Dag Lohmann

The Global Land Data Assimilation System

A Global Land Data Assimilation System (GLDAS) has been developed. Its purpose is to ingest satellite- and ground-based observational data products, using advanced land surface modeling and data assimilation techniques, in order to generate optimal fields of land surface states and fluxes. GLDAS is unique in that it is an uncoupled land surface modeling system that drives multiple models, integrates a huge quantity of observation-based data, runs globally at high resolution (0.25°), and produces results in near–real time (typically within 48 h of the present). GLDAS is also a test bed for innovative modeling and assimilation capabilities. A vegetation-based “tiling” approach is used to simulate subgrid-scale variability, with a 1-km global vegetation dataset as its basis. Soil and elevation parameters are based on high-resolution global datasets. Observation-based precipitation and downward radiation and output fields from the best available global coupled atmospheric data assimilation systems are employe...

Predicting the discharge of global rivers

The ability to simulate coupled energy and water fluxes over large continental river basins, in particular streamflow, was largely nonexistent a decade ago. Since then, macroscale hydrological models (MHMs) have been developed, which predict such fluxes at continental and subcontinental scales. Because the runoff formulation in MHMs must be parameterized because of the large spatial scale at which they are implemented, some calibration of model parameters is inevitably necessary. However, calibration is a time-consuming process and quickly becomes infeasible when the modeled area or the number of basins increases. A methodology for model parameter transfer is described that limits the number of basins requiring direct calibration. Parameters initially were estimated for nine large river basins. As a first attempt to transfer parameters, the global land area was grouped by climate zone, and model parameters were transferred within zones. The transferred parameters were then used to simulate the water balance in 17 other continental river basins. Although the parameter transfer approach did not reduce the bias and root-mean-square error (rmse) for each individual basin, in aggregate the transferred parameters reduced the relative (monthly) rmse from 121% to 96% and the mean bias from 41% to 36%. Subsequent direct calibration of all basins further reduced the relative rmse to an average of 70% and the bias to 12%. After transferring the parameters globally, the mean annual global runoff increased 9.4% and evapotranspiration decreased by 5.0% in comparison with an earlier global simulation using uncalibrated parameters. On a continental basis, the changes in runoff and evapotranspiration were much larger. A diagnosis of simulation errors for four basins with particularly poor results showed that most of the error was attributable to bias in the Global Precipitation Climatology Project precipitation products used to drive the MHM.

Regional scale hydrology: I. Formulation of the VIC-2L model coupled to a routing model

Abstract A grid network version of the two-layer Variable Infiltration Capacity (VIC-2L) macroscale hydrological model is described. The VIC-2L model is a hydrologically-based SVAT (Soil Vegetation Atmospheric Transfer) scheme designed to represent the land surface in numerical weather prediction and climate models. It is coupled to a linear routing scheme which is optimized with measured precipitation and streamflow data and is derived independently from the VIC-2L model. In this way it is possible to utilize streamflow measurements for the validation of coupled atmospheric-hydrological models. A baseflow separation routine is used to derive an equivalent description between the VIC-2L model and the routing model.

The Project for Intercomparison of Land-surface Parameterization / / Schemes PILPS Phase 2 c Red-Arkansas River basin experiment: 1. Experiment description and summary intercomparisons

Abstract Sixteen land-surface schemes participating in the Project for the Intercomparison of Land-surface Schemes (PILPS) Phase 2(c) were run using 10 years (1979–1988) of forcing data for the Red–Arkansas River basins in the Southern Great Plains region of the United States. Forcing data (precipitation, incoming radiation and surface meteorology) and land-surface characteristics (soil and vegetation parameters) were provided to each of the participating schemes. Two groups of runs are presented. (1) Calibration–validation runs, using data from six small catchments distributed across the modeling domain. These runs were designed to test the ability of the schemes to transfer information about model parameters to other catchments and to the computational grid boxes. (2) Base-runs, using data for 1979–1988, designed to evaluate the ability of the schemes to reproduce measured energy and water fluxes over multiple seasonal cycles across a climatically diverse, continental-scale basin. All schemes completed the base-runs but five schemes chose not to calibrate. Observational data (from 1980–1986) including daily river flows and monthly basin total evaporation estimated through an atmospheric budget analysis, were used to evaluate model performance. In general, the results are consistent with earlier PILPS experiments in terms of differences among models in predicted water and energy fluxes. The mean annual net radiation varied between 80 and 105 W m −2 (excluding one model). The mean annual Bowen ratio varied from 0.52 to 1.73 (also excluding one model) as compared to the data-estimated value of 0.92. The run-off ratios varied from a low of 0.02 to a high of 0.41, as compared to an observed value of 0.15. In general, those schemes that did not calibrate performed worse, not only on the validation catchments, but also at the scale of the entire modeling domain. This suggests that further PILPS experiments on the value of calibration need to be carried out.

The project for intercomparison of land-surface parameterization schemes (PILPS) phase 2(c) Red-Arkansas River basin experiment: 3. Spatial and temporal analysis of water fluxes

The energy components of sixteen Soil-Vegetation Atmospheric Transfer (SVAT) schemes were analyzed and intercompared using 10 years of surface meteorological and radiative forcing data from the Red-Arkansas River basin in the Southern Great Plains of the United States. Comparisons of simulated surface energy fluxes among models showed that the net radiation and surface temperature generally had the best agreement among the schemes. On an average (annual and monthly) basis, the estimated latent heat fluxes agreed (to within approximate estimation errors) with the latent heat fluxes derived from a radiosonde-based atmospheric budget method for slightly more than half of the schemes. The sensible heat fluxes had larger differences among the schemes than did the latent heat fluxes, and the model-simulated ground heat fluxes had large variations among the schemes. The spatial patterns of the model-computed net radiation and surface temperature were generally similar among the schemes, and appear reasonable and consistent with observations of related variables, such as surface air temperature. The spatial mean patterns of latent and sensible heat fluxes were less similar than for net radiation, and the spatial patterns of the ground heat flux vary greatly among the 16 schemes. Generally, there is less similarity among the models in the temporal (interannual) variability of surface fluxes and temperature than there is in the mean fields, even for schemes with similar mean fields.

The Rhône-Aggregation Land Surface Scheme Intercomparison Project: An Overview

The Rhone-Aggregation (Rhone-AGG) Land Surface Scheme (LSS) intercomparison project is an initiative within the Global Energy and Water Cycle Experiment (GEWEX)/Global Land-Atmosphere System Study (GLASS) panel of the World Climate Research Programme (WCRP). It is a intermediate step leading up to the next phase of the Global Soil Wetness Project (GSWP) (Phase 2), for which there will be a broader investigation of the aggregation between global scales (GSWP-1) and the river scale. This project makes use of the Rhone modeling system, which was developed in recent years by the French research community in order to study the continental water cycle on a regional scale. The main goals of this study are to investigate how 15 LSSs simulate the water balance for several annual cycles compared to data from a dense observation network consisting of daily discharge from over 145 gauges and daily snow depth from 24 sites, and to examine the impact of changing the spatial scale on the simulations. The overall evapotranspiration, runoff, and monthly change in water storage are similarly simulated by the LSSs, however, the differing partitioning among the fluxes results in very different river discharges and soil moisture equilibrium states. Subgrid runoff is especially important for discharge at the daily timescale and for smaller-scale basins. Also, models using an explicit treatment of the snowpack compared better with the observations than simpler composite schemes. Results from a series of scaling experiments are examined for which the spatial resolution of the computational grid is decreased to be consistent with large-scale atmospheric models. The impact of upscaling on the domain-averaged hydrological components is similar among most LSSs, with increased evaporation of water intercepted by the canopy and a decrease in surface runoff representing the most consistent inter-LSS responses. A significant finding is that the snow water equivalent is greatly reduced by upscaling in all LSSs but one that explicitly accounts for subgrid-scale orography effects on the atmospheric forcing.

Simulation of high-latitude hydrological processes in the Torne-Kalix basin: PILPS Phase 2(e) 1: Experiment description and summary intercomparisons

Abstract Twenty-one land-surface schemes (LSSs) participated in the Project for Intercomparison of Land-surface Parameterizations (PILPS) Phase 2(e) experiment, which used data from the Torne–Kalix Rivers in northern Scandinavia. Atmospheric forcing data (precipitation, air temperature, specific humidity, wind speed, downward shortwave and longwave radiation) for a 20-year period (1979–1998) were provided to the 21 participating modeling groups for 218 1/4° grid cells that represented the study domain. The first decade (1979–1988) of the period was used for model spin-up. The quality of meteorologic forcing variables is of particular concern in high-latitude experiments and the quality of the gridded dataset was assessed to the extent possible. The lack of sub-daily precipitation, underestimation of true precipitation and the necessity to estimate incoming solar radiation were the primary data concerns for this study. The results from two of the three types of runs are analyzed in this, the first of a three-part paper: (1) calibration–validation runs—calibration of model parameters using observed streamflow was allowed for two small catchments (570 and 1300 km2), and parameters were then transferred to two other catchments of roughly similar size (2600 and 1500 km2) to assess the ability of models to represent ungauged areas elsewhere; and 2) reruns—using revised forcing data (to resolve problems with apparent underestimation of solar radiation of approximately 36%, and certain other problems with surface wind in the original forcing data). Model results for the period 1989–1998 are used to evaluate the performance of the participating land-surface schemes in a context that allows exploration of their ability to capture key processes spatially. In general, the experiment demonstrated that many of the LSSs are able to capture the limitations imposed on annual latent heat by the small net radiation available in this high-latitude environment. Simulated annual average net radiation varied between 16 and 40 W/m2 for the 21 models, and latent heat varied between 18 and 36 W/m2. Among-model differences in winter latent heat due to the treatment of aerodynamic resistance appear to be at least as important as those attributable to the treatment of canopy interception. In many models, the small annual net radiation forced negative sensible heat on average, which varied among the models between −11 and 9 W/m2. Even though the largest evaporation rates occur in the summer (June, July and August), model-predicted snow sublimation in winter has proportionately more influence on differences in annual runoff volume among the models. A calibration experiment for four small sub-catchments of the Torne–Kalix basin showed that model parameters that are typically adjusted during calibration, those that control storage of moisture in the soil column or on the land surface via ponding, influence the seasonal distribution of runoff, but have relatively little impact on annual runoff ratios. Similarly, there was no relationship between annual runoff ratios and the proportion of surface and subsurface discharge for the basin as a whole.

Regional scale hydrology: II. Application of the VIC-2L model to the Weser River, Germany

Abstract This paper describes the application of a grid network version of the twolayer Variable Infiltration Capacity (VIC-2L) macroscale hydrological model. The VIC-2L model is implemented on a rotated grid, which is compatible with the weather forecast and climate model REMO (Regional Model), a joint project of the German Weather Service (DWD), GKSS Research Centre and the Max-Planck-Institute for Meteorology, Hamburg. Observed surface meteorological data in the Weser River basin are used to force the model off-line on a daily time step. After a 22-month calibration period, simulated and measured streamflow data are compared for a 12-year period. The resulting predictions compare well with observations at daily, monthly and annual time scales. A sensitivity analysis is presented.

Validation of the North American Land Data Assimilation System (NLDAS) retrospective forcing over the southern Great Plains

[1] Atmospheric forcing used by land surface models is a critical component of the North American Land Data Assimilation System (NLDAS) and its quality crucially affects the final product of NLDAS and our work on model improvement. A three-year (September 1996-September 1999) retrospective forcing data set was created from the Eta Data Assimilation System and observations and used to run the NLDAS land surface models for this period. We compared gridded NLDAS forcing with station observations obtained from networks including the Oklahoma Mesonet and Atmospheric Radiation Measurement/Cloud and Radiation Testbed at the southern Great Plains. Differences in all forcing variables except precipitation between the NLDAS forcing data set and station observations are small at all timescales. While precipitation data do not agree very well at an hourly timescale, they do agree better at longer timescales because of the way NLDAS precipitation forcing is generated. A small high bias in downward solar radiation and a low bias in downward longwave radiation exist in the retrospective forcing. To investigate the impact of these differences on land surface modeling we compared two sets of model simulations, one forced by the standard NLDAS product and one with station-observed meteorology. The differences in the resulting simulations of soil moisture and soil temperature for each model were small, much smaller than the differences between the models and between the models and observations. This indicates that NLDAS retrospective forcing provides an excellent state-of-the-art data set for land surface modeling, at least over the southern Great Plains region.

Simulation of high latitude hydrological processes in the Torne–Kalix basin: PILPS Phase 2(e): 2: Comparison of model results with observations

Model results from 21 land-surface schemes (LSSs) designed for use in numerical weather prediction and climate models are compared with each other and with observations in the context of the Project for Intercomparison of Land-surface Parameterization Schemes (PILPS) Phase 2(e) model intercomparison experiment. This experiment focuses on simulations of land-surface water and energy fluxes in the 58,000-km2 Torne and Kalix river systems in northern Scandinavia, during the period 1989–1998. All models participating in PILPS Phase 2(e) capture the broad dynamics of snowmelt and runoff, but large differences in snow accumulation and ablation, turbulent heat fluxes, and streamflow exist. The greatest among-model differences in energy and moisture fluxes in these high-latitude environments occur during the spring snowmelt period, reflecting different model parameterizations of snow processes. Differences in net radiation are governed by differences in the simulated radiative surface temperature during the winter months and by differences in surface albedo during the spring/early summer. Differences in net radiation are smallest during the late summer when snow is absent. Although simulated snow sublimation is small for most models, a few models show annual snow sublimation of about 100 mm. These differences in snow sublimation appear to be largely dependent on differences in snow surface roughness parameterizations. The models with high sublimation generally lose their snowpacks too early compared to observations and underpredict the annual runoff. Differences in runoff parameterizations are reflected in differences in daily runoff statistics. Although most models show a greater variability in daily streamflow than the observations, the models with the greatest variability (as much as double the observed variability), produce most of their runoff through fast response, surface runoff mechanisms. As a group, those models that took advantage of an opportunity to calibrate to selected small catchments and to transfer calibration results to the basin at large had a smaller bias and root mean squared error (RMSE) in daily streamflow simulations compared with the models that did not calibrate.