of California at Berkeley^64^Thesis^^^Graduate Fellowships for Global Change^^^^^^^63^^^^^^^^^^^^Energy and Resource Ks Group^Harte, John^Wohlpart, A.^DE-AC05-76OR00033 > ?k'R ܈ LS<L_y;:S£B_6pρ,+t>A^62^Increasing concentrations of greenhouse gases in the earth's atmosphere are expected to result in a global warming of U several degrees Celsius in the coming decades. This warming will have far-reaching impacts on the biosphere, and while Ge gneral Circulation Models (GCMs) try to predict the magnitude and scope of the warming, there is little information regardi ing the potential impacts of greenhouse warming on natural systems. An experiment currently under way in a meadow in the Co vlorado Rocky Mountains attempts to investigate the many consequences of greenhouse warming for soil ecosystems. A mathemat xical model of the soil microclimate was developed to simulate the soil temperature and moisture content of the meadow. The model simulates both treatment and control scenarios so as to investigate the potential effects of warming. Results of mo del simulation studies indicate warmer, drier soils under treatment conditions showing the greatest temperature effects of warming occurring at night. These results could have several implications regarding the dynamics of the ecosystem, and future model studies will investigate these connections.qls3\ ˕_1+(/];3XQtSۄ{Hm/Df[E 35^1^Dean,Caryn Lyn^1993^6^Interactions Between a Tropical Mixed Boundary Layer and Cumulus Convection in a Radiative-Conv ective Model^M.S.^Pennsylvania State University^159^Thesis^^^Graduate Fellowships for Global Change^^^^^^^65^^^^^^^^^^^^Me teorology^Frank, William^Wohlpart, A.^DE-AC05-76OR00033033ש:¤h)I (X*,` Ԫ'-~`gd4IS"c A^64^A radiative-convective model, combining previously developed cumulus, stable cloud and radiation parameterizations wi th a boundary layer scheme, was developed in the current study. The cloud model (Frank and Cohen, 1985) was modified to in corporate the effects of both small and large clouds. The boundary layer model, adapted from a mixed layer model (Albrecht  et al. 1979), was only slightly modified to couple it with the more sophisticated cloud model. The model was tested for a  variety of imposed divergence profiles, which simulate the regions of the tropical ocean from approximately the Intertropical Convergence Zone (ITCZ) to the subtropical high region. The sounding used to initialize the model for most of the runs is from the trade wind region of ATEX. For each experiment, the model was run with a timestep of 300 seconds for a perio d of seven (7) days. The model produced an inversion height of approximately 950 mb for each region. As the subsidence inc reased, typical of conditions towards the subtropical high region, the boundary layer became moister and warmer, and the p !eriods of deep convection became weaker and less frequent. Various cloud spectra were also introduced into the model to te %st for sensitivity to the number of cloud types represented. The model produced similar results whether one cloud or many 1clouds were used. However, the runs with more than one cloud generally displayed resultant inversion heights with more sta 4ble oscillations than the equivalent runs with only one cloud type. With the exception of higher mixing rates resulting in 9 larger clouds, the specification of initial updraft vertical velocity and mixing rate made little difference to the final < inversion height and boundary layer thermodynamic values. Finally, sensitivity tests were done for varying sea surface te Lmperatures and surface wind velocities. The model responded well to changes. As sea surface temperature increased, the inv Oersion height increased, the boundary layer became warmer and the moisture stayed approximately the same. As surface wind hvelocity increased, the inversion height decreased, and the boundary layer temperature and moisture decreased and increase ld, respectively. A FGGE sounding from a region closer to the equator was tested and the results were very similar, with an  inversion height near 960 mb. Therefore, the model is thought to be physically representative of the physical area descri bed above.L\ft阨rrjELF&eqdNR.K9>u3ӥJ4Kޝb0$`@To!@J dE2<+agI|%A/JƐ! 36^1^Jensen,Deborah Bowne^1993^6^Population Differentiation in Tree-Ring Growth Responses of White Fir (_Abies concolor_) to Climate: Implications for Predicting Forest Responses to Climate Change^Ph.D.^University of California at Berkeley^143^ Dissertation^^^Graduate Fellowships for Global Change^^^^^^^67^^^^^^^^^^^^Energy and Resources Group^Harte, John^Wohlpart, in addition to their hitherto accepted use for studying equilibrium properties of climate.Ig%,G?Bⓔ\%QjS RiKU~k#Svq1cS÷~ᄟ~ ㄍGԤ3"VJS -62`IKQfx+R*hѵNNk_"YnƊ^9(B !4vfsW$QМkGO=NGÔfvt86j jS6K2@WJT˕3}.)2bư238*:ic Ru VW 0.817 and 0.672 mol m{+-2} s{+-1}, respectively; for sorghum, these values were 0.515 and 0.339 mol m{+-2} s{+-1}, resp*?ectively. Sunflower grown with the high night temperature had a higher concentration of carbohydrates than sunflower grown-@ under the ambient night temperature. Carbohydrate concentration in sorghum grown with high night temperature was not consHAistently higher or lower than that in sorghum grown with ambient night temperature. Height of both sunflower and sorghum wKBas reduced by elevated night temperature. After 74 days of exposure to the two different temperatures, sunflower grown und\Cer the ambient night temperature was about 30 cm taller than the sunflower grown under elevated night temperature; for sor_Dghum, the difference was about 25 cm. For sunflower, flowers of plants grown under the elevated night temperature appearedlE one week earlier than flowers of plants grown under the ambient temperature. For sorghum, no flowers formed under either oFtreatment by harvest (74 days after planting). The results suggest that, if night temperatures do increase due to climate {change, development of flowers of sunflower will be hastened.of winter to summer concentration were observed to be unifor~nges between the equatorial region and shallow layers of the temperate thermocline. Parallel model runs for {+90}Sr and tr pslands introduced almost half of the present-day {+90}Sr inventory into the Pacific Ocean. An upper ocean Pacific model wiIGCarbon, Climate, and Vegetationpart, A.^DE-AC05-76OR00033odel and a box model constrain OH at northern midlatitudes to JGAtmospheric Radiation Measurement (ARM)ions of 1) non-methane hydrocarbon emissions from a northern midlatitude forest, L53^1^Chen,M^1994^6^Validation and Adjustment of Precipitation Simulated by CCM2/BATS over the Continental United States^M.MS.^University of Arizona^86^Thesis^^^Computer Hardware, Advanced Mathematics, and Model Physics (CHAMMP)^^^^^^^101^^^^^^^^^^^^Atmospheric Sciences^Dickinson, Robert E.^Dickinson, Robert E.^DE-FG02-91ER61216Q@g'5ErsTPYQ@g'5E COA^100^This study compares frequency, intensity and amount of precipitation simulated by NCAR CCM2/BATS with those observed DP over the continental United States. The emphasis is placed on both spatial and temporal variations. The analysis indicate GQs that: a) The model simulates the seasonal variation of daily intensity reasonably well, while the seasonal variation of Rfrequency is poorly produced. b) The model underestimates the amplitude of intensity, overestimates the amplitude of frequ Sency, and distorts the distribution of phase in diurnal variations. c)The simulated maximum hourly precipitation has almos Tt the same range as is observed. However, the model fails to simulate the overall pattern of maximum hourly precipitation, U maximum wet and dry periods from observations. d)The model overestimates the frequency of hourly precipitation by a factoVr of 13-48 and underestimates intensity by about 90%.These results demonstrate a necessity to adjust simulated precipitatiWon for realistic surface simulation. After adjustment, the simulated intensity is in much closer agreement to the observed value.a low carbon future with as little loss of output as is possible. The principal policy option that a government haY54^1^Delaney,Paul^1991^6^Sensitivity of Atmospheric Radiation to Temperature, Water Vapor and Ozone Profiles^M.S.^UniversiZty of Maryland at College Park^23^Thesis^^^Quantitative Links^^^^^^^103^^^^^^^^^^^^Meteorology^Ellingson, Robert G.^Ellingson, Robert G.^DE-FG05-90ER60971asing CO2 production through utilization of carbon intensive fossil fuels in a nation's e\A^102^The sensitivity of the longwave spectral radiance incident on the surface to uncertainties in the measurement of tem]perature, water vapor and ozone have been examined with the use of the line-by-line radiation code FASCODE2. Spectrally av^eraged results at one wavenumber resolution were obtained for simulated random and systematic errors in the meteorological_ data under midlatitude summer conditions. As expected, temperature errors dominate the uncertainties in the opaque region`s, whereas water vapor errors dominate the sensitivity in the atmospheric window. In general, the radiation field is most asensitive to uncertainties in water vapor, followed by temperature, tropospheric ozone then stratospheric ozone. Systematibc errors dominate radiance uncertainties, with reasonable random errors in the meteorological data resulting in radiance ejcrrors that are of the of 0.25 to 0.5 the magnitude of the systematic effects. The maximum effects of systematic errors on dthe fractional radiance error at any spectral location in the 5 to 20 micron region is the order of 2% per degree Kelvin and 1.2% per percent water vapor mixing ratio error for temperature and water vapor uncertainties, respectively.ime and hof55^1^Han,Dejiang^1993^6^Derivation of Cloud Fraction from Surface-Based Longwave Flux Observations^M.S.^University of Marygland at College Park^18^Thesis^^^Atmospheric Radiation Measurement (ARM)^^^^^^^105^^^^^^^^^^^^Meteorology^Ellingson, Robert G.^Ellingson, Robert G.^DE-FG05-90ER61075t solution difficult, this dissertation develops procedures for determining boEA^104^Theoretical studies have shown that cloud geometrical factors have a significant impact on longwave fluxes and heatiHj41^1^Morris,Gary Allen^1994^6^Demonstration and Evaluation of Trajectory Mapping^Ph.D.^Rice University^100^Dissertation^^^Graduate Fellowships for Global Change^^^^^^^77^^^^^^^^^^^^Space Physics & Astronomy^Few, Arthur, and Mark Schoeberl^WohlplA^76^The problem of creating synoptic maps from asynoptically gathered data has prompted the development of a number of scjmhemes. Most notable among these schemes are the Kalman filter, the Salby-Fourier technique, and constituent reconstructionmn. This thesis presents a new technique, called trajectory mapping. Trajectory mapping employs a simple model of air parcelo motion to create synoptic maps from asynoptically gathered data. To assess the applicability of the technique, four sourcpes of trajectory mapping errors were analyzed. The analysis revealed that 1) the computational error is negligible; 2) meaqsurement uncertainties can result in errors which grow with time scales on the order of a week; 3) isentropic approximatiorns lead to errors characterized by time scales of a week; and 4) wind field inaccuracies can cause significant errors in isndividual parcel trajectories in a matter of hours. All the studies, however, indicated that while individual trajectory etrrors can grow rapidly, constituent distributions, such as those depicted in trajectory maps, are much more robust, maintauining a high level of accuracy for periods on the order of several weeks. The trajectory mapping technique has been succesvsfully applied to a variety of problems. First, trajectory mapping was employed in the study of dynamical wave-breaking evwents. Second, trajectory mapping was applied in satellite data validation studies, both for the determination of instrumenxt accuracy and precision. Third, trajectory mapping was used to assess the accuracy of the meteorological wind fields. Sucyh demonstrations imply that trajectory mapping will become an important tool in answering questions of global change, particularly the issue of ozone depletion.@!Ε(:T89U !Nq6,]EIK>?÷gggfIaԞ,{42^1^Everham,Edwin Mckay^1994^6^Hurricane Disturbance and Recovery: An Empirical and Simulation Study of Vegetation Dynami|cs in the Luquillo Experimental Forest, Puerto Rico^Ph.D.^State University of New York at Syracuse^278^Dissertation^^^Graduate Fellowships for Global Change^^^^^^^79^^^^^^^^^^^^College of Environmental Science and Forestry^Hall, Charles A.S.^Wo~A^78^Hurricanes and other catastrophic wind events impact forests over most of the globe. These disturbances are predicted to increase in frequency and intensity if ocean temperatures rise as the result of global warming. I examined spatial patterns of damage and the factors that influence these patterns, and the resulting dynamics of recovery in a subtropical wet forest following hurricane disturbance. I developed an approach to modeling disturbance dynamics in which any point in geographical space can be represented as a position in a multiple gradient space of abiotic factors; and disturbance and recovery are quantified as movements within this gradient space. Spatial pattern analysis indicates a significant clumping of hurricane damage. Canonical correlation analysis is used to compare the relative influence of abiotic factors to biotic factors in determining the severity of damage. Spatial patterns of hurricane damage are most strongly correlated to biotic factors, i.e., species of trees. Recovery is quantified in terms of vegetation community dynamics, rates of biomass accrual, and restructuring of the canopy; and is predicted based on gradients of abiotic factors and severity of hurricane disturbance. A two-dimensional gradient space of damage severity, quantified as structural damage (percent of stems downed) and compositional damage (percent of stems killed), is effective in delineating the two major paths to recovery: regeneration (sprouting of surviving trees) and recruitment (establishment of pioneer species). A three-dimensional gradient space, of structural damage and simulated light and soil moisture, allows prediction of rates of biomass accrual. Post-disturbance biomass increases are maximized under high levels of light and moderate levels of soil moisture and damage. The vertical restructuring of vegetation is associated with community dynamics of recovery, as predicted by damage severity, and by light available to the understory, as associated with topographic position. These responses are integrated into a spatially-explicit computer model (RECOVER) which simulates patterns of damage and recovery from a hurricane of a given intensity and track. This model is used to investigate the implications of changing the hurricane disturbance regime.S H:#;T43^1^Kearney,Anne Regina^1993^6^Promoting Ride Sharing: The Effect of Information on Knowledge Structure -- A Cognitive Perspective^M.S.^University of Michigan^99+^Thesis^^^Graduate Fellowships for Global Change^^^^^^^81^^^^^^^^^^^^School of Natural Resources and Environment^DeYoung, Raymond, and Stephen Kaplan^Wohlpart, A.^DE-AC05-76OR0003305-76OR00033v5A^80^Transportation sources are a dominant source of many air pollutants which contribute to environmental problems. They contribute 70% of carbon monoxide emissions, 45% of the nitrous oxides, 34% of hydrocarbon emissions, and 73% of lead emissions (EPA, 1991). These pollutants contribute significantly to ozone depletion, global warming, and acid rain, as well as cause many adverse health effects (Graedel and Crutzen, 1990; Schneider, 1990). A major component of transportation-related emissions can be attributed to work-related trips. The length and duration of these trips has continued to grow and greatly reduces any overall pollution reduction achieved through improved auto efficiency and cleaner fuels (Schreffler and Kuzmyak, 1991; EPA, 1991). The 1990 Amendments to the Clean Air Act were designed to address this problem by mandating the implementation of transportation control measures to reduce work- related vehicle trips. Specifically, all employers with at least 100 employees who are located in areas where emissions levels are above a set standard are required to implement a trip reduction program. However, while economic-based programs have often been successful at the macro-level (e.g., utilities, large manufacturers), they have generally failed to affect long-term changes at the micro-level (i.e. the individual energy consumer) (Dennis et al., 1990). Energy users have failed to adopt energy-conserving technology even when it is highly cost-effective (Ross and Williams, 1981) and the adoption of energy conservation practices has been described as unpredictable (Yates and Aronson, 1983). Behavioral changes which do occur are usually at the minimal acceptable level and discontinue when the economic incentive is removed (Cook and Berrenberg, 1981). Certainly for this category of programs, the rational economic model is not a good description of individual behavior; reliance on this model is likely to continue to produce programs and strategies which are ineffective. It appears unlikely that the rational-economic model can provide a useful framework for promoting energy conservation among individual consumers. A more descriptive model of decision-making has been proposed by Kaplan (1991) and is based on what is known about human cognition, human behavior, and human evolution. While this model recognizes that economic incentives are one factor in the decision- making process, emphasis is also placed on the role of knowledge in the decision-making process. Support for this position is derived from numerous other studies pointing to the importance of knowledge and the impact of a lack of knowledge when making decisions (e.g., DeYoung, 1988-89; Weigel and Amsterdam, 1976). It is hypothesized that this knowledge-based model will prove useful for the examination of individual decision-making with regards to employee ride-reduction behavior. The use of case studies or narratives (or "stories" as they will be referred to in this paper) which provide interesting, vivid, concrete, and personalized information has been suggested as an effective way to transfer information (Monroe and Kaplan, 1988). The characteristics of a good story (e.g., coherence, vivid and concrete detail, a sense of mystery) readily encourage a depth of cognitive processing which makes it likely that the information will be used when making future decisions. Stories have great potential for educating people about phenomena which cannot be directly experienced and for offering behavioral solutions to environmental problems. This study will use Kaplan's (1991) knowledge-based decision-making model to explore the impact of knowledge on attitudes and intended behavior towards carpooling. The model predicts that the knowledge an individual has about a particular subject has a significant effect on his/her decisions. In general, people with little or no knowledge about a behavior (or consequences of a behavior) will tend to avoid that behavior. People with a well developed knowledge structure about a particular behavior will feel more confident in their ability to carry out the behavior and will thus be more likely to engage in that behavior. This assumption is supported by empirical studies on the differences between solo drivers and users of alternative transportation (e.g., public transit, carpools). These studies (covered in the Literature Review) suggest that a major reason why drivers will not switch to alternative modes of transportation is because they lack information on how to switch and how to deal with problems which may be encountered while using an alternative mode. Another purpose of this study is to compare two different information campaigns on their effectiveness in imparting knowledge about carpooling to the study participants. Studies (covered by Literature Review) show that pallid, relatively abstract information (e.g., factsheet, factual brochures) are not very effective at transferring information. In contrast, using stories to transfer information may be very effective. In this study, a story-based information campaign will be compared to a factual-based information campaign. Both of the groups receiving the information campaigns will be compared to a control group. Carpooling was chosen as the target behavior because it traditionally has a low level of adoption and because of its wide applicability (e.g., as compared to public transit). Carpooling has been shown to be an effective means of reducing employee vehicle trips, thereby reducing environmental degradation due to commuting (EPA, 1991).+%K-/avR,C=44^1^Zhang,Qing^1990^6^Test of Transilient Turbulence Theory Against a Field Experiment^M.S.^University of Wisconsin - Madison^88^Thesis^^^Atmospheric Radiation Measurement (ARM)^^^^^^^83^^^^^^^^^^^^Atmospheric and Oceanic Sciences^Stull, Roland B.^Stull, Roland B.^DE-FG02-92ER61361^@6j)PM lSf5٨@6j)PM lSf5٨@6j)PM lSfA^82^A simple 1-dimension grid-point boundary-layer simulation model based on transilient turbulence theory is tested against BLX83 experiment data. Transilient coefficients are parameterized using a simplified form of the total kinetic energy (TKE) equation instead of Richardson numbers. The model is absolutely numerically stable. May 28, 1983, a cloudless day with sufficient measurements, was chosen as the case-study day. The early morning radiosonde sounding is used to prescribe the initial conditions. Surface heat and moisture flux from an aircraft and flux tower, and surface geostrophic wind, thermal wind and horizontal temperature gradient are computed from surface temperature, pressure, and wind data from portable automated mesonet (PAM) stations, and are used to prescribe the forcings and boundary conditions on the model. Subsidence calculated from horizontal divergence is prescribed at the top of the model. Simulated potential temperature and specific humidity profiles compare well with actual radiosonde and aircraft soundings. Modeled heat flux decreases linearly with height as expected. Moisture flux profiles show a maximum near the top of the mixed layer (ML), also expected. There is no explicit forecast equation for ML depth z{-i} in the transilient model, yet the simulation exhibits a growing ML with entrainment. Modeled z{-i} is between radiosonde and aircraft soundings and lidar measurements. Modeled surface potential temperature and specific humidity evolution are close to the PAM II and radiosonde observations. A nonlocal spectral decomposition of the flux at different levels indicates that most contributions of the total flux at height Z are made by the eddies with size Z.98d#7Gp[W>x+(kV;3 zHE.ཇ_j[DŽ5<"p_`/Bcs203^1^Santoso,Edi^1993^6^A Wind-Profile Relationship for the Unstable Surface-Layer/Mixed-Layer System^M.S.^University of Wisconsin - Madison^59^Thesis^^^Atmospheric Radiation Measurement (ARM)^^^^^^^369^^^^^^^^^^^^Atmospheric and Oceanic Sciences^Stull, Roland B.^Stull, Roland B.^DE-FG02-92ER61361^٨@6j)PM lSf5٨@6j)PM lSf5٨@A^368^Simple expressions for wind-speed, M, profiles in a convective statically-unstable surface-layer/mixed-layer system are derived empirically. The wind-speed profiles are consistent across the top of the surface layer. Wind shear decreases rapidly with height, z, in the surface layer and becomes negligible at the top of the surface layer and in the mixed-layer interior. Thus, one may assume zero wind shear as a boundary condition and definition for the top of the surface layer. The depth of the surface layer, z{-s}, and the wind speed in the mixed layer interior, M{-ML}, can then be estimated. For the surface-layer/mixed-layer system, it is hypothesized that z{-s} and M{-ML} may become relevant height and velocity scales in a similarity relationship, formulated as M/M{-ML} = f(z/z{-s}). Wind-speed data collected during the Minnesota atmospheric boundary-layer experiment are used to develop and evaluate alternative forms of the wind-speed profiles. All the data were measured during the late morning or mid afternoon, when the surface heat flux was positive and the depth of the mixed layer was nearly constant. Wind speeds near the surface increase rapidly with height, then become nearly constant in the mixed layer interior. Wind directions are nearly constant with height, with variations mostly less than 5 degrees for winds near the surface, and only some greater than 10 degrees for winds in the mixed layer. Based on close agreement with the data, there are two functional forms adopted in the surface layer (z z{-s}): M/M{-ML} = {{(z/z{-s}){+A}{}exp[A(1-z/z{-a})] with A = 0.0975 and M/M{-ML} = (z/z{-s}){+{{A(1-z/z{-s}){}} with A = 0.0840. Above the top of the surface layer (z > z{-s}), the wind speed is simply equal to M{-ML}. These expressions give good agreement to the observations and approximate the data to within a few percent. The average depth of the surface layer, z{-s}, is found to approach the traditional definition of z{-s} = 0.1*z{-i} (the depth of the mixed layer), but the relationship is weakly correlated. It is also found that the surface vertical momentum flux is highly correlated with M{-ML}{+2}, as expected. The mixed-layer drag coefficient value is found to be Cd{-ML} = 0.0014. While the Businger-Dyer formula that is extrapolated to the mixed-layer interior deviates when the height approaches the top of the surface layer, and usually is in error over all of the mixed layer, the functional forms proposed here can represent wind-profile data quite well within 99.9% of the whole surface-layer/mixed-layer domain. The functional forms above are not recommended for use below z = 1m, where they fail to represent the bottom portion of the log wind profile. This approach also does not work well when there is not a well-mixed layer of uniform wind speed, such as occurs with very weak convection and strong winds.s. y32g￳#>~g>J-?W46^1^Schrieber,Kelly M^1994^6^Observations of Surface-Layer Characteristics and Their Relationship to Boundary-Layer Cumulus^M.S.^University of Wisconsin - Madison^67^Thesis^^^Atmospheric Radiation Measurement (ARM)^^^^^^^87^^^^^^^^^^^^Atmospheric and Oceanic Sciences^Stull, Roland B.^Stull, Roland B.^DE-FG02-92ER613618]agu]6h$>xz+ A^86^The subject of subgrid scale heterogeneity is one which has continued to pose problems for the numerical modeling community. Observations of meteorological and land use conditions which may account for these subgrid variations in the case of fair weather boundary layer cumulus (BL Cu) are presented. Clear sky cases are examined from aircraft data from the HAPEX field experiment of 1987, in order to eliminate any further heterogeneities which may be caused by cloud shading, enabling a more accurate assessment of the relationship between atmospheric conditions and the underlying surface. To account for the physical processes involved in the formation of non-precipitation BL Cu, buoyancy and moisture variables are chosen. The virtual potential temperature and lifting condensation level (LCL) are calculated for each data point along the surface-layer flight path of the aircraft. The path is divided into segments according to general surface type categories, and the resulting segments are examined. The original heterogeneous surface characteristics can later be obtained by a weighted average of the homogeneous segment characteristics. Joint distributions of virtual potential temperature and LCL are presented and evaluated. These distributions are found to be primarily mono-modal over relatively homogeneous ground-surface subdomains, and exhibit a sharp peak which tails off quickly to a finite end. The mean value for each distribution is found to be representative of the desired characteristics. It has been determined that distributions of LCL and virtual potential temperature over homogeneous land surfaces are mono-modal and remain fairly consistent over time. This quality is therefore one which can be parameterized and provide reliable information about BL Cu onset time and coverage. The information required is readily available from both routine measurements and normal model output. Sub-grid variability of surface type can be obtained from satellite inventories and cataloged for future model use. Finally, a conceptual model for a simple cumulus parameterization is presented. Based on the calculated LCL and virtual potential temperature values for a given homogeneous patch, the cumulus cover can be diagnosed by comparison with an averaged area sounding. One would expect to find either the LCL to be higher than the measured inversion height (dry conditions), or the virtual potential temperature to be cooler than the environmental sounding values (non-buoyant), for all of the clear-sky cases presented. It appears initially that this conceptual model was successful for the cases examined here.0XYxQ'=NXO`f >}47^1^Schmitt,Christina C^1990^6^The Effects of Surface Temperature and Clouds on the CO{-2} Forcing^M.S.^Colorado State Un iversity^114^Thesis^^^Carbon, Climate, and Vegetation^^^^^^^89^^^^^^^^^^^^Atmospheric Science^Randall, David A.^Randall, David A.^DE-FG02-89ER69027^alPl{E\Q]0 ^#,Ofg~O~l`{?8?integrals to time-step the prognostic equations. Multigrid methods are used to solve the diagnostic equations for the stre?am function and velocity potential. The model is compared with the Arakawa-Lamb shallow water model and the NCAR spectral @transform shallow water model using the suite of seven test cases proposed by Williamson. The model performance characteriAstics are presented. The test cases show that the evolution of the fields is independent of the relative orientations of tBhe computational grid and the flow pattern. This is particularly true when flow is directed over the pole of the grid. AlsCo, the new model is tested using a Rossby-Haurwitz wave as initial conditions. The initial disturbance breaks down towards lower wavenumbers, but remains symmetric across the equator.YF3u晍OIk*]G1ivi]i1<,B.@e'">E50^1^Cripe,Douglas G^1994^6^Investigation of GCAPE Quasi-Equilibrium in the Midlatitudes^M.S.^Colorado State University^23F0^Thesis^^^Atmospheric Radiation Measurement (ARM)^^^^^^^95^^^^^^^^^^^^Atmospheric Science^Randall, David A.^Randall, David A.^DE-FG02-91ER61363^;sc{r`lS]Y]g~yDgWuguŞ݇냾;Wc,]yg]"]qGEnHA^94^Lorenz (1955, 1978, 1979) developed the concept of the "moist available energy" (MAE) of the atmosphere. This he defiIned as the portion of non-kinetic energy (NKE) available for conversion to kinetic energy (KE). Randall and Wang (1992) anJd Wang and Randall (1994) showed that it is possible to consider the component of the MAE that resides in the vertical strKucture of the atmosphere as a "generalized convective available potential energy" (GCAPE). Using data from the tropics, thLey tested the GCAPE quasi-equilibrium hypothesis (Arakawa and Schubert, 1974) which asserts that cumulus convection "consuMmes" GCAPE as quickly as it is produced by large-scale (non-convective) forcing such that the convectively active atmospheNre remains close to a state of conditional neutrality. The main purpose of this study is to also investigate the GCAPE quaOsi-equilibrium hypothesis, only this time in a midlatitude setting. This is a tougher test of the hypothesis given the sigPnificantly larger temperature and moisture fluctuations resulting in a stronger large-scale forcing in the midlatitudes, cQompared with the tropics. Data recently made available by the Atmospheric Radiation Measurement (ARM) program has been useRd. This new data comes from radiosonde measurements collected at ARM's Cloud and Radiation Testbed (CART) site located in Snorth-central Oklahoma during Intensive Operation Periods (IOPs) run periodically throughout the year. Since this is one oTf the first studies to make extensive use of this data, a further goal was to evaluate the quality of the wind and thermodUynamic measurements being produced by the CART site. Additionally, analysis data from the Mesoscale Analysis and PredictioVn System (MAPS) was used, both as a check on the reasonableness of the ARM data, and also to detect any possible errors in the MAPS model output.]qwoeW}~Lwi8§9qdb?_wϏO}];?F>>}=wX51^1^Gleckler,Peter John^1993^6^The Partitioning of Meridional Energy Transport Between the Ocean and the Atmosphere^Ph.D.Y^University of California at Davis^160^Dissertation^^^Carbon, Climate, and Vegetation^^^^^^^97^^^^^^^^^^^^Land, Air & Water Resources^Weare, Bryan C.^Gates, W. Lawrence^W-7405-ENG-48^=?T8 ??>>=}gi-Ǒ}}zc:'?[A^96^The partitioning of meridional energy transport between the ocean and the atmosphere is a fundamental feature of the \climate system, and unfortunately one that remains poorly understood. Observationally-based estimates of atmospheric and o]ceanic meridional energy transport (respectively, AMET and OMET) resulting from various methods have been intercompared. E^nergy fluxes at the top of the atmosphere and at the ocean surface have also been studied because they are required by sev_eral methods used to estimate meridional energy transport. The resulting intercomparison has demonstrated that estimates o`f oceanic and atmospheric energy transport are inconsistent with one another. The uncertainties associated with OMET estimaates made with several methods have been established, but uncertainties in AMET estimates appear too complex to accuratelyb quantify. Much of this study has focused on ocean surface energy fluxes because annual mean estimates of AMET and OMET coculd be inferred if they were accurately known. Two methods have been established to improve our understanding of OMET (andd thus indirectly AMET). Both involve combining the same independent data and their associated uncertainties. All evidence esuggests that the most commonly cited estimates of Northern Hemisphere AMET based on radiosonde data are systematically tofo low. Lesser known estimates which make use of dynamical constraints are much more consistent with the various independengt estimates of OMET. At most latitudes in the Southern Hemisphere, it is not even clear whether the global ocean OMET is nhorthward or southward. Meridional energy transport has also been examined in three atmospheric GCM simulations. There are iconsiderable differences among the various models, particularly in the Southern Hemisphere OMET. It has been demonstrated jthat the implied transports in the models are strongly linked to the simulated cloud radiative forcing. The results of thiks study suggest that the most practical way to significantly improve our understanding AMET and OMET is with a small numbelr of strategically placed high quality hydrographic transects. It has been recommended that the utility of the methods devmeloped in this research which combine independent observationally-based data and their estimated uncertainties be considered in the planning of future global climate research programs..xOOGX_lb̬nrcy Ӻ87!o52^1^Rosa,Charles H^1993^6^Modeling Investment Uncertainty in the Costs of Global CO{-2} Emission Policy^Ph.D.^University pof Michigan^126^Dissertation^^^Graduate Fellowships for Global Change^^^^^^^99^^^^^^^^^^^^Industrial and Operations Engineering^Birge, John R.^Wohlpart, A.^DE-AC05-76OR00033H&]{Ξy9Cϡovoo>ϟYm$l/OrA^98^The dissertation investigates how the United States economy will evolve during the next 110 years if policies are impslemented which limit the production of carbon dioxide as a by-product of economic activity. Increased concentrations of attmospheric carbon dioxide, a principal determinant of terrestrial weather patterns, have led many to conclude that governmeuntal intervention is necessary to induce the world's economy to rely less upon carbon based fossil fuels. Many deterministvic macroeconomic models have been developed to try and determine what specific policies are necessary to cause this shift waway from fossil fuel use as well as what the cost of such measures are in terms of lost economic output. This dissertatioxn develops a nonlinear stochastic macroeconomic model with continuously distributed random variables that determines optimyal policy while explicitly modeling the uncertainty inherent in new technology development and exhaustible resource explorzation. It compares the optimal policy developed in this stochastic model with non-optimal policy derived under the assumpt{ion of a certain future, and measures the resulting improvement in economic performance. Because the stochastic model with| continuously distributed random variables is very difficult to solve, this dissertation bounds the true solution. It uses} a parallel implementation of a nonlinear decomposition algorithm to obtain an upper bound and develops a new lower boundi~ng method based on the linear response function technique for stochastic linear programs to derive a lower bound. The quality of the resulting bounds is explored as well as the benefits of parallelization as a means of improving solution times.6PhBƹO%-,Sf5٨@6j)PM o3K_L+ѪcB''uIS돇Bmƀط>5~ë#8P5\'S yng rates, yet most models used in climate studies employ rather crude approximations of these effects. This study was designed to determine the possibility of estimating such geometrical effects with the use of surface-based infrared flux observations as might be obtained during the Atmospheric Radiation Measurement (ARM) program. The results show that pyrgeometer measurements can be used to determine the effective cloud fraction in a manner useful for geometrical parameterizations if the error variance of the observations is less than about 5 Watts per meter squared. In general, the uncertainties associated with such determinations increase with increasing water vapor burden. Overall, the pyrgeometer error restrictions are difficult to meet, but they may be possible with careful monitoring and control.~y(ο>ۧb?=H/7172^1^Santer,Benjamin David^1988^6^Regional Validation of General Circulation Models^Ph.D.^University of East Anglia^375^Dissertation^^^Carbon, Climate, and Vegetation^^^^^^^309^^^^^^^^^^^^School of Environmental Sciences^Wigley, T.M.L.^Wigley,e modelled soil organic material's turnover times, which range from decades to millennia, as a binary system having active and passive components. CO{-2} fertilization, the plant net primary productivity increase observed at elevated CO{-2} lev184^1^Harrison,Kevin Geyer^1993^6^The Impact of CO{-2} Fertilization, Changing Land Use and N-Deposition on Soil Carbon Storage^Ph.D.^Columbia University^134^Dissertation^^^Carbon, Climate, and Vegetation^^^^^^^332^^^^^^^^^^^^Lamont-Doherty EarA^331^Fossil fuel combustion and changing land use have increased atmospheric and oceanic carbon inventories less dramatic:ally than expected. This discrepancy has been termed the "missing sink." CO{-2} fertilization and N-deposition processes c84^1^Prior,Stephen Arthur^1993^6^Cotton Root Response to Free-Air CO{-2} Enrichment^Ph.D.^Auburn University^136^Dissertati9on^^^Carbon, Climate, and Vegetation^^^^^^^149^^^^^^^^^^^^Agronomy and Soils^Rogers, Hugo H.^Rogers, Hugo H.^ER60001^Agron61^1^Najjar,Raymond G.^1990^6^Simulations of the Phosphorus and Oxygen Cycles in the World Ocean Using a General Circulation Model^Ph.D.^Princeton University^190^Dissertation^^^Carbon, Climate, and Vegetation^^^^^^^147^^^^^^^^^^^^Geological & Geophysical Sciences^Sarmiento, Jorge L.^Sarmiento, Jorge L.^DOE MM-19X-27405C^|sq_}l6-QWOyO8מ?q_;Z;64^1^Arsky,Natalie^1994^6^Prediction of the Generalized Ecotypes with COLA GCM Using Four Different Bioclimatic Schemes^M.S.^University of Maryland at College Park^22^Thesis^^^Atmospheric Radiation Measurement (ARM)^^^^^^^117^^^^^^^^^^^^Meteorology^Baer, Ferdinand^Baer, Ferdinand^DE-FG02-94ER61746ώߞwvw"*8OyO?; Aߺ><gĿ1A^116^The problem of interactive coupling of the AGCM to the biospheric model is examined. Four different bioclimatic schemes, which relate the atmospheric parameters like annual temperature and precipitation to the vegetation distribution, were implemented. The schemes were applied to two global climate data sets. One represents the result of the COLA (Center for Ocean-Land-Atmosphere Interactions) model R-15 version simulation of the present-day climate (without coupled ocean). Ten years monthly averages of near-surface temperature, total precipitation, soil wetness, net long-wave radiation and short-wave radiation absorbed at the surface were used for schemes calculations. Another data set represents climatological temperature and precipitation data (Jaeger data set, NCAR). The main result is that all four models are able to reproduce the main features of the vegetation distribution. Thus, the schemes can be used as a first step in coupling of atmospheric GCM with biosphere, e.g. "interactive biosphere". In addition, they can be a useful tool in interpreting of the GCM results of future and past climates simulations and a diagnostic tool for verification of the model performance in a simulation of the present-day climate.i q8N-?W]Vqc~3ﳻk/oOj}]p`߳>?865^1^Zhang,Bing^1993^6^Use of Taylor Expansion in Computation of the Simplest Spectral Barotropic Vorticity Equation^M.S.^University of Maryland at College Park^29^Thesis^^^Computer Hardware, Advanced Mathematics, and Model Physics (CHAMMP)^^^^^^^119^^^^^^^^^^^^Meteorology^Baer, Ferdinand^Baer, Ferdinand^DEFG0591ER61219^qf&mp߰* ;  \ GM@A^118^Time integration is the basic part of the computation of meteorologic forecast models, but the time step is constrained by stability consideration, CFL condition must be satisfied for linear problems. The time steps are very short for high frequency components, this makes the integration procedures extremely time-consuming, especially for climate models. The current popular numerical integration scheme for time integration is three time-level leapfrog scheme which uses the first order derivative at the mid-level, we will refer this method as the regular leapfrog scheme. In this paper, a system of the simplest spectral barotropic vorticity equation in channel domain is experimentally computed by using Taylor's expansions of time dependent variables so that higher derivatives are included in computing the values at next time level. Taylor's expansion is applied for two computational schemes: one is "leapfrog" with the derivatives evaluated at the mid-level, the other is the two time-level forward scheme with the derivatives evaluated at level one, at which all variables values are considered as known, and the predicted outputs are given at level two. The experiment shows that on the average, when certain number of derivatives are included in the expansions, the integration time step can be more than 6 times of those given by the regular leapfrog scheme without losing any accuracy. This results tells us that it is possible to integrate the equations with a much higher time step by rewriting the forecasting equations in their Taylor's expansion forms. Although Taylor's expansion forms are usually much more complicated compared with the forms of the regular leapfrog scheme, time integration can still be fastened if each term in Taylor's expansion is assigned to one processor of the Massive Parallel Processors. The deal is the lose of simple expression form in exchange of the gaining of computational speed when MPP is used. In addition, the simple expansion form of the regular leapfrog scheme (several integration time steps are combined into one expression) is tested. Compared with Taylor's expansion forms, it will take about two orders magnitude higher computation times for the same increase of integration time step. This shows that the perspective of making use of Massive Parallel Processors with the use of Taylor's expansion forms are brighter..zǐ׿;~gO瞶Ogp>x_p,66^1^Charney,Joseph J^1992^6^Heating Rate Variability in Radiation Codes When Subjected to Different Vertical Discretizations^M.S.^University of Maryland at College Park^43^Thesis^^^Atmospheric Radiation Measurement (ARM)^^^^^^^121^^^^^^^^^^^^Meteorology^Baer, Ferdinand^Baer, Ferdinand^DE-FG05-90ER61075yy7c8?g3ۨiU ){ܐZIeiW }r@i%A^120^A collection of seven radiation codes was acquired from various organizations and intercompared. Their variability with respect to each other was determined when performing heating rate calculations on different distributions of vertical levels. Specifically, two vertical discretizations with 30 and 18 levels were intercompared. Some significant differences were found to exist by performing simple statistical tests on the models. The statistical tests were based on sets of 100 heating rate calculations. Independent determinations were made at each model level. The parameterizations of water vapor absorptivities were found to cause the largest variability between the models. Other mechanical problems were highlighted and recommendations for future steps in the ongoing study were proposed.lSf5b3BvtqcZork at Albany^86^Thesis^^^Carbon, Climate, and Vegetation^^^^^^^315^^^^^^^^^^^^Atmospheric Science^Wang, Wei-Chyung^Wang, 175^1^Zhong,Min^1994^6^The Greenhouse Effect in the Present Earth-Atmosphere Climate System^M.S.^State University of New Y308^1^Brakke,Mary P^1989^6^Gas Exchange and Growth Responses of Citrus Trees to Partial Irrigation, Soil Water, and Atmospheric Conditions^Ph.D.^University of Florida^181^Dissertation^^^Carbon, Climate, and Vegetation^^^^^^^609^^^^^^^^^^^^Agronomy^Allen, Leon H., Jr.^Allen, Leon H., Jr.^DE-AI01-81ER60001Q:xV,ǞvC{PPMt}q~y}vw" fǞin the cloud-radiation parameterization.thetic Response of Soybeans: Model Development and Elevated Carbon Dioxide Experiry to the conclusion of Wigley and Raper (1990a).^^^Carbon, Climate, and Vegetation^^^^^^^528^^^^^^^^^^^^Agricultural Engineering^Jones, J.W.^Allen, L.H., Jr.^DE-AI01-81ER60001^*4}ߞ3UgxwƱLֽ3ޣ-}ޏ G=v&;311^1^Roy,Brian L^1988^6^Development and Evaluation of an Adaptive Control Procedure for Carbon Dioxide Control in Environmental Growth Chambers^M.Eng.^University of Florida^130^Thesis^^^Carbon, Climate, and Vegetation^^^^^^^615^^^^^^^^^^^^Agricultural Engineering^Jones, Pierce H.^Allen, Leon H., Jr.^DE-AI01-81ER60001, DE-AI05-88ER69014}{>|oϾ72^1^Hussain,Mian W^1992^6^Effect of Elevated CO{-2} and Temperature on Sucrose Phosphate Synthase Activity and Carbohydrate Metabolism in Rice^Ph.D.^University of Florida^157^Dissertation^^^Carbon, Climate, and Vegetation^^^^^^^526^^^^^^^^^^^^Botany^Bowes, G.^Allen, L.H., Jr.^DE-AI05-88ER69014^>G?7𑟏dqoWH?o7<$ǂ~313^1^Kamuru,Freeman^1992^6^Ammonia Production by a Mutant Strain of _Anabaena variabilis_ and Its Contribution to Growth and Yield of Rice (_Oryza sativa_ L.)^Ph.D.^University of Florida^155^Dissertation^^^Carbon, Climate, and Vegetation^^^^^^^619^^^^^^^^^^^^Agronomy^Allen, Leon H., Jr., and S.L. Albrecht^Allen, Leon H., Jr.^DE-AI05-88ER690146-es#0>1ߊ(7tunity to consider not only the radiative balances at the top of the atmosphere but also the clear sky radiative fluxes. Tth Observatory^Broecker, Wallace S.^Broecker, Wallace S.^DE-FG02-91ER61247UmM~|: z!g!8G3~|ration with fixed sea-surface temperature), the OSU CGCM (in years 11-20 of a 20-year control integration, in which the tw305^^^^^^^^^^^^Chemistry^Zoller, William H.^Gaffney, Jeffrey S., and Nancy A. Marley^W-31-109-ENG-38, W-7405-ENG-36kى A^304^A field site was established at Frijoles Mesa in Northern New Mexico to study the tropospheric photochemical oxidant at a Remote Location in Northern New Mexico^Ph.D.^University of Washington^183^Dissertation^^^Atmospheric Sciences^^^^^^^169^1^De Bruyn,Warren J^1994^6^The Heterogeneous Chemistry of Carbonyl Halides, Haloacetyl Halides and Biogenic Sulfur Spe2ER61453ѐs>nz4m -m -¶nC٨@6j)PM lSflTz: Fp3)^NŐBgFK97^1^Little,Jackie^1994^6^Patterns of Recovery in a Riverine Mangrove Forest^Ph.D.^University of Tennessee^183^Dissertation^^^Carbon, Climate, and Vegetation^^^^^^^158^^^^^^^^^^^^Graduate Program in Ecology^Rennie, John^Emanuel, W.R.^DE-AC05-84A^157^This research forecasts the effect of variable hurricane intensity on mangrove forest structure and evaluates the role of species composition in recovery. Baseline data estimates were made along the Shark River in Everglades National Park, Florida, USA, in 1992, 32 years after Hurricane Donna ripped up the southwestern Florida coast. Hurricane disturbance and forest recovery were simulated in an individual-based model. Results were extrapolated to forecast the effects of two fa