authorship: U.S. CLIVAR High Latitude Flux Working Group plus other contributors
Abstract : to be done last.
Outline:
Introduction
Surface fluxes in High Latitudes
Current Capabilities
Direct
Radiative
Freshwater
Heat and momentum
Gas
Measurement programs
Satellite
Direct
Radiative
Freshwater
Heat and momentum
Gas
Gridded Flux Products
Radiative
Freshwater
Heat and momentum
Gas
Desired Accuracies
Climate perspective
Ocean circulation perspective
Atmospheric dynamics perspective
Sea ice mass balance perspective
Needs
Radiative
Freshwater
Heat and momentum
Gas
IPY
Summary

1. Introduction
Countless publications, various regional and global assessments, and the popular media document rapid changes in snow- and ice-covered regions of our planet. Highly visible events such as extreme year-to-year sea ice loss in the Arctic Ocean or rapid collapse of large sections of Antarctic ice shelves apparently catch all of us, scientists and the public, by surprise. As scientists we must quickly develop and improve predictive skills, partly to recover from these recent surprises but mostly because we suspect that changes in the polar regions will reverberate throughout the physical, ecological and social systems of the entire planet. Predicting the rate and trajectory of polar changes will require enhanced collaboration among meteorologists, oceanographers, ice physicists and climatologists, new combinations of in situ measurements and satellite remote sensing, and close interaction between observationalists and modelers. The surface fluxes at high latitudes, the exchanges of heat, momentum and material among ocean, atmosphere and ice, constitute the essential language of these enhanced collaborations and present the most serious technical challenges to improved predictions.
In this paper we introduce the unique challenges of determining surface fluxes between oean, ice, and atmosphere at high latitudes, which we define to include the Arctic, Antarctica, and the Southern Ocean. High latitudes differ markedly from temperate regions because of two conditions: very high winds and intense spatial variability along ice margins. We evaluate the current capabilities of direct measurements, remote sensing, and gridded numerical estimations to provide high latitude fluxes, and we outline from several perspectives the requirements for improved measurements and improved modeling. We focus on regions poised near the freezing temperature where small changes in fluxes can induce dramatic changes in air-sea interaction or radiative balances. We touch briefly on radiative fluxes over the high-latitude ice sheets. Here we do not consider energy or material fluxes over land surfaces or freshwater fluxes from land to ocean; the SWIPA assessment (Snow, Water, Ice, Permafrost of the Arctic) will shortly provide an up-to-date description of surface and lateral fluxes and net mass changes of the Greenland ice sheet and various international carbon plans have described the measurement needs and challenges for terrestrial permafrost and tundra regions.
2. Surface Fluxes in High Latitudes
Radiative, freshwater, momentum, heat and gas fluxes all pose challenges in high latitude regions. Radiative fluxes include the shortwave electromagnetic radiation from the sun impinging on the ocean (or ice) surface and the longwave electromagnetic radiation emitted from the surface and from within the atmosphere. Freshwater fluxes occur as the balance between precipitation and evaporation and, unique to polar regions, during processes of ice formation and ice dissolution. The remaining exchanges, including momentum, sensible heat, latent heat and gas exchange, occur through combinations of diffusive and turbulent fluxes. Substantial efforts are required to quantify these fluxes in any natural environment, terrestrial or marine, tropical, temperate or polar. For a recent summary see e.g. Smith et al. (2009).
Polar regions present particularly challenging logistical constraints: extreme cold, long periods of darkness, presence of ice (and icing, on exposed surfaces), extreme distances from support services. These constraints result in a relative paucity of standard surface or upper air

meteorological data and the almost complete absence of moored or free-drifting sensor systems in large areas of the polar oceans covered with regular seasonal or durable ice. Two additional challenges in polar regions are high winds and marginal sea-ice zones.
Winds over polar oceans are among the strongest in the world (e.g. Liu et al. 2008;
Renfrew, 2008). Extrapolation of wind-based flux parameterizations developed and calibrated for temperate conditions becomes highly uncertain in extreme winds, particularly in large regions of the Southern Ocean, North Atlantic storm tracks, and near high orography. Wind-wave interactions and their impacts on surface stresses become acutely important when strong winds accompany or oppose large swells in the polar oceans. Oceanographic instruments which might provide surface flux validation must withstand high winds and rough seas as well as cold temperatures and icing conditions.
In the relatively calm conditions of sea ice margins, intermixed areas of open water, organic slicks, new ice, existing bare ice, snow-covered ice, and ice ponds, in an exquisite variety of fine-scale textures and patterns, interacting with overlying regions of haze, low cloud, and clear sky, modulate globally-important processes of ice formation, brine formation, sunlight reflection, and gaseous deposition (e.g. of mercury). Subtle changes in heat or momentum fluxes cause, and respond to, rapid phase changes of freezing or melting. Even the ‘margins’ themselves lose definition, as large areas of summer sea ice in both hemispheres develop extensive leads, channels, ponds and holes. Determining instantaneous local fluxes or validating average regional fluxes over the footprint area of a remote sensing instrument becomes almost impossible.
In many other aspects, determining surface fluxes in polar regions presents the same difficulties as elsewhere: extrapolating from point measurements to regional or global coverage; achieving temporal resolution across daily, monthly and annual scales; understanding the errors and uncertainties in various derived or composite products. Because direct high-quality flux measurements generally derive from geographically- and temporally-limited field programs, climate researchers and ocean modelers often use gridded flux fields inferred by applying flux parameterizations to satellite or numerical weather prediction grids of basic physical variables, such as air-sea temperature difference and surface wind speed, available with nearly (exceptions in polar regions) global coverage. Comparison of fluxes as represented in several widely-used gridded flux products (Figure 2) shows clear inconsistencies on sub-synoptic time scales at high latitudes, understandable because of differing representations of storms, but also large differences on monthly time scales particularly in high latitude regions. These substantial differences in fluxes reflect differences in parameterization and biases of the input data (Smith et al. 2009) but also demonstrate fundamentally different geospatial distributions. Figure 2 thus represents the two challenges to our group: to understand what we can do at present for high latitude surface fluxes and to identify what we can and must do better in the future.

3. Current capabilities
Direct (Radiative, Freshwater, Heat and Momentum, Gas)
In the tropics, extensive ship observations along with flux moorings have provided information on the variability and physical properties of surface fluxes. At high latitudes, in situ observations are considerably more limited. a. Radiative Fluxes
Most radiative flux information for high latitudes comes from radiometer measurements around the Arctic and on Antarctica. There are almost no observations of radiative fluxes (SW or
LW) over high latitude water bodies. The high albedo of snow and ice together with a large loss of long-wave radiation through clear and dry atmospheres result in a net loss of radiation in most months of the polar year. Cloud cover typically reduces the radiative loss of energy (Pietroni et al. 2008); however, the characteristics of clouds and other aerosols in polar latitudes are relatively poorly known (Lubin and Vogelmann, 2006). Pietroni et al. (2008) concluded that differences in long wave radiation and net long wave flux occurrence distribution between two
Antarctic sites, one near the coast and one on the continent, were strongly related to the differences in cloud cover. b. Freshwater Fluxes
Precipitation is notoriously difficult to measure in high latitudes; snowfall measurement errors may reach 100% in winter. Serreze et al. (2005) estimate that at a coarse grid cell resolution of 175 km, obtaining an accurate assessment of the grid cell precipitation requires typically 3-5 stations within the cell, more in topographically complex areas. However, for the
Arctic terrestrial drainage, only 38% of 175 km grid wells contain even a single station. The situation is much worse over Antarctica and the Arctic Ocean.
Water and salt fractionate when ice forms from seawater, resulting in substantial fluxes of freshwater into the ice and of salt into the brine and underlying seawater. Wind- and currentinduced transport of sea ice, particularly out of the Arctic basin, represents a substantial meridional freshwater flux. Concurrently, the salt flux into the ocean represents a dominant factor in deep water formation and associated global heat and CO
2
fluxes occurring through meridional oceanic circulations. Net depletion, annually or seasonally, of sea ice, as in the
Arctic, and net erosion of land ice, as perhaps occurs beneath Antarctic ice shelves, represent substantial freshwater inputs to polar oceans, with probable adverse impact on deep water formation and circulation. The accurate assessments of sea ice mass balance (sidebar) and of salinity budgets for ice-covered seas thus represent important, if indirect, measurements of these crucial high latitude fluxes. c. Heat and momentum
Direct in situ flux measurement are used to calibrate and validate indirect methods so that fields of fluxes can be determined from variables, such as wind speed and sea-surface temperature, that are available on the required space/time scales. The indirect methods are known as bulk flux algorithms (e.g. Fairall et al, 2003); they form the basis for ocean-surface

boundary conditions in virtually all climate and NWP models and in retrievals of turbulent fluxes from satellite observations.
In bulk algorithms the turbulent fluxes are represented in terms of mean wind speed, air and sea surface temperature, and air humidity, with appropriate transfer coefficients. These same basic approaches, modified to account for the frozen surface, can be applied to turbulent flux parameterization over ice (see Brunke et al. 2009). The intercomparison of transfer coefficients from different times, location and surface conditions is difficult. In terms of wind speed dependence of momentum and moisture transfer coefficients, flux algorithms (fig. 6) tend to agree for wind speeds between 2-14 ms
-1
where (not coincidently) there is a lot of data.
However, computing means and averages tends to mask real physical variability of fluxes in space and time at a given mean wind speed. Clearly, a steady wind blowing straight into large swells will generate more surface stress than the same wind going with the swells (observations suggest the difference is about a factor of two).
While scalar transfer in the atmosphere is dominated by turbulent flux, molecular diffusion must contribute heavily to the transfer at mm-scales near the interface, particular in ice margin zones. Because the molecular diffusion coefficients for water vapor and heat are 20% different, we expect their bulk transfer coefficients to be slightly different (order of 5%) but not sufficiently different to be unambiguously detectable by today’s observations. At the other extreme, evaporation of sea spray at high wind speeds will effectively enhance a moisture flux and reduce a sensible heat flux. Progress on enhancing bulk algorithms to characterize spray contributions continues (e.g., Andreas et al. 2008) but direct measurements of these effects remain difficult. d. Mass fluxes
Mass fluxes are often reported in terms of piston velocities, dependent on diffusivities and viscosities, forced by bulk air-sea effective concentration differences. Frew et al. (2004) proposed a relationship of gas piston velocities to the mean square slope of cm-scale waves, an approach that allows estimations of exchange velocities remotely using satellite altimetry (Frew et al 2007) or scatterometry (Bogucki et al 2009). Given the paucity of in situ observations of gas transfer (or the necessary bulk variables), remote sensing is likely to remain the primary source of global data. One difficulty is that surface and near surface CO
2
concentrations are not yet measured through remote sensing. Only two campaigns have measured gas fluxes at high latitudes: the Southern Ocean Gas Exchange experiment (2008), and the Circumpolar Flaw Lead experiment (2007-2008), both part of IPY. Other efforts, notably those of Miller et al (1999,
2002) and Skjelvan et al (1999), looked at the annual carbon budget in the Greenland and
Norwegian Seas and Baffin Bay. While these studies assumed ice to be a barrier to air-sea transfer, subsequent work has indicated that brine channels in sea ice may act as a pathway for carbon (Semiletov et al 2004). This is currently an active area of research (e.g. Delille et al
2007). e. Measurement programs
Ideally, we desire data from intregrated programs that measure all components of the surface energy budget plus gas exchanges across a representative range of conditions. Flux observations were successfully collected from the Surface Heat Budget of the Arctic Ocean
(SHEBA) ice camp in 1997 to 1998, but necessarily over perennial sea ice without leads (Perrson

et al. 2002). By the time of IPY 2007-2009, ice conditions in the Arctic became much less predictable or reliable for human on-ice deployments.
Several IPY projects measured components of surface fluxes in the Arctic and the
Antarctic, most often from ships. The sailing vessel Tara measured bulk meteorological and oceanographic parameters during its rapid drift across the Arctic (Developing Arctic Modeling and Observing Capabilities for Long-term Environmental Studies, DAMOCLES) while the
Arctic Sea Ice Properties and Processes and Antarctic Sea Ice Projects measured many of the flux terms needed to determine sea ice mass . The Ocean-Atmosphere-Sea Ice-Snowpack
Interactions (OASIS) project measured gas exchange processes over open water and ice in the
Arctic while the Southern Ocean Gas Exchange Experiment measured many of the processes that determine ocean-atmosphere CO
2
fluxes in high winds at southern high latitudes. Many other research vessels will have measured useful bulk meteorological and oceanographic parameters during IPY cruises while autonomous gliders and acoustically-tracked drifters demonstrated promising new capabilities for under-ice measurements. Because none of these IPY Projects had the measurement of surface fluxes as their primary mission, and there is not program to archeive the collective IPY surface observations, we identify the need for a post-IPY effort to identify and compile all the relevant bulk and flux measurements that occurred during IPY.
Satellite (Radiative, Freshwater, Heat and Momentum, Gas) a. Radiative
At present, large-scale estimates of radiative fluxes from satellite observations disagree most in Polar Regions (Fig. 5). None of the current satellite inference schemes accounts for the variability in the extent of sea ice and as such, do not correctly represent the boundary conditions in the radiative transfer computations. Several authors have focused on verification of parameterizations of DLW radiation data year-round data (König-Langlo and Augstein, 1994), polar summer data (Key et al., 1996) or polar winter/late-autumn data (Guest, 1998, Makshtas et al., 1999). The parameterization of König-Langlo and Augstein (1994) reproduced the observations with root mean square (RMS) deviations of less than 16 W/m
2
. Liu et al. (2005) indicate that the surface downward shortwave radiative fluxes derived from satellites are more accurate than the two main reanalysis datasets (NCEP and ECMWF), due to the better information on cloud properties in the satellite products. The SHEBA project showed that satellite-based analysis may provide downward shortwave (long wave) radiative fluxes to within
~ 10-40 (~10-30) W m
-2
of surface observations (Perovich et al., 1999). b. Freshwater
Alternative sources of high latitude precipitation include satellite retrievals, output from atmospheric reanalyses, and various blends of satellite, reanalysis and station data (Huffman et al., 1997; Serreze et al., 2005; Xie and Arkin, 1997). Satellite retrievals are uncertain over cloudy, snow and ice-cover surface and it appears that better estimates can be obtained from atmospheric reanalyses (Serreze et al., 2005). However precipitation biases in reanalysis field can be very large (Serreze aand Hurst, 2000)
.c. heat and momentum
Satellite-derived turbulent sensible and latent heat flux products are based primarily on application of the bulk aerodynamic flux formulations using as input satellite observations of

wind speed, sea surface temperature (SST), and near-surface air temperature and specific humidity. The high latitude challenges to the satellite-based turbulent heat flux products are largely due to limited numbers of direct observations with which to derive and validate the methods and products in these regions. Atmospheric temperature and humidity retrievals are particularly poor for cold conditions. Atmospheric stability and air-sea temperature differences can be very different near ice edges and cold land masses; few data exist to confirm the accuracy of satellite retrievals over these relatively small scales and under these conditions. SST retrievals are also subject to increased uncertainties in the high latitudes due to difficulties in discriminating open ocean, clouds, and ice in infrared brightness temperatures, and proximity to ice in microwave retrievals.
Momentum fluxes (stress) are typically derived from satellite observations of surface wind.
The sampling from a single wide swath scatterometer (e.g., QuikSCAT) is approximately sufficient to determine monthly average stresses with an accuracy of better than 0.01Nm
-2
. The differences between monthly scatterometer observations and monthly merchant marine
(Bourassa et al. 2005) observations are remarkably small for regions with good ship traffic.
Although scatterometer surface coverage is much better at high latitudes, calibration of scatterometers for very high wind speeds remains a serious problem due to the paucity of good comparison data for wind speeds greater than roughly 20 ms
-1
. Such wind speeds are often associated with rain, which modifies retrievals (Draper and Long 2004; Weissman et al. 2002;
Weissman and Bourassa 2008). The rapid translational motion and evolution of high latitude weather systems results in poor temporal sampling, even from a wide swath polar orbiting satellite (e.g., QuikSCAT). d. Gas
Wind inputs can easily be observed from space; however, near surface partial pressures of
CO2 are still in situ observations.
Gridded Flux Products (Radiation, Freshwater, Heat and momentum, Gas) a. Radiation
Radiative flux products inferred from numerical model outputs show substantial discrepancies in polar regions (Sorteberg et al. 2007). The comparison of the surface energy budget over the Arctic (70-90°N) from 20 coupled models for the IPCC fourth Assessment with
5 observationally based estimates and reanalysis showed a large bias in the climate models with the largest differences located over marginal ice zones. As shown in Figure 1 (Wild et al. 2005), there is a large discrepancy between model estimates of DLW and ground observations over the poles; GCMs tend to underestimate DLW. Significant underestimates are found at observation sites in cold and dry climates with low DLW emission, which implies an excessively strong meridional gradient of DLW in the GCMs. Iacono et al. (2000) found a substantial increase in the DLR at high latitudes in GCMs that used improved formulations of the water vapor continuum. Estimates of surface energy fluxes over the Arctic Ocean from atmospheric reanalyses, which might constrain the modeled fluxes, are unfortunately also of questionable accuracy in terms of individual components and the net surface flux. The annual net surface flux

averaged over the Arctic Ocean from ERA-40 is 11 W m
-2
(Serreze et al. 2007), compared to 6 W m
-2
from JRA-25, representing an effective annual ice thickness difference of 0.5 m.
Freshwater
Covered above
Heat and momentum
In the Southern Ocean State Estimate (SOSE), an ocean data assimilation model, air-sea heat and freshwater flux estimates (Mazloff et al., 2009) are produced consistent with the techniques of Stammer et al. (2004) for a global ocean state estimate. Any misfit between ocean observations and the model results is reduced through an iterative process of adjusting the control variables (prescribed atmospheric state and the initial condition), but not the dynamical parameters of the model. SOSE Southern Ocean (south of 25

S) air-sea heat and freshwater flux fields are provided at relatively high horizontal resolution (1/6

) and relatively short temporal resolution (five day averages). Cerovecki et al. (2009) show that the adjustments made by SOSE to the control variables tend to reduce the known biases in heat and freshwater flux estimates
(e.g. Taylor et al., 2000). SOSE flux estimates moreover have the advantage over alternative flux estimates that they are mutually consistent with one another and with the oceanic variables in that they fit into closed model budgets of heat, freshwater, momentum, and particularly salinity, which is especially important for an accurate estimate of the SO surface buoyancy fluxes where very cold temperatures in the southern part of the southern ocean can make the fresh water contribution to the total buoyancy flux more important then the heat flux contribution (Warren et al., 1996). The downside of this approach is that model errors in non-flux processes also result in adjustments to surface fluxes.
For many polar regions, event-driven fluxes dominate annual or even seasonal averages; storm-driven heat and moisture fluxes probably exceed all other terms in an Arctic energy budget. Polar events include intense mesoscale cyclones (polar lows) and cold air outbreaks from land to ocean. The generation of comprehensive air-sea flux products for these conditions confronts the usual polar constraints: a scarcity of in situ observations for validation. Other difficulties include uncertainties in bulk flux parameterizations at high winds (already mentioned), smaller Rossby numbers and a reduction in scale of circulations leading to resolution problems in data assimilation (e.g., Chelton et al. 2006) and in simulation, and, in the case of cold air over warm ocean, very high sensible heat fluxes (Shapiro et al. 1987). For example, in Fig. 7 over the sea-ice (0-30 km) the SST and wind speed are too high in the NCEP analyses, leading to an overestimate in the heat fluxes; while just off the ice-edge (40-120 km) the 2-m temperature and the 10-m wind speed are both too low in the ECMWF analyses, leading to an underestimate in the heat fluxes there.
Gas
Something short mentioning Wannikof’s averages and the limitations.

4. Desired Accuracies, Now and Future a. From a climate perspective
How accurate do we want surface fluxes to be? If our goal were to diagnose long-term climate change by observing shifts in surface fluxes, then the requirements would be stringent.
Observed long-term warming trends in the ocean from 1993-2003 can be explained by an ocean heat gain of just 0.86 ± 0.12 W m -2
(Hansen et al., 2005). For sea ice, a 1 W m
-2
flux imbalance can melt 10 cm of ice in a year. However, the climate change signal of O(1 W m
-2
) is, at present an unachievable standard for directly measured surface fluxes, and long-term changes in fluxes are more effectively diagnosed through other means. While gridded flux products with 1 W m -2 accuracy may be unachievable, significant scientific gains could be made if we could improve the accuracy of heat flux estimates by an order of magnitude.
Basin-scale changes in salinity associated with global change are associated with very small changes in air-sea freshwater flux, on the order of 0.05 to 0.1 psu/decade (Boyer et al., 2005) concentrated in the top 200 m. This trend is equivalent to a change in precipitation-evaporation of 0.003 cm/yr, far below any expected accuracy globally or in polar regions. (xx better check this number before publishing xx)
Thus, for heat and freshwater fluxes associated with climate change, observing the ocean temperature and salinity changes as integrators of heat and freshwater fluxes remains a better approach than attempting to determine, globally or in polar regions, actual fluxes. b. From an ocean circulation perspective
The ocean mixed layer mediates fluxes between the deeper ocean and the atmosphere. Airsea buoyancy (heat and freshwater) fluxes change water properties, the waters then circulate at the surface and at depth to accomplish planetary transports of heat and freshwater. At high latitudes, surface cooling produces deeper mixing. Salinity becomes a major or even dominant factor in extreme polar regions where temperatures approach the freezing point. Thus highly accurate surface heat and freshwater fluxes, including freshwater fluxes linked to processes of ice formation, export and melt, are critical to high latitude ocean processes. For example, buoyancy gain by excess precipitation and buoyancy loss by ocean heat loss are of comparable importance in estimating Sub-Antarctic mode water formation (Cerovecki et al., 2009).
Calculation of water mass formation rates requires accurate and balanced fluxes (e.g without a net warming or cooling bias). Using the best available data products, Dong et al (2007) found that the zonally averaged imbalance can be 50 W m
-2
, and locally, the upper ocean heat balance can have an root mean squared misfit of more than 200 W m
-2
at any given location , and 130 W m -2 in a global rms averaged sense. Such large errors make it difficult to discern the details of the upper ocean heat storage and meridional overturning circulation. If root mean squared errors could be reduced to 10 W m
-2
for a xx time scale, the situation would clearly improve. Achieving such an accuracy would require much better sampling, and a reduction in biases, particularly for high wind conditions.
Freshwater flux anomalies documented in the North Atlantic, sufficient to inhibit or slow deep convection, are on the order of 0.05 Sv (Curry and Mauritzen 2005). Although these freshwater anomalies derive from river runoff and ice melt, we calculate an equivalent precipitation/evaporation since variations in P-E can also impact the surface salinity. A variation

of 0.05 Sv over the area the Arctic and North Atlantic converts to almost 1 cm/yr. The (global?)
P-E error reported by Taylor et al. (2000) appears to be on the order of 0.3 cm/yr, but the adjustments required in the SOSE ocean assimilation model are on the order of 1 cm/yr, equivalent to the desired signal. Here, improvement in accuracy of NWP products is clearly required. c. From an atmospheric dynamics perspective
A poleward trend in storm-track location has been detected in the southern hemispheric summer (reference). Hakkinen et al. (2008) show a positive trend in storm activity in the high
Arctic both in summer and winter over the past 50 years. In both cases increased turbulent surface fluxes can feed back on large scale atmospheric flows, affecting heat and moisture transports, ocean mixing, and sea ice movement. Turbulent energy fluxes resulting from the opening up of previously ice covered areas of the Arctic are especially large (area average of
O(50-70 W m
-2
) in winter. Recommendation – 10 W m
-2
, 5-10 km, hourly? ; which will again require much more frequent sampling from satellites, and increased accuracy in mean values and reduced random errors. d. From a sea ice mass balance perspective
We identify improved assessment of the Arctic sea ice mass budget as a key need and urgent challenge. The range in recent and projected future ice extent and volume trends from different models remains large, reflected in both initial (20th century) and evolving surface energy fluxes.
Inter-model scatter in absorbed solar radiation, due in part to differences in the surface albedo simulation, is a particular concern (Holland et al., 2009). When Bitz et al (2006) adjusted sea ice albedo by about 8%, resulting in a change in the net surface shortwave flux by about 5-10 W m
-2 on the annual mean, the resultant equilibrium (e.g. with a net flux into the sea ice of zero) sea ice thickness was thinner by 1-2 m in the central Arctic and about 1/4 m in the Antarctic. Perovich et al. (2009) show that substantial sea ice mass loss occurs from below, through fluxes between ice and ocean; Rintoul et al (comm.) suggest a similar process underway beneath Antarctic ice shelves. In both hemispheres, reduced ice thicknesses will feed back on ocean-atmosphere processes by changing the conductive as well as the sensible and upward surface longwave fluxes through the ice. For the under-ice situation, measurements of mass loss (thickness change) and of salinity will necessarily substitute for direct ice-ocean flux measurements. Above the ice, where a net surface flux of 1 W m
-2
over a year is the energetic equivalent of melting 0.1 m of ice, predictions of ice mass balance require net flux precisions and accuracies of that order
(1 W m -2 ).
5. Key needs
Radiation
Despite the challenges, the accuracy of the radiative fluxes in polar regions are likely to improve with the utilization of newly available satellite observations, improved inference schemes, and improved in situ observations, particularly of clouds and LW, as ground truth. In particular, more accurate data on surface conditions such as ice extent, atmospheric conditions such as aerosol optical properties, improved models of narrow to broadband transformations with realistic surface models and newly available bi-directional distribution functions (BRDF) models
(e.g., from CERES or MISER) need to be utilized. Daily average values from the Moderate

Resolution Imaging Spectro-radiometer (MODIS) instrument onboard the Terra and Aqua satellites (King et al., 1992) have allowed development of a new inference scheme to estimate shortwave radiative fluxes (Wang and Pinker, 2009) that agree well with ground measurements
(Fig. 6) over oceanic sites (agreement is better over land). The improvement is very significant in problematic areas for most inference schemes such as the Tibet Plateau and Antarctica.
Freshwater
More reference quality surface measurements.
Heat and momentum
More high quality in situ comparison data for at wind speeds greater than 14 m s
-1
, with and without rain contamination, are needed to greatly improve the calibrations. Wave effects on surface fluxes remain the single most important and confounding problem. Analyses will require more frequent satellite observations. For high latitudes, at least two wide swath scatterometers are desired for sampling typical mid-latitude storms and polar lows.
The formation and presence of ice provokes a step-function change in radiative, heat, momentum and gas fluxes. Ice formation and accumulation processes, which can include snow refreezing (common in the Antarctic) and vertical migration of frazil ice, and dissolution, erosion, and break-up processes remain highly complicated, closely linked to chemical and microbiological processes, and beneath the footprint of most remote sensing and below the resolution of most parameterizations. As more-stable multi-year ice declines, annual ice processes and extent will become increasingly important terms in air-sea interaction and high latitude fluxes.
Gas
IPY
We have listed several IPY projects in this discussion, those with measurements of surface fluxes as a prominent feature. We believe that many other IPY projects in both hemispheres will have collected useful in situ data bulk or in some cases flux data as routine or necessary components of a larger measurement program. Therefore, we identify the need for a post-IPY effort to identify and compile all the relevant bulk and flux measurements that occurred during
IPY.
6. Summary
1.
We have identified substantial deficiencies in direct measurements, remote sensing, and derived flux products at high latitudes. The current array of gridded flux products do not agree. Virtually every aspect of improvement, in measurements, paramerizations, extrapolations and modelling, requires more and better in situ observations, particularly in the difficult high wind and ice margin conditions unique to polar regions. In most cases, the improvements themselves seem likely to fall far short of the accuracies required for skilful predictions. Improvements in high latitude regions present extreme

technical, analytical and conceptual challenges. The combination of what we need most where we face the greatest difficulties pushes us toward new idea and technology.
As fuel costs rise, the number of research or logistics ships operating at high latitudes will decrease, certainly from the peak IPY years and possibly (although one hopes not) from pre-IPY levels. We therefore need to make better use of fewer opportunities. Tourist ship trips will increase, albeit to only a small portion of the Arctic and a minute portion of the
Antarctic; the tourist ships do cross some interesting straits, however. Autonomous gliders provide very interesting possibilities; one glider operating off of Norway profiled the upper ocean and penetrated the air-sea interface nearly 200 times in a three-month period during IPY. The measurement of under-ice salinity as an indication of ice-toocean freshwater flux likewise requires tethered or autonomous sensor systems.
Reference sensors on research vessels, smart sensors on volunteer ships, autonomous gliders, all coupled to advance array communication and data assimilation systems, represent an important step forward; progress in meeting the high latitude needs will enable better global measurements as well.

References
Bitz et al 2006.
Bogucki et al. 2009.
Boyer, T.P., J.I. Antonov, S. Levitus, and R. Locarnini, 2005: Linear trends of salinity for the world ocean, 1955-1998. Geophys. Res. Lett., 32(L01604), doi:1029/2004GL021791.
Brunke, M. A., M. Zhou, X. Zeng, and E.L. Andreas, 2009: An intercomparison of bulk aerodynamic algorithms used over sea ice with data from the SHEBA experiment J.
Geophys. Res.
, to appear.
Cerovecki, I., L. Talley and M. Mazloff, 2009. Subantarctic Mode Water and Antarctic
Intermediate Water Formation using a Walin framework. In preparation.
Chelton, D.B., M.H. Freilich, J.M. Sienkiewicz, and J.M. Von Ahn, 2006: On the Use of
QuikSCAT Scatterometer Measurements of Surface Winds for Marine Weather Prediction.
Mon. Wea. Rev.
, 134, 2055–2071.
Curry and Mauritzen, 2005.
Delille et al, 2007.
Dong et al., 2007.
Draper, D.W. and D.G. Long, 2004; Evaluating the effect of rain on SeaWinds scatterometer measurements, J. Geophys. Res ., 109 , doi:10.1029/2002JC001741
Fairall, C. W., E. F. Bradley, J. E. Hare, A. A. Grachev, and J. B. Edson, 2003: Bulk parameterization of air-sea fluxes: Updates and verification for the COARE algorithm. J.
Clim.
, 16 , 571-591.
Frew et al, 2004.
Guest, 1998.
Hakkinen, S., A. Proshutinsky and I. Ashik, 2008: Sea ice drift in the Arctic since the 1950s.
Geophys. Res. Lett., xx, xx.
Hansen, J., L. Nazarenko, R. Ruedy, M. Sato, J. Willis, A. Del Genio, D. Koch, A. Lacis, K. Lo,
S. Menon, T. Novakov, J. Perlwitz, G. Russell, G. A. Schmidt, N. Tausnev, 2005. Earth's
Energy Imbalance: Confirmation and Implications, Science, doi: 10.1126/science.1110252.
Holland. M.M., M.C. Serreze and J. Stroeve (2009), Projected changes in the sea ice mass budgets of the Arctic, Climate Dynamics (in press).
Huffman, G. J. et al. (1997), The Global Precipitation Climatology Project (GPCP) combined precipitation data set, Bull. Amer. Meteor. Soc., 78, 5-20.
Iacano et al, 2000.
Key et al., 1996.
King et al., 1992.
Konig-Langlo and Augstein, 1994.
Liu, W. T., W. Tang, and X. Xie, 2008: Wind power distribution over the ocean. Geophysical
Research Letters, 35, doi:10.1029/2008GL034172.
Lubin and Vogelmann, 2006
Makshtas et al., 1999.
Mazloff, M. R., P. Heimbach, and C. Wunsch, 2009: An eddy permitting Southern Ocean state estimate. Submitted to JPO.
Miller et al, 1999.
Miller et al. 2002.
Perovich, D. K., et al., 1999. Year on ice gives climate insights. Eos Trans. AGU, 80, 481.

Perovich, D.K., J.A. Richter-Menge, B. Elder, K. Claffey, and C. Polashenski, Observing and understanding climate change: Monitoring the mass balance, motion, and thickness of
Arctic sea ice, http://www.crrel.usace.army.mil/sid/IMB/, 2009
Persson, P. O. G., C. W. Fairall, E. Andreas, P. Guest, and D. K. Perovich, 2002. Measurements near the atmospheric surface flux group tower at SHEBA: Near-surface conditions and surface energy budget. J. Geophys. Res., 107(C10), 8045, doi: 10.1029/2000JC000705.
Pietroni et al. 2008
Renfrew, 2008???
Semiletov et al. 2004.
Serreze, M.C. and C.M. Hurst (2000), Representation of mean Arctic precipitation fron NCEP-
NCAR and ERA Reanalyses, J. Climate, 13, 182-201.
Serreze, M.C., A.P. Barrett, and F. Lo (2005) Northern high-latitude precipitation as depicted in atmospheric reanalyses and satellite retrievals, Mon. Wea. Rev., 133, 3407-3430,
Serreze, M.C., A.P. Barrett, A.J. Slater, M. Steele, J. Zhang, and K.E. Trenberth (2007a), The large-scale energy budget of the Arctic, J. Geophys. Res., 112, D11122, doi:10.1029/2006JD008230.
Shapiro, M. A., L. S. Fedor and T. Hampel (1987), Research aircraft measurements of a polar low over the Norwegian Sea, Tellus 37A, 272-306.
Skjelvan et al. 1999.
Smith et al., 2009.
Sortebeg et al, 2007.
Stammer D, Ueyoshi K, Large W, Josey S, Wunsch C., 2004: Estimating air-sea fluxes of heat, freshwater and momentum through global ocean data assimilation. J. Geophys. Res., 109, doi:10.1029/2003JC002082.
Taylor, P., (Ed.), 2000. Final report of the joint WCRP/SCOR Working Group on air-sea fluxes: intercomparison and validation of ocean-atmosphere flux fields, WCRP-112, WMO/TD-No.
1036, World Climate Research Programme, 303 pp.
Wang, H., and R. T. Pinker, 2009. Radiative Fluxes from MODIS. JGR-Atmospheres, in revision.
Warren, B. A., J. H. LaCasce, and P. E. Robbins, 1996: On the obscurantist physics of "form drag" in theorizing about the Circumpolar Current. J. Phys. Oceanogr., 26, 2297-2301.
Weissman, DE; Bourassa, MA; Tongue, J, 2002: Effects of rain rate and wind magnitude on
SeaWinds scatterometer wind speed errors. JOURNAL OF ATMOSPHERIC AND OCEANIC
TECHNOLOGY , 19 , 738-746, doi:10.1175/1520-0426(2002)019<0738:EORRAW>
2.0.CO;2.
Weissman, D. E., and M. A. Bourassa, 2008: Measurements of the Effect of Rain-induced Sea
Surface Roughness on the Satellite Scatterometer Radar Cross Section. IEEE Trans. Geosci.
Remote Sens.
, 46 , 2882-2894
Wild et al, 2005.
Xie, P. and P.A. Arkin (1997), Global precipitation: A 17-year monthly analysis based on gauge observations, satellite estimates and numerical model output, Bull. Amer. Meteor. Soc., 78,
2539-2558.