C11B-0495
Development of a Climate Data Record of the Clear Sky Surface Temperature of the Greenland Ice Sheet
The objective of this work is to create a climate data record (CDR) of the surface temperature of the
Greenland Ice Sheet from 1981 to the present, on a weekly and monthly time scale at 0.05 deg resolution for
the entire record, and 1 km resolution from 2000 to present. We will use AVHRR data (from 1981) and
MODIS data (from 2000), to the present. A single algorithm with an optimum cloud mask (for each
instrument) will be selected. AVHRR-specific issues include intercalibration within the record (instrument drift,
and different instruments (satellites) used). Also there is a clear-sky bias in the surface temperature
measurements that is a limitation of LST products.
We show preliminary comparison of AVHRR (Comiso, 2006) and MODIS (Wan et al., 2002; Hall et al., 2008)
monthly "clear sky" surface temperature products at Swiss Camp in Greenland for 2000 through 2007.
Results show good agreement between the products, with a coefficient of determination of 0.987, however
the mean MODIS LST is 0.63 deg C higher. The MODIS winter, spring, summer and fall LSTs average 0.75,
0.62, 1.05 and 0.05 deg C, respectively, higher than the AVHRR LSTs. The AVHRR algorithm does not take
into account atmospheric effects, while the MODIS algorithm does. Since atmospheric effects are greatest in
the summer due to enhanced atmospheric water vapor, this may explain why the largest difference between
mean monthly AVHRR and MODIS LST occurs during the summer. In addition, small differences in the
emissivity values used in the different algorithms may impact the resulting LSTs.
Our initial analysis of ICESat data from Laser 2A operations (9/25-11/19/2003) reveals clouds where the
MODIS mask shows clouds on the standard LST map. However, the ICESat profiles also identify areas where
clouds are likely (laser pulses appear impacted) but are not masked as cloud on the LST map. This
indicates that some of the daily LSTs pixels may be contaminated by cloud.
We will validate the CDR using ASTER and ETM+ data as was done in Hall et al. (2008). The CDR will be
useful for monitoring surface-temperature changes of the ice sheet, and as input to general-circulation
models.
References
Comiso, J.C., 2006: Arctic warming signals from satellite observations, Weather, 61(3):70-76.
Hall, D.K., J.E. Box, K.A. Casey, S.J. Hook, C.A. Shuman and K. Steffen, 2008: Comparison of satellite
derived ice and snow surface temperatures over Greenland from MODIS, ASTER, ETM+ and in situ
observations, Rem. Sens. Environ., doi:10.1016/j.rse.2008.05.007.
Wan, Z., Zhang, Y., Zhang, Q. and Li, Z.L., 2002: Validation of the land surface temperature products
retrieved from Terra Moderate Resolution Imaging Spectroradiometer data, Rem. Sens. Environ., 83:163-
180.
class="ab'>
C11B-0496 INVITED
A Northern Hemisphere Snow Extent Climate Data Record
Satellite-derived maps of Northern Hemisphere snow cover extent (SCE) began being produced by National Oceanic and Atmospheric Administration meteorologists in late 1966. Map data have been used in international assessments of climate variability and change, and in investigations regarding the role of snow cover in the climate system. Despite their proven climate utility, meteorological forecasting has been the driving force behind producing these maps. As such, changes (documented and undocumented) in mapping methodologies have occurred over time, without a focus on their climatological continuity. In particular, 1999 brought a change from weekly to daily maps and a greatly increased resolution to the map's digitized grid. Members of our Global Snow Lab have kept a watchful eye on changes in this satellite environmental data record (EDR). From this EDR, we have developed a satellite SCE climate data record (CDR). This presentation will introduce this CDR and discuss efforts that have begun to merge it with other visible and microwave satellite and station-observed estimates of extent and depth over Northern Hemisphere lands, as well as with CDRs being developed for snow melt atop Arctic sea ice and the Greenland ice sheet as part of NASA's Making Earth System Data Records for Use in Research Environments (MEaSURES) program.
C11B-0497
Passive Microwave Bad Scans and the Importance of Satellite Data Version Numbers
One of the most important considerations in scientific research is the quality of the data used for new and important findings. However, a problem with the data, even if it is un-denounced to the researcher can completely nullify the results, especially after it is published. Therefore a version number or another way of identifying the data is needed for all data sets used in research. In creating a climate data record, the quality of the input data also needs to be the best available data at that time with a version number. Several examples can be made where data used in research were not correct and a new and improved version was later released. In the world of satellite data, the need for a version number or build number is increasingly important since data are being released almost as fast as they are produced. An example of this can be made with the SSM/I passive microwave daily gridded data product, archived and available from NSIDC. A detection algorithm was run for each daily SSM/I grid from 1 March through 30 September in 1989. A count of all bad scans that were partially or fully detected was completed for the 19H and 37H grids. The 19H grids had 67 days with at least one bad scan while the 37H data contained 51 contaminated days. Those totals are 28% and 21% of the 19H and 37H days investigated, respectively. Bad scans may occur in multiple channels on the same day, and may also be co-located or partially overlapping. Algorithms and routines used to track changes in polar regions that are sensitive to channel differences or individual channels in the DMSP SSM/I daily gridded brightness temperatures will be affected by the bad scans present in the NSIDC data set. Although the effect of bad scans on individual days may be small, the cumulative effect of bad scans over multiple days or in small subset regions may be significant. Contact with NSIDC revealed that these bad scans are the result of at least two errors: bad data that were not flagged prior to the drop in the bucket averaging and/or differences in the SSM/I gridding routines used before 2000. The problem of bad scans in data was addressed by re-gridding the data using the system employed for data sets for 2001 and beyond. Therefore, the SMM/I gridded data were reprocessed for 1988-2001. The reprocessed brightness temperatures from NSIDC were reviewed and comparisons showed that the bad scans have been removed. Users of the SSM/I gridded data should update their sources with the newly gridded versions. The paper will show how the reprocessed data has changed the results in the melt onset algorithm. However, notifying other users of the data is a difficult task and the use of version numbers at least allows one to identify which data set is being used in the research. The generation of climate data records need to have the capability to use the best data available, but also must be able to utilize new and improved data sets via updates.
C11B-0498
Global Snow Extent Climate Data Records and Trends Derived from Satellite Passive Microwave and Visible Data
The extent and variability of seasonal snow cover are important parameters in climate and hydrologic systems due to effects on energy and moisture budgets. Northern Hemisphere snow cover extent, comprising about 98 percent of global seasonal snow cover, is the largest single spatial component of the cryosphere, with a mean maximum extent of 47 million square kilometers, nearly 50 percent of the land surface area. During the past four decades much important information on Northern Hemisphere snow extent has been provided by the NOAA weekly snow extent charts derived from visible-wavelength polar-orbiting and geostationary satellite imagery. NSIDC distributes these data as the Northern Hemisphere EASE-Grid Weekly Snow Cover and Sea Ice Extent Version 3. Since 1978, satellite passive microwave sensors have provided an independent source for snow monitoring, with the ability to penetrate clouds, provide data during darkness and the potential to provide an index of snow water equivalent. The historic microwave record spans a thirty year period and data are available from NSIDC as the Global EASE-Grid Monthly Snow Water Equivalent Climatology Product. Both data sets have been updated through spring, 2008. Trend analysis on the passive microwave record is complicated by the short overlap period of SMMR and SSM/I in 1987. To derive a consistent map of passive microwave snow cover, we examined the temporally closest overpasses from each sensor at selected targets and derived regression equations to cross-calibrate the sensors. Passive microwave snow algorithms have also consistently overestimated snow cover on the Tibet Plateau. We attribute the overmeasure to the use of algorithms that have assumed a thick atmosphere. These algorithms overmeasure snow extent when applied to very high elevation surfaces. We have derived an atmospheric correction to compensate for the influence of the reduced atmospheric thickness on snow extent estimates. Using the latest improvements to our passive microwave algorithms, we present time series of monthly anomalies for Northern Hemisphere snow extent derived from passive microwave and visible data from 1978 to 2008. We see both positive and negative trends in snow extent derived from these data, depending on region and time of year.
C11B-0499
Expanding research capabilities with sea ice climate records for analysis of long-term climate change and short-term variability
Recent sea ice analysis is leading to predictions of a sea ice-free summertime in the Arctic within 20 years, or even sooner. Sea ice topics, such as concentration, extent, motion, and age, are predominately studied using satellite data. At the National Snow and Ice Data Center (NSIDC), passive microwave sea ice data sets provide timely assessments of seasonal-scale variability as well as consistent long-term climate data records. Such data sets are crucial to understanding changes and assessing their impacts. Noticeable impacts of changing sea ice conditions on native cultures and wildlife in the Arctic region are now being documented. With continued deterioration in Arctic sea ice, global economic impacts will be seen as new shipping routes open. NSIDC is at the forefront of making climate data records available to address the changes in sea ice and its global impacts. By focusing on integrated data sets, NSIDC leads the way by broadening the studies of sea ice beyond the traditional cryospheric community.
C11B-0500
Air and Soil Temperature Climatology of the Appalachian Mountain Range
The Appalachian Mountain Range extends nearly 2000 km and contains summit elevations greater than 2000 m. Considerable diversity of climate exists within this complex terrain, ranging from temperate deciduous forests to alpine tundra conditions suitable for the development and maintenance of permafrost. In August 1996, the Appalachian High-Elevation Network (AHEN) of climate stations was established to collect baseline air and soil temperature data at twenty summit and other high-elevation locales where climate monitoring has traditionally been absent owing to logistical difficulties. Each site was instrumented with miniature temperature loggers that record hourly measurements of air and ground temperature at the surface and depths of 10, 25, and 50 cm. The sites are arrayed on a 1500 km transect along the spine of the Appalachians with endpoints in Maine and North Carolina. Analysis of these ten-year climate records has revealed strong relations between air temperature, snow cover, and the depth and duration of ground freezing events. Observed lapse rates diverge significantly from the standard environmental lapse rate, and as such, are more suitable for use in the construction of regional air-temperature fields that are used in a wide range of applied studies, including ecology, forestry and permafrost.
C11B-0501
Simulation of Long-Term Response of the Greenland Ice Sheet to Global Warming With an Ice Sheet Model Coupled to a Regional Energy-Moisture Balance Climate Model
Using the 3D, thermo-mechanical ice sheet model SICOPOLIS coupled to a simple, regional energy-moisture balance climate model, we simulated the response of the Greenland ice sheet under various global warming scenarios. Until now, the usual approach to specify surface boundary conditions for ice sheet models has been to use present day temperature and precipitation distributions in combination with anomalies to scale the values to the past or future. This method is only justified, however, when the ice sheet area and elevation remain similar to present day, because it assumes that the patterns of temperature and precipitation remain similar to present ones. However, it is likely that the distributions of temperature and, especially, precipitation would be much different for a partially or completely ice-free Greenland. In our approach, the climatology used to force the ice sheet model explicitly accounts for albedo feedback and elevation changes. This is important for long-term (multi-centennial to multi-millennial) climate change scenarios, in which the Greenland ice sheet could melt completely, since the albedo feedback would produce higher temperatures in the interior of Greenland, altering the temperature pattern from the current distribution, and the precipitation pattern, particularly in Southern Greenland, would be strongly affected by elevation changes. We present results of simulations for several long-term global warming scenarios and compare them with those found using the traditional (anomalous) approach. We also performed a stability analysis of the Greenland Ice sheet in CO2 phase space by performing a set of equilibrium experiments for different CO2 concentrations and initial conditions.
C11B-0502
The Effects of Summertime Melt on Stable Water Isotopes: Implications for Paleotemperature Reconstructions from Firn Core Records
A firn core was drilled on the Prince of Wales (POW) Icefield, Ellesmere Island, Nunavut, in the spring of 2001. While this firn core was drilled on the high northern plateau of the Icefield, its location is still prone to occasional summertime melt events. Annual δ18O amplitudes of 5.7±2.3‰ persist in the firn core stratigraphy indicating that the degree of melt at the site is insufficient to destroy annual isotopic signals. However the degree of isotopic modification resulting from meltwater and its effects on average annual temperature estimates from the site are not well understood. A series of four early melt-season snow and dye pits were established along an altitudinal gradient extending from 380 m a.s.l. to 1000 m on the Leffert Glacier of the POW Icefield. These snowpit sites are intended to serve as proxies for the degree of isotopic modification occurring at the firn core drill site. Two factors are considered potentially important in determining the degree of isotopic modification experienced at a site, 1) the absolute amount of melt, and 2) the percent melt. The two lowest elevation sites, TM380 and TM600, experience both absolute and percent melt values similar to those observed at the firn core drill site over the past 28 years, as determined from NARR 2-m surface air temperature data and firn core stratigraphies. Since TM380 and TM600 had average δ18O enrichments of 1.1 and 0.5‰ respectively, we infer that the average annual temperatures derived from the firn core drill site are overestimated by 0.5-0.2°C based on the local δ18O-temperature relationship. Results from this analysis suggest average annual temperatures are overestimated when derived from firn core records that experience even occasional melt events.
C11B-0503
Records of Iceberg Rafting at the North Siberian Margin Linked to Regional ice cap History
Concentration and composition of lithic grains >500 microns regarded to be largely an ice-rafted debris (IRD) were analyzed in two AMS14C-dated cores from the western Laptev Sea. The core from the upper continental slope (270 m) dates back to c. 17.6 cal.ka, and the core from the outer shelf (60 m) to 12.4 cal.ka. On the continental slope, two evident spikes in IRD concentration are recorded, one in the basal unit with extrapolated age of 17.6-16 cal.ka, and another one in topmost sediments younger than 7 cal.ka. Although quartz is dominant throughout the sequence, the composition of rock fragments differs between the spikes with the diversity of rocks being higher in the older unit. The distinguishing feature of the lower IRD-rich interval is the presence of numerous concretions of vivianite and rhodochrosite formed diagenetically under anaerobic conditions. The anaerobic bottom water environment was a result of a strong water stratification caused by meltwater input from the Barents-Kara ice sheet and also by river runoff as the former coastline and river mouths were located at the shelf break close to the core site. This explains the diverse composition of lithics due to a contribution of iceberg-rafted fragments from the decaying Barents-Kara ice sheet and material brought by rivers from the continental hinterland. After 16 cal.ka, the IRD concentration in the core sharply decreases likely evidencing the retreat of the ice sheet from the shelf break and cessation of iceberg production. The upper IRD-rich interval in the core from the continental slope correlates in time (<7 cal.ka) with the spike of IRD concentration in the core from the outer shelf. In both cores the composition of the IRD points to archipelago Severnaya Zemlya as the major source region. Ice caps on Severnaya Zemlya were considerably reduced in size during the LGM. Their regrowth and increasing iceberg production were induced by climate cooling and strong intrusions of Atlantic derived waters accompanied by moisture supply to the western Siberian continental margin.
C11B-0504
Evaluating the Performance of a New-Generation Numerical Weather Prediction Model in Representing the Antarctic Surface Layer
Recent atmospheric boundary layer literature has documented that the numerical weather prediction models are not very reliable under stably stratified conditions. The discrepancies between the observations and corresponding model forecasts become more pronounced in high latitude regions where stable boundary layer persists (almost) throughout the year. It is worth pointing out that most of the present-day mesoscale and large-scale atmospheric models use stable boundary layer parameterizations, which are not physically based but inspired by model performance. Some of these ad-hoc parameterizations alleviate problems like 'runaway-cooling' by artificial enhanced mixing, but at the same time create unphysical consequences, such as unreasonably deep boundary layers and weaker low-level jets. More importantly, most of these parameterizations are in general tuned to give reasonable overall model performance in mid-latitudes, and they do not provide satisfactory accuracy for the Polar regions. In this presentation, an attempt has been made to evaluate the performance of a new-generation numerical weather prediction model – called the Weather Research and Forecasting (WRF) model in representing the stable boundary layer characteristics over South Pole. High-resolution turbulence observations from the field campaign called 'Investigation of Sulfur Chemistry in the Antarctic Troposphere' (ISCAT) are used for validation. The WRF configuration used in this study is quite similar to that of the operational Antarctic Mesoscale Prediction System (AMPS). Some changes have been made in the model to give optimum performance over the Polar regions. The main focus of the presentation will be on comparing the WRF model-derived surface layer parameters such as near-surface wind speed, sensible heat flux, frictional velocity etc. with observations. The authors will also briefly touch upon the intriguing dual character of the surface sensible heat flux in the Antarctic surface layer.
C11B-0505
Comparison of Navier-Stokes, Pattyn's higher-order, and McAyeal's vertically-integrated finite element models on Pine Island Glacier, West Antarctica.
To improve predictions of the evolution of the Antarctic Ice Sheet in a warmer climate, it is critical
to implement new models with a full description of the glacier physics. This work compares
three different finite element models of a major West Antarctic ice stream, the Pine Island Glacier.
The first model is the vertically-integrated model of McAyeal (1989), the second one is Pattyn's (2002)
higher-order model where horizontal and vertical velocity components are computed separately, and the
third one is a full Navier-Stokes solution. First, we examine differences in predicted speed from
all three methods using the same constraints on basal drag and ice rigidity. Many discussions have
addressed
the applicability of these models and their domain of validity. In particular, we examine three-dimensional
effects
where ice flows over bumps and hollows in bottom topography which can only be faithfully reproduced with
the
full Navier-Stokes equation. We also examine the flow evolution predicted by all three models if the grounding
line is abruptly retreated while keeping other parameters constant. The result illustrates the level of
sensitivity of the different solutions to perturbations in grounding line position. We present conclusions
on model recommendations based on the reproductibility of the flow field, strain rates and stresses, the
sensitivity to grounding line migration, and computational cost.
This work was performed at the California Institute of Technology's Jet Propulsion Laboratory under a
contract
with the National Aeronautics and Space Administration's Cryosphere Science Program.
C11B-0506
Spatially Distributed Temperature-index Melt Modelling of Glaciers in the Donjek Range, St. Elias Mountains, Yukon Territory
We apply both classical and enhanced spatially-distributed temperature-index melt models to two small valley glaciers in the Donjek Range of the St. Elias Mountains (Yukon, Canada) in an effort to accurately model glacier melt in this region. The enhanced model is an extension of the classical approach with the addition of potential clear-sky direct solar radiation as developed by Hock (1999). The two study glaciers are located ~10 km apart and are of similar size (~5 km2), but different aspect (north/south) and dynamic regime (normal/surge-type). We calibrate the melt models on one glacier in 2007 and evaluate the models on the same glacier in 2006 and the second glacier in 2007, providing insight into the potential for model extension in both time and space. To drive the melt models we use four to seven temperature records from distal and proximal automatic weather stations (AWS), including AWS located on the glacier surfaces. Model output consists of hourly simulated melt rates at 30 m spatial resolution across each glacier surface. Model skill is examined through comparisons with cumulative surface lowering measured at ablation stakes, as well as with hourly measured surface lowering and hourly melt simulated by an energy balance model at AWS locations on each glacier. Results show that although air temperature lapse rates vary in time and space, daily air temperature over the glaciers can be estimated from off-ice temperature records with reasonable accuracy (maximum r2 = 0.97) when constant lapse rates are applied. With the enhanced model, high model skill in simulating both daily and seasonal melt rates can be achieved even in the absence of accurate estimates of air temperature. Off-ice temperature records generally yield higher model skill than on-ice records. Optimized model parameters (i.e. melt factors) are similar for the two glaciers in 2007, but differ significantly from 2006 to 2007. As a result, the models show greater skill in predicting cumulative melt on the second glacier than predicting melt on the calibration glacier in a different year: mean absolute errors (MAE) are 18 and 23%, respectively. This difference in model skill is partly explained by the surface energy balance computed at the glacier AWS locations: the energy balance is similarly partitioned on the two glaciers in 2007 but shows higher net radiative flux and lower turbulent fluxes in 2006 compared to 2007 on the calibration glacier. The classical 'degree-day' approach matches the skill of the enhanced temperature- index model for cumulative melt simulation during model calibration, but the enhanced model exhibits superior performance in the model evaluation (MAE 11 to 14% lower than classical model) and for daily and sub- daily melt-rate simulations. The spatial transferability of enhanced temperature-index model parameters in this study suggests the potential for reasonable estimation of distributed glacier melt within the study region with minimal data input.
C11B-0507
Effects of snow accumulation on soil temperature and change of salinity in frozen soil from laboratory experiments
In order to clarify the effect of snow depth on the ground temperature, snowfalls were occurred on soil samples using an artificial snowfall machine in the laboratory and variations of soil temperatures up to 30cm were measured during snowfall. The snow types used here were dendrites (type A) and sphere (type B). The snow depths on the soil surface were 10cm and 30cm for each snow type, so four deferent experimental results were obtained. At each experiment, two samples with deferent initial volumetric water content were prepared, about 10% and 20%. The initial soil temperature was set to 5°C and temperature in the laboratory was kept at -10°C. Soil temperatures were measured at the depths of 0cm, 10cm, 20cm and 30cm during the snowfall, and continuous measurements were conducted for ten hours after the stop of snowfall. From the experiments, it is confirmed that the soil temperature strongly depended on the depths of snow on the surface, density and water content. The soil sample using the type A with the depth of 30cm snow accumulation had the highest temperature at the surface, followed by the type A with 10cm snow, type B with 30cm snow and type B with 10cm snow. It was also pointed that temperature of the high water content samples showed the high temperature decrease compared with the low water one due to the high heat capacity except for the sample using type A with 10cm snow. Numerical calculation will be needed to explain these results. In addition, another experiment will be carried out to clarify the change of salinity during soil freezing with snow accumulation. The method to measure the salinity of soil is to measure the electrical conductivity of soil and volumetric water content at the same depth. The temperature condition in the cooling bath is ranged between -10 and 5°C and changed in 24 hours. Firstly, the temperature profiles will be measured to detect the frozen front, then measurements will start and discuss the results.
C11B-0508
Quantification of Glacimarine Sediment Yields Using Multibeam Sonar in Alaskan Fjords
Accurate measurement of sediment yields and inferred denudation rates is critical to assess the role of glaciation in the complex interconnection between climate and tectonics in orogenic regions. Glaciers have long been recognized as efficient modes of erosion, yet questions persist regarding potential variations in total effective erosion during different phases of tidewater glaciation. It has been argued that sediment accumulation is greatest during the retreat phase as both calving rate and ice flow velocity increase. This acceleration can potentially cause overestimation when projecting sediment yields and denudation rates beyond modern conditions. In this study we present a method to quantify temporal bathymetric changes to measure glacial sediment flux. We find that after simple processing, historical bathymetric data can be used in conjunction with multibeam bathymetry to accurately and effectively measure sea floor sediment accumulation using geospatial software. This method provides the greatest accuracy in quantifying spatial variance in sediment accumulation and therefore has advantages over point and line-derived measurements. From these data we calculate accumulation volumes, yields, and inferred denudation rates from two contrasting modern glacial systems in Southeast Alaska. Hubbard Glacier has steadily accumulated mass and advanced since at least 1895. This provides a unique opportunity to measure accumulation rates dominated by a glacier in its advance phase. Conversely, as with most Alaskan glaciers, Muir Glacier has dramatically retreated and lost mass over the last few centuries. Measured rates associated with these glaciers are compared and demonstrate sedimentation and denudation rates are analogous during both advance and retreat phases of the tidewater glacier cycle. This analysis allows for a better understanding of sedimentation patterns with tidewater glacier cycles in addition to insight on the interplay of glacial landscape denudation associated with tectonic uplift. This has ramifications not only to tectonic uplift but also to creating differential relief and over-deepening of fjord basins.
C11B-0509
Glacier and lake variations in the inland closed basins at high altitude on the Tibetan Plateau using remote sensing and GIS technologies
As a contribution to studies of the impact of climate change on glaciers and lakes in high-altitude closed basins of the western Himalayas, we present spatial and temporal variations of glaciers and lakes in the Mapam Yumco Basin on the Tibetan Plateau, by means of Geographical Information System and Remote Sensing techniques. Our results show that both glacier and lake areas in Mapam Yumco Basin decreased from 1974 to 2003. Glaciers in the basin have receded due to the warmer climate, in total by ~ 7.53 km2 (0.26 km2 a-1 or 0.25 % a-1) during 1974¡§C2003 (c.f. ~ 0.07 % a-1 in nearby Yamzhog Yumco Basin, and 0.18% a-1, the mean glacier recession rate over China since the 1960s). During the same period, lake area decreased by 34.16 km2 (1.18 km2 a-1 or 4.37 % of whole lake area in the basin) in total, where decreased by 1.43 km2 a-1 on average (with lake shrinkage amounting to 1.70 km2 a-1 in some areas and lake growth to 0.27 km2 a-1 in others) during 1974¡§C1990, by 1.55 km2 a-1 (with lake shrinkage amounting to 2.15 km2 a-1 in some areas and lake growth to 0.60 km2 a-1 in others) during 1990--1999, while enlarged by 0.66 km2 a-1 (with lake shrinkage amounting to 2.24 km2 a-1 and lake growth to 2.89 km2 a-1) during 1999--2003 over the past three decades. It is suggested that both enlargement and reduction of lakes were accelerated, which might be an indicator for an accelerated water cycle process over the Tibetan Plateau in a warming climate.
C11B-0510
Stress and strain-rate comparisons between observed in situ buoy arrays and ice models
The modeled material behavior of sea ice traditionally assumes some form of elastic/viscous/plastic rheology. To date, however, this assumption has never been validated against observed measurements at the scale of the assumptions. Using an archive of in situ stress buoys with GPS positioning from Beaufort Sea ice campaigns in 1993, 1998, 2001, and 2007, we present a comparison between the observed stress and strain-rate archive and the Los Alamos CICE model version 3. The formulation of observations in a format compatible with model output and preliminary comparisons with the CICE model will be shown.
C11B-0511
Comparison between Ground and Helicopter Electromagnetic Sea Ice Thickness Measurements from SEDNA during APLIS07 in the Beaufort Sea
The Sea-ice Experiment: Dynamic Nature of the Arctic (SEDNA) project took place in April 2007 in the
Beaufort Sae as part of the first ice camp of IPY. Central to the SEDNA project was an array of six 1km
survey lines, laid out across all available ice types. Among the many experiments conducted along the survey
lines was a comparison between a helicopter-borne electromagnetic induction device (HEM), ground EM
profiles using an EM-31, and snow and ice calibration drill holes. Here we report and compare all these
measurements of ice thickness, made over the same place at the same time in the Beaufort Sea. We extend
this comparison to similar measurements made in another experiment conducted in 2004 in the Baltic Sea
over seasonal level and deformed ice. Results are used to assess the capabilities and natural limitations of
the instrumentation including insight into the geophysical interpretation of their differences. Despite the
footprint limitation of EM systems to resolve the structure of deformed sea ice on a local scale, histogram
analysis shows that the mean and median of sea ice derived from these surface-only measurements provide
a good representation of the mean and median thickness of sea ice as well as the thickness distribution of all
first year ice (both level and deformed).
http://research.iarc.uaf.edu/SEDNA/
C11B-0512
Long-Term Changes In The Behaviour Of Jakobshavns Isbrae, West Greenland During The Late Quaternary-Holocene
This poster shows the initial results of a joint scientific project to reconstruct the Late Quaternary-Holocene behavior of Jakobshavns Isbrae in central west Greenland, one of the largest ice streams draining the modern Greenland Ice Sheet. The underlying rationale for this research is to determine if recent observed changes to the mass balance of the Greenland Ice Sheet are part of the natural variability in ice-sheet dynamics, or if they relate to anthropogenically-induced climate warming. Key to resolving this question is an understanding of long-term changes in ice sheet behavior during the Late Quaternary and the Holocene. This research will allow assessment of the links between deglaciation and internal and external environmental controls, such as the influence of inflowing Atlantic Water, and will facilitate modelling of the likely future behavior of the GIS. Currently, four marine sediment cores arrayed along a transect from the Disko Bugt Fan to Disko Bay are providing information on changes in sediment flux and sedimentation style, such as abrupt intervals of iceberg-rafting vs. "normal" hemipelagic sedimentation, as well as the paleoceanographic setting and ice sheet-ocean interactions. The cores are being analysed using a variety of proxies including IRD, mineralogy, oxygen isotopes, foraminiferal assemblages, lithofacies analysis and AMS radiocarbon dating. Data are presented from two piston cores from the continental slope at the trough-mouth fan collected during the HE0006 'shakedown' cruise to Baffin Bay and from two gravity cores recovered in 2007 during MS Merian cruise MSM 05/03 to West Greenland. Slope cores contain sequences of laminated facies interpreted as fine-grained turbidites and intervals of massive, bioturbated, hemipelagic mud. The two Merian cores, contributed to this project by the Baltic Sea Research Institute, were collected from the southern entrance to Disko Bugt and the Vaigat channel north of Disko. Radiocarbon dates from the Disko Bugt core show that it contains a full Holocene record of glacial activity and paleoceanography. The poster will present the initial analyses, including radiocarbon dating, XRF compositional data, magnetic susceptibility, lithofacies and IRD analyses determined from x-radiography, foraminiferal analyses and sediment mineralogy. Additional cores and seismic data for this project will be obtained from a cruise on the Canadian research vessel, CSS Hudson in September 2008, and on the British ship, the RRS James Clark Ross in 2009.
C11B-0513
Surface roughness heights derived from wind and temperature profile measurements: comparison with eddy correlation. Case study of a high altitude glacier in the Bolivian Andes.
The turbulent fluxes remain poorly known on tropical glaciers. Different studies based on the bulk method have shown that sublimation can be important during the dry season, reducing the energy available for melting. However, uncertainties on the bulk method are large, especially when the katabatic wind causes a wind speed maximum at low height. Eddy correlation measurements have been conducted at 5050 m elevation on the Zongo Glacier, Bolivia (16°S) in July-August 2007. Concomitant measurements of all radiation components, snow surface temperature and of vertical gradients of air temperature and wind speed were made. The site was approximately level within several hundred meters, with drainage winds prevailing at night and most of the day. The surface roughness heights for temperature and momentum were derived from the profile measurements at the hourly time scale. Results indicate values from 0.1 to 1 cm in rough agreement with terrain observations. However, the uncertainties on the roughness height for temperature are large and the relation of Andreas (1987) can not be thoroughly tested. The results show that in this dry and thin atmosphere, surface temperatures below freezing are maintained with significant cooling of the surface due to sublimation and low long-wave radiation input. The agreement between the bulk and the eddy correlation methods is rather good, except during periods of calm wind when errors are large in both methods. The roughness height for momentum derived from the eddy correlation is also in good agreement with the heights derived from the profile measurements. However, the measurement period of the eddy correlation is short and more work must be done to quantify the errors on the method.