A Macroscale Hydrological Model for the Arctic Basin

Despite the importance of Arctic freshwater fluxes to the thermohaline circulation of the world ocean and the global heat balance, the hydrology of the Arctic basin land surface is not well understood. The Arctic basin, defined as the land area draining to the Arctic Ocean (17,400,000 km²), spans 37 degrees latitude from 46N to 83N (Figure 1). Vegetation consists primarily of boreal forest in the southern lowlands, transitioning to taiga and tundra with increasing latitude and/or elevation.

The objective of this research is to estimate freshwater fluxes into the Arctic Ocean, and the hydrologic controls of the interannual and seasonal variability of such fluxes, through macroscale hydrologic modeling of the entire Arctic Basin. The current understanding of the freshwater balance of the Arctic basin from a land surface hydrological perspective is reviewed. The VIC hydrologic model is applied to two of the Arctic river basins, the Ob and the Mackenzie, prior to extension to the entire Arctic Basin.

Hydrologic modeling of the Arctic Basin requires special consideration of several key land surface factors:

• 40% of Arctic drainage lies south of 60N; vegetation is typically boreal forest, therefore forest effects represent substantial influence (Figure 2).
• Sublimation and wind redistribution of snow are important components of the snow available for runoff in tundra areas (e.g. Pomeroy et al. 1997).
• Snow is present for 3-4 months, even at southernmost latitudes; non-frozen precipitation still significant even at northernmost latitudes (Figure 3).
• Seasonal snow, lake and surface storage exert strong controls on interseasonal and interannual variability leading to differing responses for Eurasia and N. America (Figure 4).
• Precipitation and evaporation both decrease with latitude; runoff ratios are fairly constant (0.25 - 0.60).

The VIC model (Liang et al. 1994) is a grid based SVAT scheme developed as a macroscale hydrological model and for inclusion in GCMs.

• The soil column is partitioned into a variable number of soil layers. Infiltration into the top soil layer is controlled through a variable infiltration curve. Release of baseflow from the pixel is controlled through a baseflow curve (Figure 5).
• Runoff and baseflow for each cell is routed to the cell outlet using a unit hydrograph (Lohmann et al. 1998a; 1998b). The hydrographs for the individual cells are routed to the basin outlet through a channel network using the linearized Saint-Venant equations (Figure 6).
• Snow accumulation and melt are solved via an energy balance. Interception of snow in the canopy, sublimation from the ground snow pack, and ground heat flux are included.
• The effect of sub-grid topography on snow precipitation and melt is represented through snow elevation bands (Figure 7).

At a minimum, the VIC model requires time-series of temperature and precipitation and spatial maps of vegetation and soil characteristics. Data sources for this application include:

• 2^o global daily precipitation data set:
• Monthly data from Hulme (1995), disaggregated to daily by Schnur and Lettenmaier (1997).
• Corrected for gauge catch deficiencies after Larson and Peck (1974).
• 2^o global daily Tmin/Tmax data set:
• Gridded mean monthly temperature climatology from Jones (1994), interpolated to 2^o by Schnur and Lettenmaier (1997).
• Incoming shortwave and longwave radiation and humidity parameterized from air temperature.
• 10 m wind speed data set:
• NCEP/NCAR reanalysis 6 hour forecast fields interpolated from gaussian grid.

Our general approach is to estimate freshwater fluxes into the Arctic Ocean by representing hydrologic processes at large spatial scales (e.g. degrees) utilizing a physical representation to the extent possible. This involves:

• Initial calibration of the Ob and Mackenzie, and the Irtish and Athabasca subbasins, evaluated using remotely sensed snow cover extent and snow depth (from the EOSDIS NSIDC Distributed Active Archive Center), in addition to observed streamflow hydrographs. (Figure 8, Figure 9, Figure 10 and Figure 11).
• Extension to other continental-scale basins (Yenesei and Lena) to determine the degree of transferability of the calibrated model (Figure 12).
• Application to the entire Arctic basin (Figure 13 and (Figure 14).

The initial results (shown here) for the Mackenzie and Ob Basins indicate the following:

• While the major features of seasonal streamflow variation are represented, the runoff peak occurs 2-3 weeks early at the mouth of both the Mackenzie and Ob Rivers (Figure 8 and Figure 9). However, there is no consistent bias in the timing of snowmelt (Figure 11).
• The timing of the snowmelt peak is better for two southern sub-basins of the Mackenzie and Ob Rivers (the Athabasca and Irtish, respectively), however the magnitude is overestimated (Figure 8 and Figure 9).
• Comparisons of the simulated and remotely sensed snow cover extent indicate that the model represents annual snow cover extent well for both large river basins (Figure 10) and the entire Arctic (Figure 14).

Hulme, M. (1995). Estimating global changes in precipitation. Weather, 50, 34-42.

Jones, P.D. (1994). Hemispheric surface air temperature variations: a reanalysis and update to 1993. Journal of Climate, 7, 1794-1802.

Larson, L.W. and E.L. Peck (1974). Accuracy of precipitation measurements for hydrologic modeling. Water Resources Research, 10 (4).

Lettenmaier. D.P., L.C. Bowling and B.V. Matheussen (1997). Hydroclimatology of the Arctic Drainage Basin, NATO ARW, Arctic Ocean Fresh Water Budget, in review.

Liang, X.D., D.P. Lettenmaier, E.F. Wood and S.J. Burges, (1994). A simple hydrologically based model of land surface water and energy fluxes for GCMs. Journal of Geophysical Research, 99(D7), 14415-14428.

Lohmann, D., E.Raschke, B.Nijssen and D.P. Lettenmaier (1988a). Regional scale hydrology I: formulation of the VIC-2L model coupled to a routing model. Hydrological Sciences Journal, 43(1), 131-141.

Lohmann, D., E.Raschke, B.Nijssen and D.P. Lettenmaier (1988a). Regional scale hydrology II: application of the VIC-2L model to the Weser River, Germany. Hydrological Sciences Journal, 43(1), 143-157.

Pomeroy, J.W., P. Marsh, and D.M. Gray (1997). Application of a distributed blowing snow model to the Arctic. Hydrological Processes, 11 (11), 1451-1464.