Land Surface Model Parameters Research Articles

Improving soil moisture assimilation efficiency via model calibration using SMAP surface soil moisture climatology information

To minimize systematic differences between soil moisture (SM) time series derived from remote sensing (RS) and land surface model (LSM), RS-based SM climatology information is typically discarded during land data assimilation (DA). However, recent studies have demonstrated that SM climatology estimates provided by L-band microwave RS retrievals can significantly outperform comparable estimates derived from LSMs. Consequently, neglecting a RS-based SM climatology may lead to degraded SM spatial patterns generated by land DA. Here, we propose a climatology-optimized SM DA framework, which first calibrates LSM parameters to leverage the RS SM climatology information. Next, multi-sensor RS SM retrievals are assimilated into the calibrated LSM using Ensemble Kalman Filter (EnKF). This framework is demonstrated using the Variable Infiltration Capacity (VIC) model and RS SM retrievals derived from the L-band Soil Moisture Active Passive (SMAP) and C-band Advanced SCATterometer (ASCAT) sensors. Here, both SMAP and ASCAT SM retrievals are assimilated into the VIC model after calibrating VIC to match spatial variations captured in the SMAP SM climatology. The DA SM results are validated using in-situ SM observations derived from 820 stations. Results show that SMAP-based SM climatology calibration directly improves the quality of SM spatial patterns estimated by VIC. In addition, the SMAP-based calibration also benefits our model-error representation and thereby yields better Kalman gains that improve temporal SM DA accuracy. Relative to typical DA approaches, this newly proposed DA framework improves both spatial (0.26 versus 0.51 (−)) and temporal correlations (0.48 versus 0.52 (−)) versus in-situ SM observations. In addition, SM improvements are effectively propagated into improved streamflow estimates – leading to an average increase of Nash and Sutcliffe coefficient from 0.74 to 0.76 (−). Overall, we demonstrate that RS SM climatology information is valuable for land DA and our climatology-optimized framework successfully retains such information to the benefit of land DA performance.

Inverse Modeling of Hydrologic Parameters in CLM4 via Generalized Polynomial Chaos in the Bayesian Framework

In this work, generalized polynomial chaos (gPC) expansion for land surface model parameter estimation is evaluated. We perform inverse modeling and compute the posterior distribution of the critical hydrological parameters that are subject to great uncertainty in the Community Land Model (CLM) for a given value of the output LH. The unknown parameters include those that have been identified as the most influential factors on the simulations of surface and subsurface runoff, latent and sensible heat fluxes, and soil moisture in CLM4.0. We set up the inversion problem in the Bayesian framework in two steps: (i) building a surrogate model expressing the input–output mapping, and (ii) performing inverse modeling and computing the posterior distributions of the input parameters using observation data for a given value of the output LH. The development of the surrogate model is carried out with a Bayesian procedure based on the variable selection methods that use gPC expansions. Our approach accounts for bases selection uncertainty and quantifies the importance of the gPC terms, and, hence, all of the input parameters, via the associated posterior probabilities.

Uncertainties in Evapotranspiration Estimates over West Africa

An evapotranspiration (ET) ensemble composed of 36 land surface model (LSM) experiments and four diagnostic datasets (GLEAM, ALEXI, MOD16, and FLUXNET) is used to investigate uncertainties in ET estimate over five climate regions in West Africa. Diagnostic ET datasets show lower uncertainty estimates and smaller seasonal variations than the LSM-based ET values, particularly in the humid climate regions. Overall, the impact of the choice of LSMs and meteorological forcing datasets on the modeled ET rates increases from north to south. The LSM formulations and parameters have the largest impact on ET in humid regions, contributing to 90% of the ET uncertainty estimates. Precipitation contributes to the ET uncertainty primarily in arid regions. The LSM-based ET estimates are sensitive to the uncertainty of net radiation in arid region and precipitation in humid region. This study serves as support for better determining water availability for agriculture and livelihoods in Africa with earth observations and land surface models.

Application of the Particle Swarm Optimization in the Land Surface Model Parameters Calibration

Application of the Particle Swarm Optimization in the Land Surface Model Parameters Calibration

The Impact of Error Accounting in a Bayesian Approach to Calibrating Modeled Turbulent Fluxes in an Open-Canopy Forest

Abstract Land surface models are notorious for containing many parameters that control the exchange of heat and moisture between land and atmosphere. Properly modeling the partitioning of total evapotranspiration (ET) between transpiration and evaporation is critical for accurate hydrological modeling, but depends heavily on the treatment of turbulence within and above canopies. Previous work has constrained estimates of evapotranspiration and its partitioning using statistical approaches that calibrate land surface model parameters by assimilating in situ measurements. These studies, however, are silent on the impacts of the accounting of uncertainty within the statistical calibration framework. The present study calibrates the aerodynamic, leaf boundary layer, and stomatal resistance parameters, which partially control canopy turbulent exchange and thus the evapotranspiration flux partitioning. Using an adaptive Metropolis–Hastings algorithm to construct a Markov chain of draws from the joint posterior distribution of these resistance parameters, an ensemble of model realizations is generated, in which latent and sensible heat fluxes and top soil layer temperature are optimized. A set of five calibration experiments demonstrate that model performance is sensitive to the accounting of various sources of uncertainty in the field observations and model output and that it is critical to account for model structural uncertainty. After calibration, the modeled fluxes and top soil layer temperature are largely free from bias, and this calibration approach successfully informs and characterizes uncertainty in these parameters, which is essential for model improvement and development. The key points of this paper are 1) a Markov chain Monte Carlo calibration approach successfully improves modeled turbulent fluxes; 2) ET partitioning estimates hinge on the representation of uncertainties in the model and data; and 3) despite these inherent uncertainties, constrained posterior estimates of ET partitioning emerge.

Land surface model performance using cosmic-ray and point-scale soil moisture measurements for calibration

Abstract. At very high resolution scale (i.e. grid cells of 1 km2), land surface model parameters can be calibrated with eddy-covariance flux data and point-scale soil moisture data. However, measurement scales of eddy-covariance and point-scale data differ substantially. In our study, we investigated the impact of reducing the scale mismatch between surface energy flux and soil moisture observations by replacing point-scale soil moisture data with observations derived from Cosmic-Ray Neutron Sensors (CRNSs) made at larger spatial scales. Five soil and evapotranspiration parameters of the Joint UK Land Environment Simulator (JULES) were calibrated against point-scale and Cosmic-Ray Neutron Sensor soil moisture data separately. We calibrated the model for 12 sites in the USA representing a range of climatic, soil, and vegetation conditions. The improvement in latent heat flux estimation for the two calibration solutions was assessed by comparison to eddy-covariance flux data and to JULES simulations with default parameter values. Calibrations against the two soil moisture products alone did show an advantage for the cosmic-ray technique. However, further analyses of two-objective calibrations with soil moisture and latent heat flux showed no substantial differences between both calibration strategies. This was mainly caused by the limited effect of calibrating soil parameters on soil moisture dynamics and surface energy fluxes. Other factors that played a role were limited spatial variability in surface fluxes implied by soil moisture spatio-temporal stability, and data quality issues.

Improving Weather Predictability by Including Land Surface Model Parameter Uncertainty

AbstractThe land surface forms an important component of Earth system models and interacts nonlinearly with other parts such as ocean and atmosphere. To capture the complex and heterogeneous hydrology of the land surface, land surface models include a large number of parameters impacting the coupling to other components of the Earth system model.Focusing on ECMWF’s land surface model Hydrology Tiled ECMWF Scheme of Surface Exchanges over Land (HTESSEL), the authors present in this study a comprehensive parameter sensitivity evaluation using multiple observational datasets in Europe. The authors select six poorly constrained effective parameters (surface runoff effective depth, skin conductivity, minimum stomatal resistance, maximum interception, soil moisture stress function shape, and total soil depth) and explore their sensitivity to model outputs such as soil moisture, evapotranspiration, and runoff using uncoupled simulations and coupled seasonal forecasts. Additionally, the authors investigate the possibility to construct ensembles from the multiple land surface parameters.In the uncoupled runs the authors find that minimum stomatal resistance and total soil depth have the most influence on model performance. Forecast skill scores are moreover sensitive to the same parameters as HTESSEL performance in the uncoupled analysis. The authors demonstrate the robustness of these findings by comparing multiple best-performing parameter sets and multiple randomly chosen parameter sets. The authors find better temperature and precipitation forecast skill with the best-performing parameter perturbations demonstrating representativeness of model performance across uncoupled (and hence less computationally demanding) and coupled settings.Finally, the authors construct ensemble forecasts from ensemble members derived with different best-performing parameterizations of HTESSEL. This incorporation of parameter uncertainty in the ensemble generation yields an increase in forecast skill, even beyond the skill of the default system.

Parameter sensitivity analysis and optimization of Noah land surface model with field measurements from Huaihe River Basin, China

This study aims to identify the parameters that are most important in controlling the Noah land surface model (LSM), the analysis of parameter interactions, and the evaluation of the performance of parameter optimization using the parameter estimation software PEST. We found it necessary to analyze parameter sensitivity in order to properly simulate hydrological variables such as latent heat flux in the Huaihe River Basin, China. The parameters under study in the Noah LSM link thermodynamic and hydrological parts into a complete model. To our knowledge, this parameter interaction in the Noah LSM has never been studied before. There are, however, several studies concerning the influence of vegetation types and climate conditions on parameter sensitivity of the Noah LSM. Three sensitivity analysis methods, the including local sensitivity analysis method SENSAN, regional sensitivity analysis, and Sobol’s method, were tested. Five experimental sites in the Huaihe River Basin were chosen to perform the simulations. The results show that the Noah LSM parameter sensitivities were impacted by the choice of the analysis method. The local method SENSAN often produced significant differences in results compared to the two global methods. The parameter interactions investigated made a significant contribution towards elucidating how one process influences another in the Noah LSM. The results show that parameters were not transferable solely based on vegetation types but also rely on climate conditions. According to the sensitivity analysis results, four sensitive parameters were chosen to be optimized using the PEST method. PEST is a widely used method for estimating parameters in models. Root-mean-square error was used to evaluate the effect of the optimization. Generally in all sites, the optimized parameters values perform better than the original parameter values.

An updated treatment of soil texture and associated hydraulic properties in a global land modeling system

AbstractThe advent of new data sets describing soil texture and associated soil properties offers the promise of improved hydrological simulation. Here we describe the composition of a new soil texture data set and its implementation into a specific land surface modeling system, namely, the Catchment land surface model (LSM) of the NASA Goddard Earth Observing System version 5 (GEOS‐5) modeling and assimilation framework. First, global soil texture composites are generated using data from the Harmonized World Soil Database version 1.21 (HWSD1.21) and the State Soil Geographic (STATSGO2) project, with explicit consideration of different levels of organic material. Then, the LSM's soil parameters are upgraded using the new texture data, with hydraulic parameters derived for the more extensive set of texture classes using pedotransfer functions. Other changes to the LSM parameters are included to further support simulations at increasingly fine resolutions. A suite of simulations with the original and new parameter versions shows modest yet significant improvements in the Catchment LSM's simulation of soil moisture and surface hydrological fluxes. The revised LSM parameters will be used for the forthcoming Soil Moisture Active Passive (SMAP) soil moisture assimilation product.

An alternate and robust approach to calibration for the estimation of land surface model parameters based on remotely sensed observations

[1] Atmospheric models include land surface parameterizations of heat and moisture fluxes. Most parameterizations derive from early models of layered soil and vegetation canopy that account for liquid, vapor and heat diffusion, radiation processes, and surface turbulence. The number of parameters in these models ranges roughly from 10 to 50. Some parameters are known to vary over a small range and/or not significantly impact predictions. Others are highly influential (e.g., Leaf Area Index), but are currently well estimated from satellite data. Many parameters, however, are highly influential, vary over a large dynamic range, cannot be estimated from satellite data, and cannot be readily upscaled from in-situ measurements. For such parameters (e.g., maximum stomatal conductance and soil hydraulic conductivity), values are assigned based on look-up tables sorted by land cover and soil texture. Here we present a method for estimating these parameters by minimizing a measure of nonstationarity of model-predicted moisture state variable tendencies. This method has advantages over calibration: 1) it does not require flux data (e.g., evapotranspiration); and 2) the tendency terms are evaluated at the model grid and thus yield parameters that are effective for that scale. The method is demonstrated with the Noah Land Surface Model, using remotely-sensed soil moisture, at a site in California. Preliminary results indicate that the method is robust and performs better than both: 1) calibration to soil moisture observations, which can lead to large, compensating errors in drainage and evaporation; and 2) minimizing the sum of squares of innovations of soil moisture updates.

Using conditional nonlinear optimal perturbation method in parameter optimization of land surface processes model

In this paper, we attempted to entend the application of conditional nonlinear optimal perturbation(CNOP) to the optimization of parameters in land surface model. We used the common land model and data of Tongyu station,which is a reference site of the CEOP in the semi-arid regions, and used three key parameters (soil color, soil sand/lay proportion and leaf area index) as parameters to be optimized. Two experiments are designed in our work, namely the single-parameter optimization and the triple-parameter optimization. Notable improvements in simulating sensible heat flux (SH), latent heat flux (LH), soil temperature (TS) and moisture (MS) at shallow layers were achieved by using the optimized parameters. In addition, the latter experiment shows a better performance than the former. All results above illustrate that the application of CNOP method can be extended to parameters optimization of land surface model. And what is more, due to its other advantages, such as the clear mathematical meaning, the simple design structure, and the fast computing speed, it shows a great potential for further applications in parameters optimization of related problems.

Likelihood parameter estimation for calibrating a soil moisture model using radar bakscatter

Land surface model parameter estimation can be performed using soil moisture information provided by synthetic aperture radar imagery. The presence of speckle necessitates aggregating backscatter measurements over large (> 100 m × 100 m) land areas in order to derive reliable soil moisture information from imagery, and a model calibrated to such aggregated information can only provide estimates of soil moisture at spatial resolutions required for reliable speckle accounting. A method utilizing the likelihood formulation of a probabilistic speckle model as the calibration objective function is proposed which will allow for calibrating land surface models directly to radar backscatter intensity measurements in a way which simultaneously accounts for model parameter- and speckle-induced uncertainty. The method is demonstrated using the NOAH land surface model and Advanced Integral Equation Method (AIEM) backscatter model calibrated to SAR imagery of an area in the Southwestern United States, and validated against in situ soil moisture measurements. At spatial resolutions finer than 100 m × 100 m NOAH and AIEM calibrated using the proposed radar intensity likelihood parameter estimation algorithm predict surface level soil moisture to within 4% volumetric water content 95% of the time, which is an improvement over a 95% prediction confidence of 10% volumetric water content by the same models calibrated directly to soil moisture information derived from synthetic aperture radar imagery at the same scales. Results suggest that much of this improvement is due to increased ability to simultaneously estimate NOAH parameters and AIEM surface roughness parameters.

Improving parameter estimation and water table depth simulation in a land surface model using GRACE water storage and estimated base flow data

Several previous studies have shown the significance of representing shallow groundwater in land surface model (LSM) simulations. However, optimal methods for parameter estimation in order to realistically simulate water table depth have received little attention. The recent availability of Gravity Recovery and Climate Experiment (GRACE) water storage data provides a unique opportunity to constrain LSM simulations of terrestrial hydrology. In this study, we incorporate both GRACE (storage) and estimated base flow (flux) data in the calibration of LSM parameters, and demonstrate the advantages gained from this approach using a Monte Carlo simulation framework. This approach improves parameter estimation and reduces the uncertainty of water table simulations in the LSM. Using the optimal parameter set identified from the multiobjective calibration, water table simulation can be improved due to close dependence of both base flow and total subsurface water storage on the water table depth. Moreover, it is shown that parameters calibrated from short‐term (2003–2005) GRACE and base flow data can be validated using simulations for the periods of 1984–1998 and 2006–2007, which implies that the proposed multiobjective calibration strategy is robust. More important, this study has demonstrated the potential for the joint use of routinely available GRACE water storage data and streamflow records to constrain LSM simulations at the global scale.

The role of land surface processes on the mesoscale simulation of the July 26, 2005 heavy rain event over Mumbai, India

A record-breaking heavy rain event occurred over Mumbai, India on July 26th, 2005 with 24-h rainfall exceeding 944 mm. Operational weather forecast models failed to predict the intensity and amount of heavy rainfall. The objective of this study was to test the impact of the three different land surface models when coupled to the Weather Research Forecasting (WRF), and also to investigate the ability of the WRF model to simulate the Mumbai heavy rain event. Numerical experiments were designed using the WRF model, with three nested domains (30, 10, and 3.3 km grid spacing). Results confirmed that the simulated rainfall is sensitive to the grid spacing (with finer grids leading to higher rainfall). Results also suggest that simulated precipitation amounts are sensitive to the choice of cumulus parameterization (with Grell–Devenyi cumulus scheme performing relatively best). To reduce the confounding impact of cumulus parameterization in studying the impacts of land surface models, we evaluated results for the 3.3 km grid spacing domain with explicit convection. Simulations were performed from 12Z, July 25th to 00Z, July 27th with identical boundary conditions and model configurations for three different land surface models (the Slab, the Noah, and a modified version with photosynthesis module—the Noah-GEM). The model results were compared with observed rainfall, surface temperature, and operational soundings over three locations: Mumbai, Bangalore and Bhopal. Model results showed that: (i) The simulated rainfall was sensitive to the chosen land surface model. The rainfall spatial distributions, as well as their temporal characteristics, were different for each of the three WRF runs with different LSMs. (ii) In contrast to the findings over mid-latitudes, the relatively simpler Slab model had a relatively better performance than the modestly complex Noah and Noah-GEM LSMs. For example, the highest observed rainfall over Mumbai was 944 mm and the simulated amounts for Slab, Noah and Noah-GEM runs were 781 mm, 733 mm and 678 mm, respectively. (iii) Overall, the Slab model simulated a relatively cooler surface and a shallower boundary layer. Most significantly, the Slab model resulted in a convergence hotspot at both the 850 mb and 500 mb levels, which lead to high moisture accumulation and higher rainfall activity over Mumbai. Noah and Noah-GEM, on the other hand, resulted in a divergence zone over Mumbai and the Western Ghats leading to more widespread runs but relatively lower rainfall amounts over Mumbai. Additional synthetic experiments were performed to test the sensitivity of land use land cover, the model start time and run duration. Results indicated that the WRF model was able to reproduce several features of the Mumbai rain event, and that the land surface representations would have substantial impact on the heavy rain simulations. Future studies with more up to date land use land cover data, and regional calibration of the land surface model parameters, show the potential for improving the performance of the Noah-WRF over the Indian monsoon region.

Dynamic Calibration with an Ensemble Kalman Filter Based Data Assimilation Approach for Root-Zone Moisture Predictions

AbstractIn the presence of uncertain initial conditions and soil hydraulic properties, land surface model (LSM) performance can be significantly improved by the assimilation of periodic observations of certain state variables, such as the near-surface soil moisture (θg), as observed from a remote platform. In this paper the possibility of merging observations and the model optimally for providing robust predictions of root-zone soil moisture (θ2) is demonstrated. An assimilation approach that assimilates θg through the ensemble Kalman filter (EnKF) and provides a physics-based update of θ2 is developed. This approach, as with other common soil moisture assimilation approaches, may fail when a key LSM parameter, for example, the saturated hydraulic conductivity (ks), is estimated poorly. This leads to biased model errors producing a violation of a main assumption (model errors with zero mean) of the EnKF. For overcoming this model bias an innovative assimilation approach is developed that accepts this violation in the early model run times and dynamically calibrates all the components of the ks ensemble as a function of the persistent bias in root-zone soil moisture, allowing one to remove the model bias, restore the fidelity to the EnKF requirements, and reduce the model uncertainty. The robustness of the proposed approach is also examined in sensitivity analyses.

Description and Evaluation of the Characteristics of the NCAR High-Resolution Land Data Assimilation System

AbstractThis paper describes important characteristics of an uncoupled high-resolution land data assimilation system (HRLDAS) and presents a systematic evaluation of 18-month-long HRLDAS numerical experiments, conducted in two nested domains (with 12- and 4-km grid spacing) for the period from 1 January 2001 to 30 June 2002, in the context of the International H2O Project (IHOP_2002). HRLDAS was developed at the National Center for Atmospheric Research (NCAR) to initialize land-state variables of the coupled Weather Research and Forecasting (WRF)–land surface model (LSM) for high-resolution applications. Both uncoupled HRDLAS and coupled WRF are executed on the same grid, sharing the same LSM, land use, soil texture, terrain height, time-varying vegetation fields, and LSM parameters to ensure the same soil moisture climatological description between the two modeling systems so that HRLDAS soil state variables can be used to initialize WRF–LSM without conversion and interpolation. If HRLDAS is initialized with soil conditions previously spun up from other models, it requires roughly 8–10 months for HRLDAS to reach quasi equilibrium and is highly dependent on soil texture. However, the HRLDAS surface heat fluxes can reach quasi-equilibrium state within 3 months for most soil texture categories. Atmospheric forcing conditions used to drive HRLDAS were evaluated against Oklahoma Mesonet data, and the response of HRLDAS to typical errors in each atmospheric forcing variable was examined. HRLDAS-simulated finescale (4 km) soil moisture, temperature, and surface heat fluxes agreed well with the Oklahoma Mesonet and IHOP_2002 field data. One case study shows high correlation between HRLDAS evaporation and the low-level water vapor field derived from radar analysis.

Numerical investigation of the impact of uncertainties in satellite rainfall estimation and land surface model parameters on simulation of soil moisture

This study aims at evaluating the uncertainty in the prediction of soil moisture (1D, vertical column) from an offline land surface model (LSM) forced by hydro-meteorological and radiation data. We focus on two types of uncertainty: an input error due to satellite rainfall retrieval uncertainty, and, LSM soil-parametric error. The study is facilitated by in situ and remotely sensed data-driven (precipitation, radiation, soil moisture) simulation experiments comprising a LSM and stochastic models for error characterization. The parametric uncertainty is represented by the generalized likelihood uncertainty estimation (GLUE) technique, which models the parameter non-uniqueness against direct observations. Half-hourly infra-red (IR) sensor retrievals were used as satellite rainfall estimates. The IR rain retrieval uncertainty is characterized on the basis of a satellite rainfall error model (SREM). The combined uncertainty (i.e., SREM + GLUE) is compared with the partial assessment of uncertainty. It is found that precipitation (IR) error alone may explain moderate to low proportion of the soil moisture simulation uncertainty, depending on the level of model accuracy—50–60% for high model accuracy, and 20–30% for low model accuracy. Comparisons on the basis of two different sites also yielded an increase (50–100%) in soil moisture prediction uncertainty for the more vegetated site. This study exemplified the need for detailed investigations of the rainfall retrieval-modeling parameter error interaction within a comprehensive space-time stochastic framework for achieving optimal integration of satellite rain retrievals in land data assimilation systems.

Using different hydrological variables to assess the impacts of atmospheric forcing errors on optimization and uncertainty analysis of the CHASM surface model at a cold catchment

Estimation of parameters for land‐surface models, along with their corresponding uncertainties, relies on the input data for the atmospheric forcing variables including atmospheric pressure, temperature, humidity, wind speed, precipitation, and incoming shortwave and longwave radiation. Most studies assume that forcing data are accurate and contain no random or systematic observational errors. In practice, there are indeed systematic errors in precipitation measurements, especially for snowfall, due to wind‐caused undercatch. Incoming shortwave and longwave radiation fluxes are often not directly measured, but estimated from empirical formulations. Uncertainties in these forcing data may substantially affect optimization and uncertainty estimates of land surface models. In this study, we used 18‐year forcing and calibration data as well as information about the uncertainties in the forcing variables at Valdai, Russia, to study the impacts of forcing errors on selection of optimal model parameters and their uncertainty estimates when three different hydrological variables were used for calibration. The results show that forcing errors have few effects on the selection of optimal model parameter sets when monthly evapotranspiration and runoff are calibrated. However, forcing errors do introduce significant effects on the selection of optimal model parameters when daily snow water equivalent is calibrated. Forcing errors also significantly affect uncertainty estimates of the land surface model parameters. In addition, constraints of forcing errors are different when different hydrological variables are calibrated. All three hydrological variables constrain the incoming longwave radiation error well, and the snow water equivalent and runoff constrain winter snowfall errors well. However, all three hydrological variables cannot constrain the incoming solar radiation error well. We highlight in this study that runoff is shown to be a good observable to use for calibration, the reason being that it integrates multiple hydrological processes; and the results support the theory that typical rain/snow gauges have 10–20% undercatch.

Calibrating a Land Surface Model of Varying Complexity Using Multicriteria Methods and the Cabauw Dataset

Abstract The multicriteria methodology, which provides a means to estimate optimal ranges for land surface model parameter values via calibration, is evaluated. Following calibration, differences between schemes resulting from effective parameter values can be isolated from differences resulting from scheme structure or scheme parameterizations. The method is applied to the Project for the Intercomparison of Land Surface Parameterization Schemes (PILPS) phase-2a data from the Cabauw site in the Netherlands using the Chameleon Surface Model (CHASM) as the surrogate for a range of land surface schemes. Simulations are performed calibrating six modes of CHASM, representing a range of land surface complexity, against observed net radiation and latent and sensible heat fluxes. The six modes range from a simple bucket model to a complex mosaic-type structure with separate energy balances for each mosaic tile and explicit treatment of transpiration, canopy interception, and bare-ground evaporation. Results demon...