MasS of the Polar Ice Caps Experiment (MSPICE)
L. Avallone, A. Gettelman, D. Noone, D. Toohey
Draft: March 1, 2006
Purpose
The following is a brief science outline of the MSPICE project for the International Polar Year. The PI's are soliciting community interest and participation in shaping this document. Please send comments to Andrew Gettelman (andrew@ucar.edu)
Introduction
Ninety-five percent of the freshwater on Earth is stored in the ice sheets of Antarctica and Greenland. Greenland ice accounts for an equivalent of seven meters of sea level, while Antarctica 60 meters (Houghton et al., 2001). The amount of snow deposited annually on the Antarctic ice sheet is equivalent to about 6.5 mm of global sea level, which approximately equals the mean annual discharge of ice back into the ocean. Despite all available measurements of snow accumulation, ice velocities, surface altimetry, surface and basal melting, and iceberg discharge, it is still not known for certain whether the Antarctic ice sheet as a whole is growing or shrinking. On the other hand, the Greenland ice sheet is shrinking at a rate that is uncertain, and a recent study showed that the rate of shedding of icebergs has doubled in past years. The uncertainty in the estimate of the total mass balance in Antarctica is at least 20% of the mass input as atmospheric moisture convergence, which is equivalent to a global sea-level change of 1.5-2 mm per year.
We propose a cross-disciplinary investigation that combines the expertise of investigators who work on climate, paleoclimate, atmospheric chemistry, cloud microphysics, and polar and tropical hydrology in an effort to better understand the mass balance of the polar ice sheets and the implications of changing that balance for climate and paleoclimate. More than just an independent assessment of surface mass balance, this project will be a detailed examination of the atmospheric conditions and feedbacks that govern ice formation and stability under very low temperature conditions throughout the atmosphere. This has relevance for understanding the processes that determine deposition rates of ice in the polar regions as well as those that produce the driest air in the atmosphere, which is found in the tropical upper troposphere and lowermost stratosphere. Finally, these studies will address the roles of natural and anthropogenic particles that may either enhance or inhibit the formation of ice clouds.
A change in the rate of iceberg discharge from Greenland and Antarctic ice streams in recent decades and centuries suggests that the surface mass balance (precipitation minus evaporation, or simply the accumulation) also may change rapidly. Recent work has shown that the Greenland ice sheet is ablating at a rate that gives rise to serious concern about the future of the ice sheet (Steffen et al., 2004). The Antarctic ice sheet is more stable, in part due to its proximity to the pole. Indeed while the increase in temperature seen in instrumental records near Greenland is larger than the global average, Antarctica shows little trend at all, if not a cooling. These observations have been confirmed by inferred temperature from isotopic composition of shallow ice cores (Schneider et al., in press). These results highlight that the sensitivities of these two regions to global climate forcing, is dramatically differently, such that investigation of both is necessary to adequately understand the mechanisms and processes that drive change in global hydrology and climate.
Recent work in the Arctic has shown that processes influencing nucleation of ice in the atmosphere can substantially change ice cloud properties, hence affecting strongly the longwave radiative fluxes (Garrett et al., 2001) in this highly sensitive region. Aerosol loading in the Arctic has been shown to be important in this context. As a case in point, it had been suggested that with increased aerosol concentration, increased nucleation of cloud particles would increase the optical depth of arctic clouds and thus cause a net cooling of the surface. This effect has recently been observed in some areas of the Arctic, and the cooling is indeed greatest near large industrial sources of sulfate in Northern Russia. Should a larger number of ice condensation nuclei be available, the degree to which vapor is supersaturated decreases. As such, the radiative properties of clouds, via an aerosol- and cloud condensation nuclei-dependence, are strongly linked to the degree of supersaturation. Thus the influence of aerosol effects on the surface mass balance comes first from a direct change to the hydrologic budget, but second as a result of radiation and energy balance changes, which in turn require a modified hydrology. The role of aerosols and supersaturation in controlling the surface mass balance of ice sheets is presently unknown.
The conditions for the formation of ice in the atmosphere and aerosol perturbations to ice formation processes are not important only in polar regions. There are important implications for how ice particles form in the tropical upper troposphere, where they affect climate through modifications to ice cloud (e.g., cirrus) dynamics. Ice cloud formation in the tropics may also significantly influence the water and hydrogen budgets of the upper atmosphere, with resulting impacts on stratospheric chemistry, and ultimately on stratospheric ozone.
Stable isotope composition of ice cores from the ice sheets has long been used to diagnose the integrated history of temperature and mass accumulation in the polar regions. Isotopic fractionation occurs during condensation, with an efficiency dependent on temperature. Isotopic fractionation is also affected by supersaturated conditions in clouds, which may complicate interpretation of ice core records. Indeed the difference between the two most common minor water isotopes is thought to be dominated by supersaturation in the Antarctic interior. Measurements of saturation conditions and simulations of these conditions and their affect on isotopes over the ice sheets can help better understand isotopic composition of polar snow and provide a solid foundation based on governing physics, rather than simple empirical associations, to allow for more accurate interpretation of polar climate proxy records.
Science Questions
The investigation into ice and snow formation over the ice sheets of Antarctica and Greenland has several important implications. We will focus the discussion on the following linked questions and hypotheses.
1. Ice sheet mass balance: What is the rate of accumulation or shrinking of the Antarctic and Greenland ice sheets?
Hypothesis: It is assumed that the Greenland ice sheet is shrinking, but the rate is uncertain. The sign of accumulation for the Antarctic is unknown. Based on newly available satellite humidity observations, constrained by targeted in situ validation, and comprehensive earth system models, we should be able to estimate the accumulation.
2. Ice formation in the atmosphere: How does ice form in the atmosphere over the ice sheets? How important is the anthropogenic impact on ice formation?
Hypothesis: It is known that nucleation of ice is a complex process that does not occur at 100% saturation over ice. It is likely strongly influenced by the distribution of atmospheric particles that act as ice nuclei. With new data from high altitudepolar regions on humidity, particles and aerosols, we will probe the conditions that favor ice formation in the atmosphere. We will learn how changes to the composition and size of aerosol particles impact ice formation. This has important implications not just for polar regions, but also for tropical ice cloud formation, and hence for global climate.
3. Interpretation of ice core records: What is the isotopic composition of snow deposited onto the ice sheets and how does it vary with changes in the meteorological factors that govern ice cloud formation?
Hypothesis: Stable isotopes of water provide an integrated history of the environmental conditions when condensation occurred. Observations and modeling of the isotopic composition of water vapor and condensed-phase water (precipitation, snow, and ice) in polar regions will enable us to better understand the origins and formation history of current ice, which has significant bearing on attempts to interpret ice core records of isotopes, which give paleoclimate temperatures.
Methodology
We propose an integrated, multi-disciplinary investigation based on in situ sampling, remote sensing and modeling to answer the questions posed above. The MSPICE project has these three major components, and will be integrated with other international activities surrounding the International Polar Year.
1. In-situ sampling
MSPICE will coordinate measurements in Greenland and Antarctica of specific and relative humidity, ice content and ice particle number, size, and mass, aerosol number and size, and the isotopic composition of ice and vapor. Some of the investigations will be conducted as part of the MSPICE project, and some observations will be taken by other projects to be conducted during the IPY. Data will be from ground-based and balloon profiling platforms. MSPICE investigations will focus on field campaigns at the South Pole and in Greenland. Cooperation with other projects will also include observations at other polar locations during the same time period (for example: radiosonde launches, snow sample information, basic meteorological data).Specific humidity - Due to the reliance upon highly accurate humidity data, measurements of water vapor will be carried out with multiple complementary techniques. Numerous measurements from research balloons and aircraft have demonstrated that these measurements can be made routinely with ~20% accuracy at the extremely dry conditions prevailing in the high altitude polar regions. However, there are some systematic differences in the observations reported by different techniques. By using several techniques that differ in the fundamental parameter upon which the specific humidity measurement is based, higher accuracy overall can be obtained. The most desirable techniques for these studies are frost-point hygrometers and narrow-band infrared laser spectrometers. Observations will be obtained from fixed (e.g. tower) ground stations and occasional balloon launches.
Relative Humidity- RH is necessary to help understand nucleation processes and thresholds. Calculation of accurate temperatures will be critical, and so the MSPICE project will also make sure that high-quality temperature measurements accompany specific humidity measurements.
Ice content û These observations will be carried out using two complementary techniques, evaporation followed by quantification of resulting water vapor ("total water") and laser particle scattering.
Ice ûparticles - There are many possible in situ ice particle measurements. We are currently soliciting community participation In situ measurements will also be used to constrain and validate satellite retrievals of clouds and cloud optical properties over the ice sheets.
Aerosol There are many possible in situ aerosol measurements. We are currently soliciting community participation in this component. û Satellite retrievals of aerosol properties over the polar caps and adjacent oceans will be used to help characterize the background population of nucleating aerosols. These observations will be validated with in-situ observations as part of MSPICE, as well as using data from other projects taken during the IPY.
Water Isotopes/Vapor to be measured by TDL, snow/ice to be measured by mass spectrometryàmore here from Darin and JimWhite.
2. Remote sensing
MSPICE will also use satellite humidity observations and global analyses to develop an estimate of the mass balance of the Greenland and Antarctic ice caps by computing the atmospheric water budgets at high latitudes. This will include evaluating the importance of ice supersaturation and, by inference, the importance of particle nucleation in the processes governing the accumulation. The budgets will be validated by the in-situ data collected as part of this project, and extended backwards and forwards beyond the IPY period as the satellite data permit. We expect that, if the satellite data are well validated by MSPICE and related in situ data, we will have up to 7 years of detailed mass budgets.
Specifically, we expect to rely on retrievals of atmospheric humidity from the NASA Atmospheric Infrared Sounder (AIRS), and cloud and aerosol properties from AIRS and the Moderate resolution Imaging Spectroradiometer (MODIS), which are in polar orbits and provide good daily coverage of polar regions, starting from mid 2002 and expected to continue through the IPY period and beyond.
3. Polar Modeling
We will perform various scales of modeling to better understand the in-situ and satellite data. Detailed microphysical models will be used to understand the ice formation process, potential aerosol impacts, and the impacts on the isotopic composition of models. We will also use global models to cover the entire polar region. These simulations, in conjunction with observations will be used to determine the source regions of water deposited to the ice sheets, to guide evaluation of the isotopic records from ice cores and to study how the hydrologic cycle over the ice sheets changes with climate. The detailed in-situ data sets and microphysical models will also be used to help improve the ice phase of the global hydrologic cycle in these models, which is also critical for understanding climate, not just in polar regions but globally.
Specific models include the NCAR Community Atmosphere Model (CAM), a state of the art General Circulation Model which includes an isotope package. CAM also has options for including a detailed microphysical scheme that can be compared with in situ observations
More information on the International Polar Year and MSPICE can be found at http://www.ipy.org.
Contact: David Noone (dcn@colorado.edu)