1. Introduction
Given the pressing demand for local climate change information by resource management and impact assessments, both global and regional climate models have incorporated increasingly sophisticated physics representations run at higher resolutions. As a key coupled component, land surface models (LSMs) have also evolved from simple bucket to comprehensive dynamic hydrology–ecosystem representations (Gochis et al. 2004). Most current LSMs, however, still contain various defective parameterizations for terrestrial hydrologic processes, such that continuous modifications are inevitable (Stieglitz et al. 1997; Chen and Kumar 2001; Warrach et al. 2002; Niu and Yang 2003; Niu et al. 2005; Niu and Yang 2006; Niu et al. 2007; Oleson et al. 2008). Although crucial to surface water and energy prediction, representations of terrestrial hydrologic processes remain relatively simplistic. The crude and often unrealistic assumptions underpinning these parameterizations introduce substantial deficiencies into the LSMs and eventually limit the predictive skill of their coupled climate models. These deficiencies manifest as nonlinear drifts in the dynamic responses of land surface processes to external forcings (such as climate anomalies or future changes), which in turn feed back to the coupled modeling system and ultimately lead to significant errors in surface water and energy prediction.
Runoff is one of the most important components in the terrestrial hydrologic cycle. LSMs partition incoming precipitation into evapotranspiration, surface runoff, and subsurface runoff and allocate the residual to change soil moisture. Although the representation of the hydrologic cycle becomes more detailed (Stieglitz et al. 1997; Chen and Kumar 2001; Warrach et al. 2002; Niu and Yang 2003; Niu et al. 2005; Niu and Yang 2006; Niu et al. 2007; Oleson et al. 2008), the runoff prediction by most current LSMs is still problematic owing to the use of various crude parameterizations. This is no exception for the Common Land Model (CLM), which is chosen in this study as the basic LSM to develop improved runoff prediction.
The CLM is a state-of-the-art soil–vegetation–atmosphere transfer model (Dai et al. 2003). It has been incorporated into the mesoscale Climate–Weather Research and Forecasting model (CWRF; Liang et al. 2005a,b,c,d, 2006; Choi 2006; Choi et al. 2007) with numerous crucial updates for land processes. The original CLM has been extensively evaluated against field measurements in a stand-alone mode as driven by the observational forcing (Dai et al. 2003; Niu and Yang 2003; Niu et al. 2005; Qian et al. 2006; Niu and Yang 2006; Niu et al. 2007; Lawrence and Chase 2007; Lawrence et al. 2007; Oleson et al. 2008). All have indicated that the CLM realistically simulates the land state variables (soil moisture, soil temperature, and snow water equivalent) and fluxes (net radiation, latent and sensible heat, and runoff). On the other hand, our own experience has found that a direct application of the CLM at the CWRF 30-km grid resolution leads to serious problems in predicting the hydrologic cycle, especially runoff. Solutions to some of these problems are the focus of this study.
Note that in the past literature, most hydrologic parameterization schemes in LSMs, including CLM, have been tested with field measurements at small-scale catchments (Stieglitz et al. 1997; Warrach et al. 2002; Dai et al. 2003; Niu et al. 2005; Niu and Yang 2006), while directly applied at much larger climate model grids, such as the Community Climate System Model (CCSM), European Centre for Medium-Range Weather Forecasts (ECMWF), Goddard Earth Observing System, version 5 (GEOS-5), among others. Given the strong scale dependence, these parameterization schemes, especially for runoff, must be tuned and evaluated at the same resolution as their host climate models. This scale inconsistency may explain why the CLM application at the CWRF 30-km grid produces substantial errors in runoff.
Therefore, the present study develops and evaluates various improvements to the terrestrial hydrologic schemes in the CLM for their designated application at the CWRF 30-km grid resolution, focusing on the representation of surface and subsurface runoff. Following Choi (2006) and Choi et al. (2007), all schemes are implemented at the grid points rather than basins or catchments to avoid downscaling and upscaling exchanges between CWRF atmospheric forcings and CLM hydrologic predictions. The performance improvement of the new terrestrial hydrologic scheme in predicting runoff is evaluated using weekly observations at a catchment-scaled study domain around the Ohio Valley region. The improvements include realistic bedrock depth, dynamic water table, effective hydraulic conductivity, minimum residual soil water, maximum surface infiltration, and saturated lateral runoff limit on baseflow. The use of the realistic bedrock depth allows a better estimate of soil moisture memory that affects both soil water dynamics and runoff. The dynamic prediction of the water table produces more reasonable variations of the saturated zone depth than the equilibrium approximation. All of the other changes improve soil moisture availability and hence the terrestrial water balance. As demonstrated below, together the new schemes generate runoff variations much closer to observations.
2. Experiment design for model evaluation
Because of the strong scale dependence, the performance of the terrestrial hydrologic schemes developed below are compared to the existing ones in predicting runoff and are evaluated over a relatively large catchment using the actual CWRF–CLM 30-km grid cells targeted by climate applications. To facilitate result comparison with observations, the model experiments are conducted in a stand-alone mode, where the CLM is driven by the most realistic surface boundary conditions (SBCs) and meteorological forcings. This avoids the complication from errors of atmospheric processes and surface–atmospheric feedbacks in the fully coupled CWRF.
a. Study catchment
To appropriately evaluate the performance of the terrestrial hydrologic schemes in the CLM, a study catchment of medium size around the Ohio Valley within the CWRF U.S. domain is chosen for which the observed streamflow discharges are available from the U.S. Geological Survey (USGS) National Water Information System (available online at http://waterdata.usgs.gov/nwis/sw). Located near the drainage outlet of the study catchment, there is one USGS gauge station (03287500)—named Kentucky River—at Lock 4 near Frankfort, Kentucky, with excellent records of flow discharges (for convenience, this station is referred to as KRL4). The Kentucky River basin has a drainage area of 13 706 km2. It is modeled as an area of 13 500 km2 by 15 computational grid boxes with a 30-km horizontal spacing. Figure 1 shows the computational domain with a rectangular size of 240 km (eight grid cells) by 150 km (five grid cells) covering the entire Kentucky River basin. This is a subset of the actual computational domain used in the current CWRF for U.S. climate applications (Liang et al. 2004). The whole U.S. domain, centered at (37.5°N, 95.5°W), includes a total of 196 (west–east) by 139 (south–north) grids using the Lambert conformal conic map projection.
b. Surface boundary conditions
The CLM, as coupled with CWRF, incorporates the comprehensive surface boundary conditions based on the best observational data over North America constructed by Liang et al. (2005c). Having removed numerous limitations and inconsistencies, the SBCs can be readily implemented into any regional climate model suitable for mesoscale regional climate and terrestrial hydrology modeling. For this study, the required SBCs include surface topography, bedrock depth, soil sand and clay fraction profiles, surface albedo localization factor, surface characteristic identification, land cover category, fractional vegetation cover, and leaf and stem area index. Figures 1a and 1b depict the distributions at the CWRF 30-km grid over the Kentucky River basin for terrain elevation ranging from 225 to 687 m and land cover types consisting of cropland/woodland mosaic, deciduous broadleaf forest, and mixed forest. See Liang et al. (2005c) for the details of data sources and construction methods.
c. Meteorological forcings and initial conditions
The stand-alone CLM simulations with the old or new terrestrial hydrologic parameterizations are driven by the same meteorological forcings constructed from the best available observational North American Regional Reanalysis (NARR) (Mesinger et al. 2006). The NARR adopts a 32-km grid, close to that of CWRF, and provides 3-hourly atmospheric and land data over an extensive area that completely includes our U.S. computational domain. The outcome represents a major improvement upon the earlier global reanalysis datasets in both resolution and accuracy. The required atmospheric variables are listed in Table 1. They are remapped onto the CWRF grids by linear spatial interpolation. Since the NARR data for soil temperature and moisture are given only in four layers of 0–10, 10–40, 40–100, 100–200 cm below the surface, a CLM spinup strategy is adopted. Specifically, the model is started at 0000 UTC 1 January 1995 and run continuously throughout the whole year of 1995 as driven by the NARR data. This is repeated for two cycles with the same 3-hourly NARR forcings of 1995. The resulting conditions at the end of the second cycle are considered to be fully consistent with the atmospheric forcings and hence used as the initial conditions for the subsequent CLM simulation to be evaluated against observations.
3. New parameterizations and result improvements
The new parameterizations for the CLM terrestrial hydrologic processes developed here can be separated into five major groups. The description of the new schemes and the evaluation of their effects on runoff against observations are presented below. This is done in a progressive manner in which each subsequent improvement includes all previous changes.
a. Bedrock and drainage
The bedrock depth is defined as the depth of soil and/or unconsolidated material that lies between the land surface and the geologic substratum. It acts as the bottom lid that effectively prevents downward water flux and hence is a key factor affecting the subsurface moisture dynamics. It raises the water table and limits moisture storage available in the soil column, which has significant impact on surface energy and water flux dynamics as well (Chen and Kumar 2001). Most LSMs have, however, generally neglected the bedrock or simply fixed it at the bottom of the lowest soil layer. As such, these models may overestimate soil moisture memory in deeper zones, resulting in unrealistic representation of the terrestrial hydrology and regional water recycling processes. To improve estimates of the actual soil water capacity, we adopt the geographically distributed bedrock depth as constructed at the CWRF 30-km grid cells by Liang et al. (2005c). It is based on the conterminous U.S. multilayer soil characteristics (CONUS-SOIL) 1-km dataset developed by Miller and White (1998) from the U.S. Department of Agriculture (USDA) State Soil Geographic (STATSGO) database. Figure 1c illustrates the bedrock depth distribution over the study catchment, ranging from 90 to 132 cm.
Although water rarely penetrates the solid bedrock, certain moisture fluxes can still occur through fractures, fissures, and cracks in the bedrock. It is, however, neither easy to model these fracture flow mechanisms nor sufficient to use available data for bedrock and rock geological characteristics. To approximate the water drainage through the bedrock, we assume that the bedrock has its porosity of 0.05 and its saturated hydraulic conductivity of Df = 1% of the soil layer right above. A similar drainage factor Df has been introduced in many LSMs, such as Goddard Institute for Space Studies (GISS) (Abramopoulos et al. 1988), Simplified Simple Biosphere Model (SSiB) (Xue et al. 1991), version 2 of the Simple Biosphere Model (SiB2) (Sellers et al. 1996), and Parameterization for Land–Atmosphere–Cloud Exchange (PLACE) (Boone and Wetzel 1996).
Figure 2 compares the weekly time series of specific discharges (discharges per unit drainage area) during 1995 as simulated by the CLM and observed at the USGS gauge station KRL4. The model simulations use the original CLM except for three treatments on bottom drainage: 1) free drainage, where water drains out of the bottom layer without any bedrock limit; 2) no drainage, as if the bottom layer is impermeable strata over the entire catchment; and 3) bedrock drainage, as predicted by using the actual bedrock depth distribution (Fig. 1c) along with the prescribed porosity and saturated hydraulic conductivity. The free drainage method cannot capture peak discharges, while the other two approaches produce more realistic runoff variations. The bedrock drainage treatment predicts a runoff pattern similar to the no-drainage assumption but with somewhat smaller magnitudes. Their difference is not large, mainly because, under the impermeable bottom condition (no drainage), lower soil layers are almost fully saturated and thus act very much like bedrocks.
b. Water table depth
Our implementation has two modifications to the scheme of Niu et al. (2007) and Oleson et al. (2008). First, all calculations related to soil moisture allow for both liquid and ice phases of water. Second, instead of roughly estimating deep aquifer storage, we extend one more layer with the bottom at 5.68 m to contain the water table mostly within the predictive soil column. Note that both their and our methods neglect the possible contribution to the baseflow from the deeper aquifer underlying the bottom of the LSM soil column.
Figure 3 compares the simulations using the equilibrium and dynamic methods for the water table depth at a check grid where soil has a constant porosity of 0.425 in layers 1–9 and bedrock in layers 10 and 11. Soil moisture content is saturated (=porosity) in layer 9 during weeks 8–12 but not saturated (<porosity) in layers 1–8 above for the entire integration period. The dynamic method predicts a realistic water table depth located at the bottom of layer 8 during weeks 8–12 when layer 9 is saturated and declines toward the lower portion of layer 9 for the remaining period as the soil becomes drier. The equilibrium method generates a water table depth that is much shallower than the actual interface between the saturated and the unsaturated soil. Consequently, the shallower water table depth leads to greater runoff using the equilibrium than the dynamic method.
Figure 4 compares weekly time series of specific discharge (or runoff) and basinwide mean water table depths for the study catchment during 1995 as simulated by CLM using the dynamic versus equilibrium methods for the water table depth. Consistent with the result at the check grid cell (Fig. 3), the dynamic method simulates a deeper water table and thus generates less surface runoff than the equilibrium method as averaged over the entire catchment. The current CLM represents subsurface runoff as bottom drainage plus saturation excess (see details in section 3e). In late spring and early summer (weeks 18–21), high flow conditions cause more saturation excess and thus larger subsurface runoff, as bottom drainage is confined by the bedrock layers. This results in greater and hence improved total (surface plus subsurface) runoff in response to the deeper water table as predicted by the dynamic method rather than the equilibrium method. Under low flow conditions for the remaining period, the runoff differences between the two methods are relatively small, with the dynamic one somewhat smaller.
c. Soil hydraulic conductivity
The CLM defines thinner upper layers to better resolve details of near-surface processes and progressively thicker lower layers as an exponential function of the depth to integrate deep soil contributions (Dai et al. 2001). Table 2 lists the thickness, depth, and Ksz versus Ks ratio for the 11-layer soil profile. Clearly, the saturated hydraulic conductivity is much greater using Ksz than Ks above the bedrock depth, with the enhancement by a factor of 7.13 (layer 1) to 1.41 (layer 7). In contrast, the conductivity at saturation is reduced below the bedrock by a factor of 0.47 (layer 8) to almost at the bottom. As such, vertical transport of soil moisture near the surface is much faster with the Ksz profile due to the effect of macropores.
Figure 5 compares weekly time series of runoff and basinwide mean water table depth for the study catchment during 1995 as simulated by the CLM using the Ksz versus Ks profile. Here the realistic bedrock depth implementation (section 3a) and dynamic water table solution (section 3b) are incorporated in both simulations. Obviously, soil moisture responses to rainfall changes are faster through vertical transport because of the enhanced hydraulic conductivity when using Ksz rather than Ks. As such, less water is held instantly in upper soil layers, which results in deeper water table and less surface runoff. Note that, at peak discharges (weeks 18–21), total runoff is dominated by the subsurface component of saturation excess by abnormal supersaturation due to a numerical problem in the existing parameterization (see section 3d). The total runoff under the high flow conditions is actually increased (closer to observations) by using Ksz. The result sensitivity to the value of the decay factor f will be discussed later.
d. Surface runoff
Note that our Ke formulation of Eq. (21) has important differences from two existing schemes: the original CLM using total area (i.e., Fliq = 1) and Oleson et al. (2008) using the average of the unfrozen areas in the two adjacent layers. In contrast, we use the minimum of the unfrozen areas in the two adjacent layers. Figure 6 illustrates an example comparing the CLM numerical stability associated with the three Ke formulations. Given the same initial state profile and no external forcing, the two old schemes predict at the next time step unrealistic soil water contents, including negative and supersaturation values (Figs. 6a and 6b). Negative values are numerically adjusted to zero and thus add artificial water into the system. Supersaturation errors are propagated into other soil layers or depleted as subsurface runoff. Our new scheme corrects these problems and produces a stable and physical solution (Fig. 6c).
Figure 8 shows that the old scheme generates soil water supersaturation and subsequently treats the excess water as subsurface runoff. This is obvious, especially at peak discharges (weeks 18–21), as total runoff coincides with surface runoff predicted by our new scheme. Given the use of Eq. (21) for the effective hydraulic conductivity and Eq. (24) for the maximum surface infiltrability, our scheme prevents the numerical supersaturation problem and predicts very little subsurface runoff for the study catchment. This is reasonable because the incorporation of the realistic bedrock depth along with exponential decay profile of the saturated hydraulic conductivity effectively limits bottom drainage.
e. Subsurface runoff
1) Bottom drainage runoff
2) Saturation excess runoff
3) Saturation lateral runoff (baseflow)
Figure 10 compares weekly time series of the model-simulated total specific runoff with observations and basinwide mean water table depth for different values of the decay factor f. Again, all new formulations described in sections 3a–3e are incorporated here. Together, they have made a significant improvement on the total runoff simulation (comparing the result in Fig. 10 with Fig. 2), especially for the peak discharges during weeks 18–21, which are now very close to observations. The new parameterizations for subsurface runoff improve the terrestrial water mass balance. On the other hand, the use of a much larger f (10 versus 1.8) does not contribute much to the baseflow generation. This insensitivity results from the tendency of our new schemes to predict smaller soil water availability. As compared with the original CLM, the dynamic water table scheme generates a deeper water table, the new scheme prevents supersaturated soil water that is recharged to the upper unsaturated layers for raising the water table, the bedrock layers contain less soil water, and the minimum residual moisture restricts the baseflow allocation. Their combinations cause a minor contribution to the baseflow.
4. Conclusions
This study addresses several deficiencies generated from unrealistic assumptions and crude parameterizations in the terrestrial hydrologic scheme of the original CLM (Dai et al. 2003) that can often limit the predictability in key processes related to the land surface water and energy budgets. New parameterizations are thus developed to improve the runoff prediction, which include the following.
Incorporating realistic geographically distributed bedrock depth (Liang et al. 2005c) to improve the estimate of the actual soil water capacity. Neglect of variation in bedrock depth or specifying it at the bottom of the lowest soil layer likely results in overestimation of soil moisture memory in deeper zones. Such a simplification, as done in many existing LSMs, causes unrealistic representation of the terrestrial hydrology and regional water recycling processes. The fracture flow through the bedrock is implemented using a drainage parameter Df as 1% of the soil conductivity immediately above the bedrock.
Replacing an equilibrium approximation with a dynamic prediction of the water table to produce more reasonable variations of the saturated zone depth. The equilibrium assumption calculates total water deficit in unsaturated layers neglecting vertical soil moisture flux and, thus, generates an unrealistic shallower water table. In contrast, the dynamic prediction depicts the groundwater recharge and discharge processes and correctly places the water table right above the saturated zone. As a result, the dynamical scheme predicts a generally deeper water table depth and less surface runoff.
Using an exponential decay function with soil depth for the saturated hydraulic conductivity to consider the effect of macropores, which is larger near the soil surface, especially in vegetated areas. As compared with the original constant conductivity, soil moisture responses to rainfall changes are faster through vertical transport and less water is held instantly in upper soil layers, which results in deeper water table and less surface runoff.
Formulating an effective hydraulic conductivity of the liquid part at the frozen soil interface and imposing a maximum surface infiltration limit to eliminate numerically generated negative or excessive soil moisture solution. The new scheme takes the minimum of the unfrozen areas in the two adjacent layers to produce a mass-conserved and numerically stable solution of soil moisture profile. Effective soil wetness for the minimum residual soil water and saturation lateral runoff limit in a baseflow allocation also make a contribution to mass-conserved terrestrial water balance. The other existing schemes can produce negative and supersaturated moisture content purely by numerical problems; they are subsequently corrected with artificial adjustments, causing serious mass balance errors.
Including additional contributions to subsurface runoff from saturation lateral runoff or baseflow that is controlled by topography. The original CLM predicts subsurface runoff as the sum of only bottom drainage and saturation excess. While other recent studies (Chen and Kumar 2001; Niu and Yang 2003) have shown that baseflow has a significant contribution to the streamflow, we have found that less available soil water predicted by the new scheme results in only a minor contribution to baseflow.
Given the above improvements, the new CLM captures well the major characteristics of the observed total runoff variations. As compared with in the original CLM, the temporal correlation between the modeled and observed weekly runoff time series at the evaluation site increases from 0.623 to 0.762. The corresponding Nash–Sutcliffe efficiency (Nash and Sutcliffe 1970) is enhanced from 0.269 to 0.424. The improvement is especially significant at peak discharges under high flow conditions. The model, however, still simulates steep declining recession curves and relatively small total runoff, primarily because it underestimates the baseflow. Compared with the observed streamflow of 457 mm yr−1, the CLM produces much less total runoff, of 257 and 223 mm yr−1 in the original and new version, respectively. As a result, the modeled evapotranspiration increases from the original 520 to the new 563 mm yr−1. The CLM, like most LSMs, currently ignores the role of surface flow depth on the infiltration rate, which may cause errors in both surface and subsurface flow modeling. Future work seeks to determine whether a conjunctive surface–subsurface flow model, representing full interactions between surface and subsurface dynamics, will solve the recession problem.
Acknowledgments
This research was supported by the Yeungnam University research grants in 2008. Model simulations were conducted using the Illinois State Water Survey facilities and the National Center for Supercomputing Applications (NCSA) supercomputing allocations. The views expressed are those of the authors and do not necessarily reflect those of the sponsoring agencies or the Illinois State Water Survey.
REFERENCES
Abramopoulos, F., Rosenzweig C. , and Choudhury B. , 1988: Improved ground hydrology calculations for global climate models (GCMs): Soil water movement and evapotranspiration. J. Climate, 1 , 921–941.
Beven, K. J., 1982a: Macropores and water flow in soils. Water Resour. Res., 18 , 1311–1325.
Beven, K. J., 1982b: On subsurface stormflow: An analysis of response times. Hydrol. Sci. J., 27 , 505–521.
Beven, K. J., 1984: Infiltration into a class of vertically non-uniform soils. Hydrol. Sci. J., 29 , 425–434.
Beven, K. J., and Kirkby M. J. , 1979: A physically based, variable contributing area model of basin hydrology. Hydrol. Sci. Bull., 24 , 43–69.
Bonan, G. B., 1996: A land surface model (LSM version 1.0) for ecological, hydrological, and atmospheric studies: Technical description and user’s guide. NCAR Tech. Note NCAR/TN-417+STR, 150 pp.
Boone, A., and Wetzel P. J. , 1996: Issues related to low resolution modeling of soil moisture: Experience with the PLACE model. Global Planet. Change, 13 , 161–181.
Brooks, R. H., and Corey A. T. , 1964: Hydraulic properties in porous media. Colorado State University Hydrology Paper 3, 27 pp.
Chen, J., and Kumar P. , 2001: Topographic influence of the seasonal and interannual variation of water and energy balance of basins in North America. J. Climate, 14 , 1989–2014.
Choi, H. I., 2006: 3-D volume averaged soil-moisture transport model: A scalable scheme for representing subgrid topographic control in land-atmosphere interactions. Ph.D. dissertation, University of Illinois at Urbana–Champaign, 189 pp.
Choi, H. I., Kumar P. , and Liang X-Z. , 2007: Three-dimensional volume-averaged soil moisture transport model with a scalable parameterization of subgrid topographic variability. Water Resour. Res., 43 , W04414. doi:10.1029/2006WR005134.
Clapp, R. B., and Hornberger G. M. , 1978: Empirical equations for some soil hydraulic properties. Water Resour. Res., 14 , 601–604.
Cosby, B. J., Hornberger G. M. , Clapp R. B. , and Ginn T. R. , 1984: A statistical exploration of the relationships of soil moisture characteristics to the physical properties of soils. Water Resour. Res., 20 , 682–690.
Dai, Y., and Coauthors, 2001: Common Land Model (CLM): Technical documentation and user’s guide. Georgia Institute of Technology Doc., 69 pp. [Available online at http://climate.eas.gatech.edu/dai/clmdoc.pdf].
Dai, Y., and Coauthors, 2003: The Common Land Model. Bull. Amer. Meteor. Soc., 84 , 1013–1023.
Elsenbeer, H., Cassel D. K. , and Castro J. , 1992: Spatial analysis of soil hydraulic conductivity in a tropical rainforest catchment. Water Resour. Res., 28 , 3201–3214.
Entekhabi, D., and Eagleson P. S. , 1989: Land surface hydrology parameterization for atmospheric general circulation models including subgrid scale spatial variability. J. Climate, 2 , 816–831.
Famiglietti, J. S., and Wood E. F. , 1994: Multiscale modeling of spatially variable water and energy balance processes. Water Resour. Res., 30 , 3061–3078.
Gochis, D. J., and Coauthors, cited. 2004: A ten-year vision for research on terrestrial-atmospheric interactions: Advancing coupled land-atmosphere prediction. CUAHSI Cyberseminar Series Vision Paper, 32 pp. [Available online at http://www.cuahsi.org/cyberseminars/Gochis-20041201-paper.pdf].
Lawrence, D. M., Thornton P. E. , Oleson K. W. , and Bonan G. B. , 2007: The partitioning of evapotranspiration into transpiration, soil evaporation, and canopy evaporation in a GCM: Impacts on land–atmosphere interaction. J. Hydrometeor., 8 , 862–880.
Lawrence, P. J., and Chase T. N. , 2007: Representing a new MODIS consistent land surface in the Community Land Model (CLM3.0). J. Geophys. Res., 112 , G01023. doi:10.1029/2006JG000168.
Levine, J. B., and Salvucci G. D. , 1999: Equilibrium analysis of groundwater–vadose zone interactions and the resulting spatial distribution of hydrologic fluxes across a Canadian prairie. Water Resour. Res., 35 , 1369–1383.
Liang, X-Z., Li L. , Kunkel K. E. , Ting M. , and Wang J. X. L. , 2004: Regional climate model simulation of U.S. precipitation during 1982–2002. Part 1: Annual cycle. J. Climate, 17 , 3510–3528.
Liang, X-Z., and Coauthors, 2005a: Development of land surface albedo parameterization bases on Moderate Resolution Imaging Spectroradiometer (MODIS) data. J. Geophys. Res., 110 , D11107. doi:10.1029/2004JD005579.
Liang, X-Z., Choi H. , Kunkel K. E. , Dai Y. , Joseph E. , Wang J. X. L. , and Kumar P. , 2005b: Development of the regional Climate-Weather Research and Forecasting (CWRF) model: Surface boundary conditions. Illinois State Water Survey Scientific Research Rep. ISWS SR 2005-01, 32 pp. [Available online at http://www.isws.illinois.edu/pubs/pubdetail.asp?CallNumber=ISWS+SR+2005%2D01].
Liang, X-Z., Choi H. , Kunkel K. E. , Dai Y. , Joseph E. , Wang J. X. L. , and Kumar P. , 2005c: Surface boundary conditions for mesoscale regional climate models. Earth Interactions, 9 .[Available online at http://EarthInteractions.org].
Liang, X-Z., Xu M. , Zhu J. , Kunkel K. E. , and Wang J. X. L. , 2005d: Development of the regional Climate-Weather Research and Forecasting Model (CWRF): Treatment of topography. Proc. Joint WRF/MM5 User’s Workshop, Boulder, CO, NCAR, 9.3.
Liang, X-Z., and Coauthors, 2006: Development of the regional Climate-Weather Research and Forecasting model (CWRF): Treatment of subgrid topography effects. Proc. Seventh WRF User’s Workshop, Boulder, CO, NCAR, 7.3.
Mahrt, L., and Pan H. , 1984: A two-layer model of soil hydrology. Bound.-Layer Meteor., 29 , 1–20.
Mesinger, F., and Coauthors, 2006: North American regional reanalysis. Bull. Amer. Meteor. Soc., 87 , 343–360.
Miller, D. A., and White R. A. , 1998: A conterminous United States multilayer soil characteristics dataset for regional climate and hydrology modeling. Earth Interactions, 2 .[Available online at http://EarthInteractions.org].
Nash, J. E., and Sutcliffe J. V. , 1970: River flow forecasting through conceptual models part I — A discussion of principles. J. Hydrol., 10 , 282–290.
Niu, G-Y., and Yang Z-L. , 2003: The versatile integrator of surface atmospheric processes: Part 2: Evaluation of three topography-based runoff schemes. Global Planet. Change, 38 , 191–208.
Niu, G-Y., and Yang Z-L. , 2006: Effects of frozen soil on snowmelt runoff and soil water storage at a continental scale. J. Hydrometeor., 7 , 937–952.
Niu, G-Y., Yang Z-L. , Dickinson R. E. , and Gulden L. E. , 2005: A simple TOPMODEL-based runoff parameterization (SIMTOP) for use in global climate models. J. Geophys. Res., 110 , D21106. doi:10.1029/2005JD006111.
Niu, G-Y., Yang Z-L. , Dickinson R. E. , Gulden L. E. , and Su H. , 2007: Development of a simple groundwater model for use in climate models and evaluation with Gravity Recovery and Climate Experiment data. J. Geophys. Res., 112 , D07103. doi:10.1029/2006JD007522.
Oleson, K. W., and Coauthors, 2008: Improvements to the Community Land Model and their impact on the hydrological cycle. J. Geophys. Res., 113 , G01021. doi:10.1029/2007JG000563.
Qian, T., Dai A. , Trenberth K. E. , and Oleson K. W. , 2006: Simulation of global land surface conditions from 1948 to 2004: Part I: Forcing data and evaluations. J. Hydrometeor., 7 , 953–975.
Salvucci, G. D., and Entekhabi D. , 1994: Equivalent steady soil moisture profile and time compression approximation in water balance modeling. Water Resour. Res., 30 , 2737–2750.
Salvucci, G. D., and Entekhabi D. , 1995: Hillslope and climatic controls on hydrologic fluxes. Water Resour. Res., 31 , 1725–1740.
Sellers, P. J., Los S. O. , Tucker C. J. , Justice C. O. , Dazlich D. A. , Collatz G. J. , and Randall D. A. , 1996: A revised land surface parameterization (SiB2) for atmospheric GCMs. Part II: The generation of global fields of terrestrial biophysical parameters from satellite data. J. Climate, 9 , 706–737.
Sivapalan, M., Beven K. J. , and Wood E. F. , 1987: On hydrologic similarity 2. A scaled model of storm runoff production. Water Resour. Res., 23 , 2266–2278.
Stieglitz, M., Rind D. , Famiglietti J. , and Rosenzweig C. , 1997: An efficient approach to modeling the topographic control of surface hydrology for regional modeling. J. Climate, 10 , 118–137.
Warrach, K., Stieglitz M. , Mengelkamp H-T. , and Raschke E. , 2002: Advantages of a topograpically controlled runoff simulation in a soil–vegetation–atmosphere transfer model. J. Hydrometeor., 3 , 131–148.
Woods, R. A., and Sivapalan M. , 1997: A connection between topographically driven runoff generation and network structure. Water Resour. Res., 33 , 2939–2950.
Xue, Y., Sellers P. J. , Kinter J. L. III, and Shukla J. , 1991: A Simplified Biosphere Model for global climate studies. J. Climate, 4 , 345–364.
(a) Plot of spatial distribution of the 30-km terrain elevation (m), (b) plot of spatial distribution of the 30-km USGS land cover type, and (c) vertically exaggerated schematic of the 11-layer model structure with 30-km DEM and bedrock profiles.
Citation: Journal of Hydrometeorology 11, 3; 10.1175/2010JHM1221.1
Comparison of weekly averaged time series of specific runoff results during 1995 as simulated by CLM with respect to free drainage, no drainage, and bedrock drainage conditions, along with observations at the USGS gauge station KRL4.
Citation: Journal of Hydrometeorology 11, 3; 10.1175/2010JHM1221.1
Comparison of (a) surface-specific runoff results simulated using the equilibrium (old) and dynamic (new) methods for the water table depth and (b) the water table depths computed from the two methods of equilibrium (old) and dynamic (new) schemes, along with soil moisture content at layers 7, 8, and 9 for a check grid cell where soil has a constant porosity of 0.425 in layers 1–9 and bedrock in layers 10 and 11.
Citation: Journal of Hydrometeorology 11, 3; 10.1175/2010JHM1221.1
Comparison of weekly time series of specific runoff and basinwide mean water table depth for a study catchment as simulated by CLM using the dynamic (zwt_new) vs equilibrium (zwt_old) methods for the water table depth, along with the total runoff observations. Both models incorporate the bedrock drainage flux scheme.
Citation: Journal of Hydrometeorology 11, 3; 10.1175/2010JHM1221.1
Comparison of weekly time series of specific runoff and basinwide mean water table depth for the study catchment during 1995 as simulated by CLM using the Ksz (K_exp) vs Ks (K_const) profile, along with the total runoff observations. Both models incorporate bedrock drainage flux and dynamic water table depth change.
Citation: Journal of Hydrometeorology 11, 3; 10.1175/2010JHM1221.1
Example comparison of the vertical profiles for liquid and ice soil water predictions with respect to the effective hydraulic conductivity equations at each soil layer interface using (a) the total liquid and ice water parts, Ke(zk) = K(zk); (b) the averaged liquid water part of the two adjacent layers, Ke(zk) = {[Fliq(k) + Fliq(k + 1)]/2}K(zk); and (c) the minimum unfrozen water part of the two adjacent layers. Ke(zk) = min[Fliq(k), Fliq(k + 1)]K(zk).
Citation: Journal of Hydrometeorology 11, 3; 10.1175/2010JHM1221.1
Schematic of the two adjacent soil layers with different liquid and ice water content in the same soil texture.
Citation: Journal of Hydrometeorology 11, 3; 10.1175/2010JHM1221.1
Comparison of weekly time series of specific runoff and basinwide mean water table depth for the study catchment during 1995 as simulated by CLM using the old scheme allowing supersaturaion (SS_old) and the new scheme preventing supersaturation (SS_new), along with the total runoff observations. Both models incorporate bedrock drainage flux, dynamic water table depth change, and exponential profile of saturated hydraulic conductivity.
Citation: Journal of Hydrometeorology 11, 3; 10.1175/2010JHM1221.1
Schematic diagram for surface runoff and subsurface runoff components in the new version of CLM.
Citation: Journal of Hydrometeorology 11, 3; 10.1175/2010JHM1221.1
Comparison of weekly time series of specific runoff and basinwide mean water table depth for the study catchment during 1995 as simulated by CLM for different values of f, along with the total runoff observations.
Citation: Journal of Hydrometeorology 11, 3; 10.1175/2010JHM1221.1
Atmospheric forcing data required to drive the CLM model.
The CLM soil layer thickness and depth (cm).
