B11A-0124 0800h
A Comparison of Landsat TM and ASTER for Equivalent Water Thickness Derivation in a Ponderosa Pine Ecosystem
Landsat TM and ASTER satellite data can be used to make physically-based estimates of equivalent water thickness (EWT) in a Pinus ponderosa ecosystem. EWT is a measure of ecosystem water status and is an important parameter for studying ecosystem dynamics, fire potential, and biological responses to climate change. Near infrared (NIR) and shortwave infrared (SWIR) reflectances were simulated using the LIBERTY and GeoSAIL leaf and canopy reflectance models; the results were used to calculate a NIR/SWIR ratio and a normalized NIR/SWIR index. Index-EWT relationships were modeled and inverted for EWT derivation. Landsat and ASTER were used to make reasonably accurate estimates of EWT ($\pm$ 17.3% and 19.4% mean error, respectively); TM band 5 and ASTER band 4 produced the best results. Exclusion of plots with dense understory vegetation reduced point scatter substantially, especially with Landsat (r2 = 0.847, $\pm$13%), indicating that this method can provide robust EWT quantification in homogeneous conifer ecosystems.
B11A-0125 0800h
How Vegetation Indices Perform in Estimating of Vegetation Biophysical Characteristics?
There is considerable interest in assessing biophysical characteristics of vegetation, such as vegetation fraction (VF), leaf area index (LAI), density / biomass and the magnitude of carbon sources and sinks for agricultural lands, grasslands, and forests. In this paper, we compare performance of widely used vegetation indices such as NDVI, ARVI, SAVI, EVI, PRI, MCARI, as well as recently developed indices for estimating of crop biophysical characteristics. We assessed accuracy of VF, LAI, biomass and net ecosystem carbon dioxide exchange retrieval for maize and soybean from remotely sensed data taken by two-heads hyperspectral radiometer. We compared sensitivity of the indices to biophysical characteristics as well as minimal value of biophysical characteristic that can be detected using each index. Then, using images from the SPOT-4 Vegetation sensor we compared performances of the indices for forests, savanna, grasslands and croplands among others.
B11A-0126 0800h
Evaluation Of Environmental Influences Of Urbanization Using Remote Sensing And Climate Observations
Urban development has experienced rapid growth in the Tampa Bay area of west-central Florida over the last century. This expansion trend is transforming the landscape from natural cover types to increasingly impervious urban land. Remote sensing data has been used to assess urban land cover and its thermal characteristics by mapping impervious surfaces at a sub-pixel resolution. Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper plus (ETM+) data were used to estimate urban imperviousness variation from 1991 to 2002 in the Tampa Bay watershed, Florida. The urban/rural boundary and urban development density are defined through the selection of imperviousness threshold values. The increasingly important influence that urbanization and its attendant imperviousness exert on the environmental health of the region have been investigated by analyzing radiant surface temperatures (Ts) and Normalized Difference Vegetation Index (NDVI) using ETM+ satellite data. Analysis of impervious surface and Ts shows significant difference in Ts values associated with different percentages of impervious coverage. An inverse relationship has been found to exist between imperviousness and Normalized Difference Vegetation Index (NDVI) for several urban areas. The relationship of imperviousness, Ts and NDVI has been evaluated, as has the urban heat island (UHI) effect on local climate change. Our results show that urban land use change has a profound impact on both seasonal-averaged and minimum surface temperature changes within the spatial scale of the watershed.
B11A-0127 0800h
Can Large Windfarms Affect Local Meteorology?
The RAMS model was used to explore the possible impacts of a large windfarm in the Great Plains region on the local meteorology over synoptic timescales under typical summertime conditions. A wind turbine was approximated as a sink of energy and source of turbulence. The windfarm was created by assuming an array of such turbines. Results show that the windfarm significantly slows down the wind at the turbine hub-height level. Additionally, turbulence generated by rotors create eddies that can enhance vertical mixing of momentum, heat and scalars usually leading to a warming and drying of the surface air and reduced surface sensible heat flux. This effect is most intense in the early morning hours when the boundary layer is stably stratified and the hub-height level wind speed is the strongest due to the nocturnal low-level jet. The impact on evapotranspiration is small.
B11A-0128 0800h
Mapping Land Cover Changes With Landsat Imagery and Spatio-Temporal Geostatistics; An application to the Pearl River Delta, China.
Accurate quantification of anthropogenic changes are a primordial input to many studies that relate processes (e.g. climate change) to land covers. Satellite images are the principal medium to detect and map changes in the landscape, both in space and time. However, the current image processing techniques do not fully exploit the data in that they do not take simultaneously into account the spatial and the temporal relations between the various land cover types. The method proposed here aims to accomplish that. At each pixel of the landscape, the time series of land cover type is modeled as a Markov Chain. That time series at any specific location is estimated jointly from the local satellite information, any neighboring ground truth land cover data, and any neighboring previously estimated time series deemed well-informed by the satellite measurements. The spatial component of the land cover types is integrated with variograms and indicator kriging. Our proposal has the advantage to account simultaneously for the spatial and temporal patterns of land cover types. Information and knowledge related to those patterns are coded such that expert opinions can easily be integrated in the process. The method is not limited by the time series length or by the number of land cover classes. Mathematically simple, it can be combined with any type of classifier that outputs a probability for a pixel to belong to a specific class, e.g., a probabilistic neural networks, or a maximum likelihood algorithm. The method is applied on six Landsat images spanning nine years to detect anthropogenic changes in the Pearl River Delta, China. This region is under going tremendous growth in population profoundly altering the landscape. The prediction accuracy of the time series improves significantly, the accuracy almost doubles, when both spatial and temporal information are considered jointly in the estimation process. The introduction of spatial continuity through indicator kriging reduces unwanted noise in the classified images, removing the need for post-processing. This improved integration of space and time produces a more accurate change detection map that defines more accurately when and where the landscape has changed.
B11A-0129 0800h
Monitoring Ecosystem Carbon and Water Variations During a Severe Drought in the Southwest With AVIRIS and MODIS Sensor Data.
We investigated the spatial and temporal variations in vegetation biologic activity across a wide range of ecosystems (desert shrub to conifer forest) in northern Arizona with carbon and water indices derived from fine resolution AVIRIS data and moderate resolution MODIS observations. Leaf level and canopy level surface moisture indices were computed over the range of ecosystems and drought-induced mortality sites with hyperspectral AVIRIS data in the 1240nm and 2100nm water absorption regions. The land surface moisture indices were combined with the vegetation index, carbon measures to map spatial and temporal patterns of above-ground net productivity and analyze ecosystem sensitivity to water availability and precipitation. The coupled water and carbon indices were scaled up to MODIS data for spatial extension and time series analysis over the past 5 years. Land surface moisture and carbon patterns behaved differently across the range of ecosystems and within drought impact sites. Drought impacts were observed in all ecosystems, particularly in tree mortality areas and the grassland and desert areas. Our results show that combined water and carbon indices offer improved sensitivity to ecosystem health assessment and drought detection and analysis. Remotely-sensed land surface water indices combined with the carbon products yielded important information useful in the prediction of vegetation health response to climate change and human land cover modifications.
B11A-0130 0800h
Modeling Hydrological Impact of Land-Use and Climate Change in West and Central Africa
Water resources are a particularly important issue in west and central Africa, where large portions of the continent are arid or semiarid and climate is highly variable. In the past 50 years, due to population growth this region has been experiencing land-use changes, including deforestation, overgrazing and reclamation. Meanwhile, there has been persistent drought since the 1960s. These land-use and climate changes have exerted a significant impact on the hydrological regime of this region. Therefore, a quantitative evaluation of the hydrological impacts of land-use and climate change is useful for water resource management. In this study, an investigation is conducted based on numerical simulations with a terrestrial ecosystem model, IBIS, and an aquatic transport model, HYDRA. The results show that deforestation (clearcutting) significantly increases the simulated stream flow by 34% ~ 65%, depending on location, despite the fact that tropical forests represent only a small portion (<5%) of the total basin area. Similarly, overgrazing (100% removal of grasslands) also increases simulated stream flow by 33% to 91%. The cultivation of crops significantly increases the simulated runoff, due primarily to shallow rooting depth, lower canopy cover in the early growing stage and lower transpiration demand. An increase of 10% in precipitation increases the simulated runoff by 26% ~ 48% (depending on the vegetation type); a similar decrease decreases the runoff by 24% ~ 32% (depending on the vegetation type). Changes in temperature by 1.5 °C do not cause any appreciable simulated hydrological changes. The numerical simulations indicate no significant impact on the water yield when deforestation (thinning) is below 50%, or the overgrazing is below 70% for savanna and 80% for grassland; however, the simulated water yield dramatically increases with land cover changes above these thresholds. This threshold effect, which has important implications for water resource management strategies, is reasonably explained by the nonlinearity of the separate response of transpiration and evaporation to the land cover changes.
B11A-0131 0800h
The Interplay Between the Carbon and Water Cycles Along a Successional Gradient in the Southeastern US
We measured net ecosystem exchange of CO$_{2}$ (NEE) and evapotranspiration (ET) for three years at adjacent old-field (OF), {\it Pinus taeda} dominated planted pine forest (PP) and mature oak-hickory forest (HW) ecosystems in the southeastern US (SE). The ecosystems serve as a model for post-agricultural succession, represent dominant ecosystem types in the SE, and experience identical edaphic and climatic conditions, which during the measurement period included severe drought. We present a sensitivity analysis on NEE estimates using eddy-covariance measurements alone, then constrain NEE using independent estimates of soil and ecosystem respiration and canopy carbon assimilation. Annual NEE differed greatly among ecosystems during mild drought (2001), with +150 (i.e., C source), -500 and -250 g C m$^{-2}$ y$^{-1}$ at OF, PP, and HW, respectively. During the severe drought of 2002, the magnitude of annual NEE was reduced by 350 g C m$^{-2}$ y$^{-1}$ (70%) at PP, but differed by less than 50 g C m$^{-2}$ y$^{-1}$ at the other two ecosystems. During the wetter-than-average year (2003), OF became a slight C sink, NEE at PP recovered slightly to -250 g C m$^{-2}$ y$^{-1}$, and the magnitude of NEE at HW decreased by 50 g C m$^{-2}$ y$^{-1}$ due in part to light limitation. Thus, we observed a progressive decoupling of the carbon and water cycles along the successional gradient. Over the next 40 years, the land cover area of pine plantations in the SE is predicted to double, and the land cover area of upland hardwood forests, to decline. Our results suggest that increasing the fractional cover of pine plantation will make variability in the SE C sink more sensitive to variability in the hydrologic cycle.
B11A-0132 0800h
Assessing the impact of lawn management practices on the US carbon and water budgets
Despite the ubiquity of turf grasses in the United States, the large-scale functioning of these ecosystems has been largely understudied. Although we know that turf grass systems are sequestering carbon at the expense of a large amount of freshwater resources, a continental carbon and water budget has yet to be attempted. The limited existing information on the total extent and spatial distribution of turf grasses and the variability in management practices are the major factors complicating this assessment. In this study, relating turf grass area to fractional impervious surface area, it was estimated that potentially 165,000 km2 (± 31,500 km2) of land are cultivated with turf grasses in the continental United States, an area three times larger than that of any irrigated crop. Using the Biome-BGC ecosystem process model, the growth of warm season and cool season turf grasses was modeled for 865 sites across the 48 conterminous states under different management scenarios, including either removal or recycling of the grass clippings, different nitrogen fertilization rates and two alternative water irrigation practices. The simulations portray potential carbon and water fluxes as if the entire turf surface was to be managed like a well-maintained lawn. The results indicate that well watered and fertilized turf grasses act as a carbon sink, even assuming removal and bagging of the grass clippings after mowing. The potential carbon sequestration that could derive from the total surface under turf (up to 17 Tg C/yr with the simulated scenarios) would require a 30% to 70% increase in current domestic and commercial water use, depending on the modeled water irrigation practices. Landscaping water conservation practices such as xeriscaping and irrigation with recycled waste-water may need to be extended as municipalities continue to face increasing pressures on freshwater resources due to continued population growth.
B11A-0133 0800h
Land Use Influence of Characteristics of Groundwater Inputs to the Great Bay Estuary, New Hampshire
Responding to the concerns that have arisen regarding elevated nitrate concentrations of groundwater discharging to the Great Bay Estuary (NH), the relationship between land use and water chemistry was investigated by coupling GIS-based land use data with CFC-derived groundwater ages, boron isotopes, and major ion chemistry. Seven submarine groundwater discharge (SGD) sites were selected and then groundwater monitoring networks installed and sampled to examine the relationship between land use and groundwater quality at the discharge zones. Field activities were performed in the summer and fall of 2003. Estuarine water intrusion in groundwater samples compromised the analysis for major ion chemistry and boron isotopes. CFC-derived and lumped parameters-modeled groundwater ages in the study area averaged 23.2 years (±15.0 years). CFC analysis allowed correlating the observed nitrate concentrations at the SGD sites with the land use coverage of 1974 (for most of the sites) or 1962 (for SGD 58.4). Two types of correlation were made: one between the agricultural and residential land use with all the observed nitrate concentrations in the SGD source areas and the other with the nitrate concentrations between developed and undeveloped land uses. Both correlations (Kendall's Tau and Spearman's Rho for each correlating group) indicated the increase of residential land use of the last three decades as being correlated with the high nitrate-bearing groundwater discharging to the Great Bay (NH). The geochemical composition of the SGD water was also investigated by using simple mixing models that attempted to explain the water chemistry characteristics of the targeted SGD sites. Overburden groundwater comprises 75% to 95% of the groundwater discharging at the SGD sites. A significant correlation (Tau's, p=0.021) between nitrate-bearing groundwater and CFC-derived groundwater ages was detected supporting the hypothesis that high nitrate-bearing groundwater will be discharged to the Great Bay in the near future accounting for the increase of residential land use of 1990's. Continuous monitoring of SGD sites was suggested to be included as part of the periodic environmental quality monitoring activities of the Great Bay. Long-term step-wise sampling for groundwater dating is required to develop a stronger chronological evolution of groundwater nitrate inputs. Further research should concentrate on detailing the overburden water chemistry, flow paths, and nitrogen loading characteristics.
B11A-0134 0800h
Impact of Deforestation on Clouds and Rainfall On the Northern Part of the Proposed Mesoamerican Biological Corridor
Central America exhibits the typical pattern of complex deforestation now seen throughout the tropics. The region is a mixture of lowlands, mostly converted to agriculture, and mountainous regions, where pristine forests still persist. To protect the biodiversity of this region from further loss, a network of biological corridors and protected areas has been proposed by the governments of Central American countries and international organizations. The present study examines the impact of deforestation in the northern part of Central America on the proposed corridor network, the Mesoamerican Biological Corridor. We use high-resolution numerical model simulations using the Colorado State University Regional Atmospheric Modeling System (CSU RAMS) to study the impact of three types of conditions: 1) pristine, 2) current and 3) extensive deforestation. In addition, GOES-8 satellite imagery is used for comparing with the numerical simulations of cloud formation. Since vegetation in the proposed protected areas would is under maximum stress in the dry season, this study is focused in the dry season month of March. During the dry season, the soil dries progressively from the soil surface down to increasing depths. Contrary to expectations, in-situ measurements of soil moisture in Costa Rica show similar values both in forests and pastures in the dry season. Measured soil moisture values in March are around 10% of the field capacity in the upper few centimeters, increasing to values of around 30% at depths of 1 m. Yet, observations show that the vegetation in pasture regions is stressed at this time while vegetation in the forested regions is not affected, implying that the forest vegetation is accessing deep soil water. Similar behavior is expected in other regions of Central America. This observation has significant implications to the design of the numerical modeling experiments. Currently the vegetation parameterization used in the RAMS does not specify rooting depth greater that 2m for any ecosystem type. To properly characterize the surface energy budget, the root water uptake was modified based on our field observations to allow for extraction of water from deeper soil layers by the trees. We also used the more recent ecosystem database generated at the University of Maryland from the Moderate Resolution Imaging Spectroradiometer (MODIS) imagery to simulate current conditions. Leaf Area Index (LAI) derived from MODIS satellite observations are used in the simulations and values prescribed for forested and deforested conditions. In addition a soil database was added to RAMS based on the one-degree Food and Agricultural Organization (FAO) soil types and depths. These additions add considerable new capabilities to accurately model conditions in Central America. These changes result in significant modification to the surface energy budgets together with the modification of the thermodynamic profile of the near surface atmosphere during the dry season. We demonstrate that deforestation has serious consequences for the proposed biological corridors. Surface air over deforested areas tends to get warmer and drier, and when these winds flow over forested patches they impact the cloud formation processes and rainfall. Numerical simulations show the locations where these impacts will be the largest, potentially making these regions climatically unstable. The managers of the Mesoamerican Biological Corridor project can utilize this information for better planning.
B11A-0135 0800h
Uptake and Loss of $^{226}$Ra in Tissues of Freshwater Mussels: A Reciprocal Transplant Experiment Between Two Florida Lakes
Round Lake near Tampa Florida has received hydrologic supplements of $^{226}$Ra-rich groundwater from the Floridan aquifer for several decades. High $^{226}$Ra activities have been measured in the water column, recent sediments, and shells and tissues of mussels from Round Lake. Commiston Lake has not received groundwater augmentation and displays relatively low $^{226}$Ra activities in its water column, sediments, and mussel shells and tissues. We relocated mussels from Round Lake to Commiston Lake, and vice versa, to examine the uptake and loss of $^{226}$Ra by mussel tissues. Transplanted mussels were collected from each lake every few days. Mussels moved to Round Lake (the $^{226}$Ra-rich lake) showed rapid uptake of $^{226}$Ra over the 68-day experiment. $^{226}$Ra activity quickly approached the levels in mussels that had always lived there. Mussels moved from Round Lake to Commiston Lake showed no significant loss of $^{226}$Ra from their tissues after 68 days.
B11A-0136 0800h
Spatial and Temporal Patterns of Soil Carbon and Nitrogen Storage Following Woody Plant Encroachment Into Grassland
Encroachment of woody plants into drylands during the past century may have significantly influenced the terrestrial carbon cycle. However, the magnitude and sign of change in soil organic carbon (SOC) pools accompanying this vegetation change is highly uncertain, ranging from positive to neutral to negative. Some of the controversy over woody plant impacts on SOC pools may be an artifact how soil properties determined from point samples are area-weighted and extrapolated. If there is substantial spatial structure in properties of soils associated with woody and herbaceous communities, extrapolations from limited point samples that do not account for this may over- or underestimate actual SOC pools. To test this possibility, we quantified near surface (0-15 cm) soil properties (bulk density, SOC, total N [TN], root biomass) at seven locations along transects extending from the tree bole to canopy edge and into adjoining herbaceous zones in replicated woody communities known to have developed on grasslands over the past 100 y. A strong gradient was found to occur: root biomass, SOC, and TN decreased exponentially with distance from tree boles, while bulk density increased. These spatial changes are consistent with temporal changes expected to occur as shrub establish and their canopies grow through time. Given the strong spatial structure of the data, it appears that area-weighted extrapolations of SOC based on near-bole samples would overestimate woody plant influences, whereas sampling soils away from boles would tend to underestimate impacts. Implications for sampling strategies to efficiently and effectively represent this non-linear spatial variation will be discussed.
B11A-0137 0800h
Analysis of Hydrologic Flowpaths in Two Meso-scale Watersheds, Mt. Mansfield, VT
The various paths by which water moves through forested watersheds are complex and not well understood. Much of our insight into runoff production processes and the effects of landuse are drawn from studies in small watersheds and on experiments involving traditional forest management activities (timber harvesting, road construction). The purpose of this research is to better understand stream water generation processes under varying seasonal and land-use conditions in two meso-scale basins in northern Vermont. An additional aspect of this research is an analysis of the impact of ski resort development on hydrologic flowpaths. We are utilizing stream and potential source water chemistries to characterize the flowpaths within the Ranch Brook (9.6 km$^{2}$) and West Branch (11.7 km$^{2}$) watersheds on the eastern slope of Mt. Mansfield. The West Branch basin encompasses an alpine ski resort while the Ranch Brook basin serves as our forested control site. Event-based samples of stream water have been collected since October 2000, and precipitation, soilwater, groundwater and snowpack samples were collected beginning in late spring of 2004. All were analyzed for their basic solute chemistry and oxygen isotopic signatures. Chemical data and principal components analysis have revealed dynamic systems of stream water generation in both watersheds. Our results provide new insights into runoff production and the effects of landuse activities in meso-scale basins.
B11A-0138 0800h
Spatial Patterns of Soil Organic Carbon Relative to Tree Size and Canopy Distribution in a Semi-Desert Grassland
The abundance of woody species in grasslands and savannas has increased globally over the past century. Recent estimates suggest that this proliferation of woody plants may account for a significant fraction of the Northern Hemisphere C sink, although a large degree of uncertainty exists in the magnitude and spatial distribution of these plant and soil pools. While field-based inventories have made progress in assessing the role of aboveground woody growth in ecosystem C inventories, the effect of woody proliferation on soil organic carbon (SOC) remains controversial, despite the fact that the majority of ecosystem C in these systems is typically belowground. Elevated levels of SOC underneath woody plant canopies have been widely reported, but little is known about the spatial distribution of SOC relative to tree canopies. Understanding the spatial distribution of SOC is critical, however, to developing accurate landscape-scale assessments of woody proliferation impacts on ecosystem C pools. We quantified the influence of encroaching mesquite trees (Fabaceae: {\it Prosopis velutina}) on the concentration of SOC and total nitrogen (TN) in a semi-desert grassland in southern Arizona. SOC concentrations near the boles of large trees (basal diameter 85-102 cm) were approximately double that of SOC in intercanopy zones (0.9% vs. 0.4% SOC by weight). SOC declined moving out from the bole to the canopy edge, at which point it was equivalent to inter-canopy spaces. Small to medium-sized trees (basal diameters less than 85 cm) had minimal influence on SOC concentrations. Patterns of TN values mirrored those of SOC in all cases, although TN values were roughly an order of magnitude lower than SOC values. These data suggest that accurate accounting of landscape-level SOC stocks will require developing area-weighting algorithms that account for tree size and bole-to-canopy gradients.