H23A-0950
Bias estimation and assimilation of land surface temperature
Satellite retrievals of land surface temperature (LST) are available from a variety of polar orbiting and geostationary platforms. Assimilating such LST retrievals into a land surface model (that is either driven by observed meteorological forcing data or coupled to an atmospheric model) should improve estimates of land surface conditions. However, LST data from retrievals and models typically exhibit very different climatologies for a variety of reasons, including model trade-offs between numerical stability and computational cost, uncertainties in land surface emissivity, satellite look-angle, and other sensor characteristics. We overcome the challenges facing LST assimilation through scaling and bias estimation approaches. In the scaling approach, the LST retrievals from each sensor are scaled to the model's LST climatology before they are assimilated into the land model. After assimilation, the merged LST product may be scaled back into the climatology of the LST retrievals if the application calls for it. Because of the strong seasonal and diurnal cycle of LST, scaling parameters must be derived separately for each 3-hour interval and for each month. The bias estimation approach dynamically estimates diurnally varying model bias parameters. This approach may be more appropriate as long as the satellite climatology of the LST retrievals is homogenenous across all the sensors that are utilized. In the presentation, we compare the different approaches by assimilating land surface temperature retrievals from the International Satellite Cloud Climatology Project (ISCCP) into the NASA Catchment land surface model with an ensemble-based data assimilation system developed at the NASA Global Modeling and Assimilation Office. The assimilation estimates are evaluated against in situ measurements of land surface temperature and surface turbulent fluxes from the Coordinated Energy and Water Cycle Observations Project (CEOP).
H23A-0951
Estimating the observation error of AMSR-E soil moisture retrievals through adaptive data assimilation
Land data assimilation systems merge estimates from land models with observations of the land surface state based on their respective uncertainties and aim to produce estimates that are superior to the model estimates and observations alone. Poorly specified model and observation error parameters (such as error standard deviations), however, negatively affect the quality of the assimilation products. For very poor input error parameters, the assimilation estimates may even be worse than model estimates without data assimilation. Adaptive data assimilation approaches dynamically estimate the input error parameters along with the state estimates by continually adjusting the error parameters in response to internal filter diagnostics. The observation error standard deviation is therefore a by-product of the adaptive assimilation procedure. In this presentation, we use an adaptive data assimilation approach developed at the NASA Global Modeling and Assimilation Office to derive error estimates for the NASA surface soil moisture product from the Advanced Microwave Scanning Radiometer for the Earth Observing System (AMSR-E). Such error estimates are not supplied with the AMSR-E data and are therefore a genuine contribution to the AMSR-E product.
H23A-0952
Surface Heat Flux Estimation Using Remotely Sensed Land Temperatures in a Variational Assimilation System With Model Uncertainty
Surface turbulent fluxes are estimated through the assimilation of remotely sensed land surface temperatures. In the current formulation both the observations and the system model are taken to be imperfect and subject to error. The approach is weak constraint variational data assimilation and it is introduced here as an effective way to combine noisy observations and imperfect models. A model error time- series vector is allowed to be an additive term in the surface energy balance equation. The dynamic heat diffusion equation with energy balance is used as the constraint in the variational formulation. The problem cost function includes misfit term that measure the difference between the observed and modeled land surface temperatures, terms that penalize the departure of the estimated parameters from their priors, and the convolution of the model error vector and the constraint. There are two main unknown parameters in the estimation of surface heat fluxes: near surface air turbulent conductivity (scales sum of the two turbulent fluxes) and evaporative fraction (partitions among the two turbulent fluxes). The problem is solved using variational methods. Results from an application to a field experiment site show that addition of unknowns, i.e. model error series, actually improves the estimation of parameters and fluxes. Over the multi-day estimation window either the forcing data or the land surface temperature observations often include bad data values, e.g. spike errors, that can adversely affect the optimization over the window. With the addition of weak constraint and possibility of additive errors the effect of the bad data remain localized and do not carry over as strongly into the window days with good data. As a test of the capability to identify and isolate errors we artificially add error to the incident solar radiation and then show that the model is able to approximate the noise. Such tests provide a convenient way to assess the practicality and feasibility of the presented approach.
H23A-0953
Adaptive soil moisture profile filtering for horizontal information propagation in the independent column-based CLM2.0
Data assimilation aims to provide an optimal estimate of the overall system state, not only for an observed state variable or location. However, large scale land surface models are typically column-based and purely random ensemble perturbation of states will lead to block-diagonal a priori (or background) error covariances. This facilitates the filtering calculations, but compromises the potential of data assimilation to influence (unobserved) vertical and horizontal neighboring state variables. Here, a combination of an ensemble Kalman filter and an adaptive covariance correction method is explored to optimize the variances and retrieve the off-block-diagonal correlations in the a priori error covariance matrix. In a first time period, all available soil moisture profile observations in a small agricultural field are assimilated into the CLM2.0 land surface model to find the adaptive second order a priori error information. After that period, only observations from single individual soil profiles are assimilated with inclusion of this adaptive information. It is shown that assimilation of a single profile can partially rectify the incorrectly simulated soil moisture spatial mean and variability. The largest reduction in the root spatial mean square error in the soil moisture field varies between 7 and 22%, depending on the soil depth, when assimilating a single complete profile every 2 days during 3 months with a single time-invariant covariance correction.
H23A-0954
Assimilation of Shortwave Radiation Measurements into a Downwelling Surface Radiation Model Using an Ensemble Kalman Smoother
Estimation of total downwelling surface radiation is necessary in the study of land surface processes and land-atmosphere exchange due to the inherent influence it has on energy availability. Significant spatial and temporal variability in radiative fluxes exist due to varying atmospheric characteristics, which can in turn influence the variability in land surface states. Clouds are a first-order modulator on downwelling radiation processes as they attenuate solar insolation while emitting longwave radiation; hence, clouds effectively couple downwelling shortwave and longwave radiative fluxes. In this study, we propose a new method for deriving high-resolution estimates of downwelling radiation via merger of several satellite-based products using a data assimilation framework. A parsimonious cloud-coupled model forced by a combination of geostationary and polar orbiting remote sensing products provides a high-resolution a priori ensemble estimate. The prior estimate can then be combined with multi-scale radiation products using the Ensemble Kalman Smoother (EnKS) to yield a high-resolution posterior estimate that is conditioned on other radiation products. Prior to assimilation, Monte Carlo simulations and error characterization studies were performed to assess cross-correlations between model parameters/inputs and to assign uncertainties to those parameters/inputs. Observing system simulation experiments and a real application were conducted to demonstrate feasibility for assimilation of the GEWEX Shortwave Radiation Budget (SRB) product [Pinker et al., 2003]. Preliminary results suggest information content within the daytime SRB measurements can be transferred to the observed daytime shortwave estimates as well as to the unobserved daytime longwave estimates. Additionally, the EnKS enables transfer of information content from the daytime SRB measurements to the unobserved longwave flux estimates during the night. Correlations between shortwave and longwave fluxes are most pronounced when clouds are present due to the process coupling, but are less pronounced during clear-sky conditions.
H23A-0955
On the Diurnal Cycle of Summertime Rainfall and Boundary Layer Conditions in the Great Smoky Mountains – The 2008 IOP at Purchase Knob Research Station
Rainfall varies greatly across the mountains of North Carolina causing widespread flooding and landslides. How the terrain modifies the microphysical and dynamical processes that govern precipitation processes as weather systems approach and pass over the mountains is not yet understood. This is also the case for localized convective storms. During the last two weeks of July in 2008, a field campaign took place at Purchase Knob Research Station (4905') in the central Great Smoky Mountains National Park with the objective of collecting observations to document summertime interactions between landform and storm systems in western North Carolina. We plan to use the observations collected during the intense field observing period (IOP) to evaluate the fidelity of existing cloud-resolving models and to test new parameterizations for the representation of boundary-layer and rainfall processes in mountainous regions. Atmospheric profiles of pressure, air temperature, relative humidity, wind speed and wind direction were collected at half-hourly and three-hourly intervals using both tethersonde and radiosonde systems. Microphysical observations relied on a high-speed camera capable of recording up to 1000 frames by second, which was collocated with a Micro Rain Radar (MRR), a disdrometer, and raingauge. The microphysical observations were integrated into a dense raingauge network designed specifically to capture orographic precipitation gradients in surrounding ridges. Near-surface radiation measurements as well as standard hydrometeorological observations were obtained at two separate locations. Here, results from an integrated analysis of the observations are presented, with emphasis on characterizing the diurnal cycle of rainfall and boundary layer evolution in the Great Smoky Mountains Data from the camera, disdrometer, raingauges and the MRR are used along with a microphysical model to simulate the rainfall events observed during the field study.
H23A-0956
Quantifying the impact of model errors on river discharge retrievals through assimilation of SWOT-observed water surface elevations
The planned NASA/CNES Surface Water and Ocean Topography (SWOT) swath satellite altimetry mission will provide highly accurate measurements of surface water slope (order on microradian over reach lengths 1-10 km) and water surface level (centimetric scale accuracy over areas order of 1 km2). Discharge would be derived through assimilation of slope and/or elevation and other quantities into a hydrodynamic model. Previous studies have demonstrated the potential for such an approach. We describe a system that includes a hydrology (Variable Infiltration Capacity, VIC) and hydrodynamics (LISFLOOD) model that provide the background predictions of river discharge and depth, which are then merged with SWOT observations. Among the key determinants of the accuracy of predictions resulting from such assimilations are the model and observation errors. The former can arise from errors in model forcings (e.g. precipitation and channel boundary inflows) as well as model parameters (e.g. channel width and roughness coefficient). In this study we make an initial attempt at quantifying the impact of model errors on river discharge estimates produced via the assimilation of SWOT observations into LISFLOOD and VIC. The study area is a 1000 km reach of the Ohio River, where synthetic SWOT WSL observations have been generated using the JPL instrument simulator. An ensemble representation is used for model errors in precipitation, hence channel boundary inflows, and roughness. Errors in precipitation are modeled by generating random fields based on the spatial probability distribution of the storm center and extent, and the error probability distribution inferred from the errors between downscaled precipitation fields from a Global Circulation Model and interpolated fields from in-situ stations. Furthermore, errors in channel boundary inflows are deduced from the discrepancies between the spatial resolutions of the hydrologic routing model and the hydrodynamics model. The relative importance of these errors in the estimation of river discharge is assessed, while the sensitivity of the latter to different model error formulations is examined.
H23A-0957 [WITHDRAWN]
New Mexico Scintillometer Network in Support of Remote Sensing and Hydrologic Modeling
In New Mexico, a first-of-its-kind network of seven Large Aperture Scintillometer (LAS) sites was established in 2006 to measure sensible heat fluxes over irrigated fields, riparian areas, deserts, lava flows, and mountain highlands. Wireless networking infrastructure and auxiliary meteorological measurements facilitate real-time data assimilation. LAS measurements are advantageous in that they vastly exceed the footprint size of commonly used ground measurements of sensible and latent heat fluxes (~100 m2), matching the pixel- size of satellite images or grid cells of hydrologic and meteorological models (~0.1-5 km2). Consequently, the LAS measurements can be used to validate, calibrate, and force hydrologic, remote sensing, and weather forecast models. Results are presented for: (1) variability and error of sensible heat flux measurements by scintillometers over heterogeneous terrain and (2) the validation of the Surface Energy Balance Algorithm for Land (SEBAL) applied to MODIS satellite imagery. Findings from this study are discussed in the context of researchers' and practitioners' data assimilation needs.
H23A-0958
Flux Contribution of Large-Scale Coherent Structures within Canopy Sub-layer
The traditional gradient-diffusion theory is problematic for estimating the turbulent transport of scalar quantities (e.g., carbon dioxide and water vapor) across the canopy-atmosphere interface due to the predominant role of large-scale coherent structures in the transport process within the canopy sub-layer (CSL). An accurate estimation of the fraction of the flux transport that the coherent structures are responsible for often suffers from the lack of a quantitative and objective identification method. In this talk, large-eddy simulation (LES) is performed to simulate canopy turbulence, and the large-scale coherent structures are identified through the use of the proper orthogonal decomposition (POD). It is shown that the results are in a good agreement with prior analysis conducted for data collected in numerical, laboratory and field experiments. The three-dimensional geometric features of the coherent structures are shown to be described by a strong sweep motion framed by a vortex pair with elliptical cross-sections inclined at a tile angle of approximately 45 degree in the spanwise direction. Furthermore, above the treetop the vortex pair curves upward in the streamwise direction. We also examine the role of these coherent structures in transporting scalar quantities. Different methods of calculating the inner product of the state vector in the POD are tested to estimate the flux contribution of the coherent structures.
H23A-0959
Canopy Components Temperature Retrieval through Bayesian inversion of Directional measurements
In the calculation of Evapotranspiration the kinematic temperature of the individual canopy components plays a crucial role. Most remote sensing algorithms, like SEBAL and SEBS, use a single surface temperature to calculate the evapotranspiration. These algorithms break down when used for canopies with a heterogeneous kinematic temperature profile. A two-source or four-source approach would result in much more realistic values of the evapotranspiration. Single view Nadir looking sensors are not able to extract the multiple kinetic temperatures with high precision. The use of multi-directional sensors is therefore essential. A bi-angular setup is sufficient to separate soil and canopy temperatures (e.g. Jia et al. 2003). For separation of sunlit and shaded soil or vegetation temperatures measurements at additional angles are needed. Calculation of the component temperatures from measured thermal radiances requires the use of more sophisticated radiative transfer models, because the use of fractional vegetation cover alone is no longer sufficient for an inversion scheme for four components. The radiative transfer model used for the calculation of the component temperatures was the four stream SAIL radiative transfer model (Verhoef et al. 2007). We present the algorithm used and the results obtained for the Bayesian inversion. The results were obtained using several directional measurement configurations. The configurations were chosen such to represent various present and future satellite-borne sensors. In this way the configurations give a clear indication of the possibilities of multi-directional thermal remote sensing. References Jia. L. Li, Z. -I., Menenti, M., Su, Z., Verhoef, W. and Wan, Z. (2003), "A practical algorithm to infer soil and foliage component temperatures from bi-angular ATSR-2 data", International Journal of Remote Sensing, 24:23, 4739-4760. Verhoef, W. Jia, L. Xiao, Q. Su, Z., (2007), "Unified optical-thermal four-stream radiative transfer theory for homogeneous vegetation canopies", IEEE Transactions on Geoscience and Remote Sensing, 45(6:2) 1808-1822.
H23A-0960
Atmospheric Boundary Layer Evening Transitions over West Texas
A systemic analysis of the atmospheric boundary layer behavior during some evening transitions over West Texas was done using the data from an extensive array of instruments which included small and large aperture scintillometers, net radiometers, and meteorological stations. The analysis also comprised some modeling resources in an attempt to take into consideration all the decisive aspects needed to achieve a precise definition of what the evening transition is as a transient process connected with the diurnal cycle as a whole, and also how it is a precondition for the subsequent stable boundary layer realization.
H23A-0961
Sensible Heat Flux Measurements Using a Large Aperture Scintillometer Over Irrigated Cotton
Numerous studies have evaluated the accuracy of sensible heat flux measurements with scintillometers using eddy covariance systems. However, the latter has energy balance closure problems up to 20%. The main objective of this study was to test the Large Aperture Scintillometer (LAS) using weighing lysimeter data. An LAS was deployed across two irrigated lysimeter fields planted to cotton. The orientation of the LAS was selected to have the path of the LAS perpendicular to the predominant wind direction and to avoid direct sunlight on the lenses. The refractive index of air was monitored during the 2008 cropping season at 15-min. intervals, synchronized with weather station and lysimeter measurements. In addition, net radiation and soil heat fluxes were measured in both lysimeter fields. Latent heat fluxes were derived from lysimeter-based evapotranspiration rates. Sensible heat fluxes were derived as a residual from the energy balance equation (H=Rn-G-LE). It is anticipated that LAS data may be useful for validating evapotranspiration maps derived from satellite data as its spatial scale (250-4500 m) is comparable to the spatial resolution of satellite images.