C43B-0226 1340h
Using a GIS to Facilitate the Reconstruction of the Scandinavian Ice Sheet
Major limitations in the successful use of ice sheet models for climate research arise from limited field data for model verification and the end result may be unrealistic simulations of ice sheet inception, growth, and decay. If a numerical ice sheet model is to be considered an accurate reconstruction of a previous glacial cycle, then the estimated ice marginal positions, subglacial regime, and ice flow charateristics must reasonably agree with field observations. A high resolution, three-dimensional, thermomechanical model was used to simulate the Fennoscandian Ice Sheet (FIS) through the last glacial maximum. Field observations from Denmark, Kola Peninsula, and the Kiruana and Stockholm regions have been digitized and are used for model verification. A Geographic Information System (GIS) was used to integrate model output and field observations (i.e., moraines, eskers, lineations) into one database. A technique that utilizes a system of buffering and overlay was used to determine the proximity and parallel conformity between model and empirical data. Through an iterative process, numerous model runs were tested to find which modeled ice margins best agree with field data. Once the "best fit" model runs were selected, field observations were then used to verify ice flow direction. The end result is a method that links a numerical ice sheet model with field evidence that constrains the marginal extent and general ice flow patterns. Results from this project can improve contemporary knowledge on the evolution of the subglacial thermal regime, as well as the inception and growth of the Scandinavian Ice Sheet during the LGM.
C43B-0227 1340h
Historic Glacier Change using a GIS: Progress and problems
We are exploring the potential wealth of information on glacier extent and topography contained in historic maps. If successful, we can examine the spatial and temporal patterns of historic glacier change where glaciological studies are typically absent. Our project is focused on the American West and data on glacier extent and topography is derived from aerial and ground-based photographs, paper maps, and digital maps. Our initial conception was that digital maps would provide a reliable base on which new data can be added. However, we encountered many errors when compiling the glacier outlines from the 100K scale USGS digital line graph (DLG) coverage. We resorted to overlaying the DLG coverage over the scanned and geo-registered paper maps (digital raster graphics) to error check the DLG coverage. At the time of this writing we are working on the 24K scale coverage, which we obtained from the US Forest Service that uses a different approach to digital map presentation. One important challenge is to define the date of the mapping. Because glaciers change in time, we need to know the date of photography used to make the USGS map. However the data accompanying the maps do not specify what part of the map was updated at the time of last printing. Experience shows that the glacier features are not updated and their outline dates to the original photography from which the first map was made. Although our immediate goal is to assess changes in area, changes in volume are important as well. We found that that historic evaluations of topography (prior to aerial photography) appear reasonable low on glaciers in the ablation zone, higher on the glaciers the topography can be way off and suggests caution in relying on such data sources. While we have found fairly good results in using historic maps, one must carefully examine every map to evaluate its reliability.
C43B-0228 1340h
Late Pleistocene Glaciation and Paleoclimate in the Rocky Mountains: Insights From GIS Analysis
Reconstruction of glaciological characteristics of Pleistocene glaciers is an important source of information on paleoclimate. Depression of equilibrium line altitudes (ELAs) and mass balance modeling have be used to assess the magnitude and character of late Pleistocene-to-present climate change, particularly for alpine glaciers, which respond rapidly to climate. Application of GIS methods to such data facilitates analysis of regional patterns of climate change. GIS studies may also reveal spatial patterns in glaciological response to climate forcing. This presentation considers two applications of GIS to the study of the paleoglaciology and paleoclimate of the southern Rocky Mountains, looking first at late Pleistocene equilibrium-line depression in Colorado and second at magnitudes of climatic change necessary for glacier inception and growth in different ranges throughout the region. The few modern glaciers in Colorado are in topographically favored locations, out of equilibrium with regional climate. Theoretical modern ELAs, more closely related to regional climate, were determined using summer temperature and snow accumulation data from 74 SNOWTEL sites, modern altitudinal gradients of accumulation and temperature, and an empirical equation relating summer temperature to accumulation at equilibrium lines of modern glaciers. A theoretical modern ELA surface was constructed across the state. Late Pleistocene ELAs were determined by accumulation area ratio methods for 149 well-mapped late Pleistocene glaciers, and a paleo ELA surface was constructed. Comparison of the surfaces indicates that the magnitude of ELA depression varied significantly across the state, with a mean of ca 900m. The greatest depression (1000-1250m) occurred in the southern portion of the San Juan Mountains, likely reflecting increased Pleistocene southwesterly moisture flow. Greater than average ELA depression (900-1150m) also occurred in the northern mountains. Central ranges of the state experienced the least ELA depression (600-900m) In most areas there was a pattern of decreasing magnitude of ELA depression away from primary moisture sources, suggesting that the central ranges of the state were buffered from the effects of climate change by the topographic barriers of the marginal ranges. Modern climate data spanning the Rockies from Wyoming to New Mexico were utilized along with the empirical relationship between summer temperature and winter accumulation at the ELA to examine the location and scale of glaciation that would develop in response to different magnitudes of climate change. The results show non-linear relationships between magnitude of climate change and pattern of glaciation, reflecting primarily the topography of areas of potential glaciation. High linear ranges initiate glaciation with small amounts of climate change, but the ultimate extent of glaciers may be limited. Broader areas of high topography, notably the Yellowstone Plateau and San Juan Mountains, do not develop significant glaciers with small amounts of climate change, but with increasing magnitudes of climate change develop extensive glaciation.
C43B-0229 1340h
GIS-Compatible Resources for Cryospheric Research and Planning
Snow, ice, and frozen ground impact a significant part of the earth's surface and natural cycles. Geographic information systems (GIS) are frequently used in studying and assessing the influence of environmental conditions on geophysical, ecological, or socioeconomic factors, but researchers, planners, and decision makers wishing to incorporate cryospheric information into GIS studies often must confront issues related to data formats, data accessibility, and data quality. The National Snow and Ice Data Center (NSIDC) is developing a repository of GIS resources that will simplify the process of identifying, obtaining, and incorporating appropriate cryospheric data into GIS studies. The initial thrust of this repository is base maps, georeferenced images, and other data layers that may be used to study interactions of cryospheric elements with the environment. Supportive narratives, in addition to existing metadata, will describe appropriate uses and explain competing products. Some products, such as the sea ice trends, are derived from satellite data, while others, such as the circumpolar soils map, are based on mapping of data from multiple ground-based and remote sensing sources. The repository will contain materials available from NSIDC as well as links to related products available from other sources. NSIDC considers feedback from existing and potential users regarding the goals, content, and structure of this repository an essential component of its development.
http://nsidc.org/data/gis
C43B-0230 1340h
Creation of integrated analytical GIS-system on the Earth Sciences
The program-technological complex is created on the basis of integrated geoinformation systems (GIS) as which are understood as three dimensional representations of various earths' layers in vector and raster form with the variable dimension. For the management, the distributed access and data processing of such volume the system of meta-computing GRID is used. For a prototype of meta-computing GRID environment the Globes system will be conducted, which is created and used in leading US scientific and computing centers. Execution and analysis of proposed model will be realized by parallel program, written by high-level language with MPI (Message Parsing Interface) using. This parallel program will be executed with user's parameters corresponding to requirements of modern computing clusters. Network components of parallel access and preview of data will be realized in the framework of multi-level model of the corporative server software and J2EE www-service, advanced by Sun Microsystems. At each investigation phase the user carries out processing and multimedia visualization of the received results in time in two and three-dimensional space. The developed multiplanimetric geoinformation system will allow to carry out the integrated analysis of geoinformation streams in an interactive mode, in particular, to reveal laws of existential distribution and dynamics of development of the basic structural lithosphere's elements, and also to establish relationship of stages of their development with epoch of formation of large and superlarge mineral deposits. Funded by RFBR (grants 02-07-90140 and 04-07-90304).
C43B-0231 1340h
Virtual Reality Visualization of Permafrost Dynamics Along a Transect Through Northern Alaska
Understanding permafrost dynamics poses a significant challenge for researchers and planners. Our project uses nontraditional visualization tools to create a 3-D interactive virtual-reality environment in which permafrost dynamics can be explored and experimented with. We have incorporated a numerical soil temperature model by Gennadiy Tipenko and Vladimir Romanovsky of the Geophysical institute at the University of Alaska Fairbanks into an animated tour in space and time in the virtual reality facility of the Arctic Region Supercomputing Center at the University of Alaska Fairbanks. The software is being written by undergraduate interns Patrick Webb and Jordanna Chord under the direction of Professors Chappell and Brody. When using our software, the user appears to be surrounded by a 3-D computer-generated model of the state of Alaska. The eastern portion of the state is displaced upward from the western portion. The data are represented on an animated vertical strip running between the two parts, as if eastern Alaska were raised up, and the soil at the cut could be viewed. We use coloring to highlight significant properties and features of the soil: temperature, the active layer, etc. The user can view data from various parts of the state simply by walking to the appropriate location in the model, or by using a flying-style interface to cover longer distances. Using a control panel, the user can also alter the time, viewing the data for a particular date, or watching the data change with time: a high-speed movie in which long-term changes in permafrost are readily apparent. In the second phase of the project, we connect the visualization directly to the model, running in real time. We allow the user to manipulate the input data and get immediate visual feedback. For example, the user might specify the kind and placement of ground cover, by ``painting'' snowpack, plant species, or fire damage, and be able to see the effect on permafrost stability with no significant time lag.
C43B-0232 1340h
GIS Tool for Real-time Decision Making and Analysis of Multidisciplinary Cryosphere Datasets.
In support of the Western Arctic Shelf-Basin Interaction Project(SBI) a web-based interactive mapping server was installed on the USCGC {\it Healy's} on-board science computer network during its 2004 spring(HLY-04-02) and summer cruises (HLY-04-03) in the Chukchi and Beaufort Seas. SBI is a National Science Foundation sponsored multi-year and multidisciplinary project studying the biological productivity in the region. The mapping server was developed by the UCAR Joint Office of Science Support(JOSS) using OpenSource GIS tools(University of Minnesota Mapserver and USGS MapSurfer). Additional OpenSource tools such as GMT and MB-Systems were also utilized. The key layers in this system are the current ship track, station locations, multibeam bottom bathymetry, IBCAO bathymetry, DMSP satellite imagery , NOAA AVHRR Sea Surface temperature and all past SBI Project ship tracks and station locations. The ship track and multibeam layers are updated in real-time and the satellite layers are updated daily only during clear weather. In addition to using current high resolution multibeam bathymetry data, a composite high resolution bathymetry layer was created using multibeam data from past cruises in the SBI region. The server provides click-and-drag zooms, pan, feature query, distance measure and lat/lon/depth querys on a polar projection map of the arctic ocean. The main use of the system on the ship was for cruise track and station position planning by the scientists utilizing all available historical data and high resolution bathymetry. It was also the main source of information to all the scientist on board as to the cruise progress and plans. The system permitted on-board scientists to integrate historical cruise information for comparative purposes. A mirror web site was set up on land and the current ship track/station information was copied once a day to this site via a satellite link so people interested SBI research could follow the cruise progress.
http://www.joss.ucar.edu/sbi/catalog_hly-04-03/
C43B-0233 1340h
Assessment of Digital Land Cover Maps for Hydrological Modeling in the Yampa River Basin, Colorado, USA
Land cover data are required to parameterize watersheds for hydrological modeling. There is a multitude of different land cover maps, and determining which input data map for the model can be unclear. The goal of this study is to quantify the differences between various publically available land cover maps to determine their relative suitability for hydrological modeling of the Yampa River Basin in northern Colorado. The land cover maps compared in this study are derived from Advanced Very High Resolution Radiometer (AVHRR), Landsat Thematic Mapper (TM), and Moderate Resolution Imaging Spectroradiometer (MODIS) imagery. These maps are compared to a 30-m land cover map modeled from ground data and MODIS imagery. This map is regarded as "truth" in this investigation due to its fine resolution and use of recent ground data and imagery, and will be used to rank publicly available AVHRR and MODIS land cover maps. In order to compare the different land cover products, all data must be degraded to the coarsest spatial resolution (1 km) and the coarsest species resolution. Once this is accomplished, the maps are compared on 4 levels. The 4 comparisons are based on: (i) the relative agreement of the total aggregated land class percentages for the 1-km data present after the data has been cross-walked; (ii) pixel accuracy; (iii) scene accuracy; and (iv) cumulative streamflow model output from the US Geological Survey Precipitation-Runoff Modeling System (PRMS) in relation to observed cumulative streamflow. The results determine the best input land cover data for modeling streamflow in the Yampa River Basin, and provide information about the required spatial, spectral, and classification resolution of these maps to optimize results for streamflow modeling.