Search Results

You are looking at 1 - 9 of 9 items for

  • Author or Editor: Jeffrey P. Walker x
  • Refine by Access: All Content x
Clear All Modify Search
Gift Dumedah and Jeffrey P. Walker

Abstract

Data assimilation (DA) methods are commonly used for finding a compromise between imperfect observations and uncertain model predictions. The estimation of model states and parameters has been widely recognized, but the convergence of estimated parameters has not been thoroughly investigated. The distribution of model state and parameter values is closely linked to convergence, which in turn impacts the ultimate estimation accuracy of DA methods. This demonstration study examines the robustness and convergence of model parameters for the ensemble Kalman filter (EnKF) and the evolutionary data assimilation (EDA) in the context of the Soil Moisture and Ocean Salinity (SMOS) soil moisture assimilation into the Joint UK Land Environment Simulator in the Yanco area in southeast Australia. The results show high soil moisture estimation accuracy for the EnKF and EDA methods when compared with the open loop estimates during evaluation and validation stages. The level of convergence was quantified for each model parameter in the EDA approach to illustrate its potential in the retrieval of variables that were not directly observed. The EDA was found to have a higher estimation accuracy than the EnKF when its updated members were evaluated against the SMOS level 2 soil moisture. However, the EnKF and EDA estimations are comparable when their forward soil moisture estimates were validated against SMOS soil moisture outside the assimilation time period. This suggests that parameter convergence does not significantly influence soil moisture estimation accuracy for the EnKF. However, the EDA has the advantage of simultaneously determining the convergence of model parameters while providing comparably higher accuracy for soil moisture estimates.

Full access
Jeffrey P. Walker, Garry R. Willgoose, and Jetse D. Kalma

Abstract

The Kalman filter assimilation technique is applied to a simplified soil moisture model for retrieval of the soil moisture profile from near-surface soil moisture measurements. First, the simplified soil moisture model is developed, based on an approximation to the Buckingham–Darcy equation. This model is then used in a 12-month one-dimensional field application, with updating at 1-, 5-, 10-, and 20-day intervals. The data used are for the Nerrigundah field site, New South Wales, Australia. This study has identified (i) the importance of knowing the depth over which the near-surface soil moisture measurements are representative (i.e., observation depth), (ii) soil porosity and residual soil moisture content as the most important soil parameters for correct retrieval of the soil moisture profile, (iii) the importance of a soil moisture model that represents the dominant soil physical processes correctly, and (iv) an appropriate forecasting model as far more important than the temporal resolution of near-surface soil moisture measurements. Although the soil moisture model developed here is a good approximation to the Richards equation, it requires a root water uptake term or calibration to an extreme drying event to model extremely dry periods at the field site correctly.

Full access
S. Allen Macklin, Nicholas A. Bond, and Jeffrey P. Walker

Abstract

During February 1982 a NOAA research aircraft investigated a cold, low-level jet blowing from a gap between mountain ranges on line west side of Cook Inlet, Alaska. The jet blew 200 km southeastward across Cook Inlet between the Kenai Peninsula and the Kodiak archipelago, passing into the Gulf of Alaska where it merged with the large-scale marine wind field. Measurement commenced ∼35 km downstream of the seaward end of the gap. The jet's internal boundary layer accelerated by 5% and grew 20% in depth for ∼50 km; thereafter, wind speed and boundary-layer depth were nearly constant for the next 100 km. The strongest winds (>20 m s−1 at a height of 80 m) were observed on the south side of the jet's thermal axis and 90 km downstream from the coast. Budget analyses reveal that the down-gradient acceleration within the jet was principally opposed by surface friction, and the cold air advection was balanced by a strong upward-directed sensible heat flux from Cook Inlet and entrainment of warmer air from aloft.

Full access
Ashley J. Wright, Jeffrey P. Walker, and Valentijn R. N. Pauwels

Abstract

An increased understanding of the uncertainties present in rainfall time series can lead to improved confidence in both short- and long-term streamflow forecasts. This study presents an analysis that considers errors arising from model input data, model structure, model parameters, and model states with the objective of finding a self-consistent set that includes hydrological models, model parameters, streamflow, remotely sensed (RS) soil moisture (SM), and rainfall. This methodology can be used by hydrologists to aid model and satellite selection. Taking advantage of model input data reduction and model inversion techniques, this study uses a previously developed methodology to estimate areal rainfall time series for the study catchment of Warwick, Australia, for multiple rainfall–runoff models. RS SM observations from the Soil Moisture Ocean Salinity (SMOS) and Advanced Microwave Scanning Radiometer for Earth Observing System (AMSR-E) satellites were assimilated into three different rainfall–runoff models using an ensemble Kalman filter (EnKF). Innovations resulting from the observed and predicted SM were analyzed for Gaussianity. The findings demonstrate that consistency between hydrological models, model parameters, streamflow, RS SM, and rainfall can be found. Joint estimation of rainfall time series and model parameters consistently improved streamflow simulations. For all models rainfall estimates are less than the observed rainfall, and rainfall estimates obtained using the Sacramento Soil Moisture Accounting (SAC-SMA) model are the most consistent with gauge-based observations. The SAC-SMA model simulates streamflow that is most consistent with observations. EnKF innovations obtained when SMOS RS SM observations were assimilated into the SAC-SMA model demonstrate consistency between SM products.

Full access
Rolf H. Reichle, Jeffrey P. Walker, Randal D. Koster, and Paul R. Houser

Abstract

The performance of the extended Kalman filter (EKF) and the ensemble Kalman filter (EnKF) are assessed for soil moisture estimation. In a twin experiment for the southeastern United States synthetic observations of near-surface soil moisture are assimilated once every 3 days, neglecting horizontal error correlations and treating catchments independently. Both filters provide satisfactory estimates of soil moisture. The average actual estimation error in volumetric moisture content of the soil profile is 2.2% for the EKF and 2.2% (or 2.1%; or 2.0%) for the EnKF with 4 (or 10; or 500) ensemble members. Expected error covariances of both filters generally differ from actual estimation errors. Nevertheless, nonlinearities in soil processes are treated adequately by both filters. In the application presented herein the EKF and the EnKF with four ensemble members are equally accurate at comparable computational cost. Because of its flexibility and its performance in this study, the EnKF is a promising approach for soil moisture initialization problems.

Full access
Andreas Colliander, Thomas J. Jackson, Aaron Berg, D. D. Bosch, Todd Caldwell, Steven Chan, Michael H. Cosh, C. Holifield Collins, Jose Martínez-Fernández, Heather McNairn, J. H. Prueger, P. J. Starks, Jeffrey P. Walker, and Simon H. Yueh

Abstract

Soil moisture retrieval is particularly challenging during and immediately after precipitation events because of the transient movement of water in the shallow subsurface. Conventional L-band microwave radiometer–based soil moisture products use algorithms that assume a static state and a constant vertical soil moisture distribution. This study assessed the retrieval performance of a SMAP radiometer-based soil moisture product during and immediately after rain events. The removal of the rain event samples systematically improved the unbiased root-mean-square error (ubRMSE) from 0.037 (all measurements) to 0.028 m3 m−3 (transitory measurements screened out), while the magnitude of the bias became larger (from −0.005 to −0.014 m3 m−3); RMSE improved from 0.047 to 0.042 m3 m−3, and the Pearson correlation saw a minor positive change from 0.813 to 0.824. The results indicate that removing samples during the transitional period causes the comparison to improve, but also suggests that the true bias may be larger than the one estimated using all the samples. Furthermore, the results revealed that the effect was stronger for areas with high clay content. An assessment of the performance of the product during the rain events (overpass within 3 h from the start of the rain) showed that the ubRMSE degraded from the benchmarked 0.036 m3 m−3 (during no rain events at all) to 0.043 m3 m−3 (during rain). The results also showed that the bias became wetter, which is expected because SMAP sensed the water on the surface before propagating to the in situ sensors. SMAP maintains its soil moisture sensitivity even during rain events and screening of rain events may not be necessary to ensure sufficient soil moisture retrieval quality.

Free access
Fan Chen, Wade T. Crow, Michael H. Cosh, Andreas Colliander, Jun Asanuma, Aaron Berg, David D. Bosch, Todd G. Caldwell, Chandra Holifield Collins, Karsten Høgh Jensen, Jose Martínez-Fernández, Heather McNairn, Patrick J. Starks, Zhongbo Su, and Jeffrey P. Walker

Abstract

Despite extensive efforts to maximize ground coverage and improve upscaling functions within core validation sites (CVS) of the NASA Soil Moisture Active Passive (SMAP) mission, spatial averages of point-scale soil moisture observations often fail to accurately capture the true average of the reference pixels. Therefore, some level of pixel-scale sampling error from in situ observations must be considered during the validation of SMAP soil moisture retrievals. Here, uncertainties in the SMAP core site average soil moisture (CSASM) due to spatial sampling errors are examined and their impact on CSASM-based SMAP calibration and validation metrics is discussed. The estimated uncertainty (due to spatial sampling limitations) of mean CSASM over time is found to be large, translating into relatively large sampling uncertainty levels for SMAP retrieval bias when calculated against CSASM. As a result, CSASM-based SMAP bias estimates are statistically insignificant at nearly all SMAP CVS. In addition, observations from temporary networks suggest that these (already large) bias uncertainties may be underestimated due to undersampled spatial variability. The unbiased root-mean-square error (ubRMSE) of CSASM is estimated via two approaches: classical sampling theory and triple collocation, both of which suggest that CSASM ubRMSE is generally within the range of 0.01–0.02 m3 m−3. Although limitations in both methods likely lead to underestimation of ubRMSE, the results suggest that CSASM captures the temporal dynamics of the footprint-scale soil moisture relatively well and is thus a reliable reference for SMAP ubRMSE calculations. Therefore, spatial sampling errors are revealed to have very different impacts on efforts to estimate SMAP bias and ubRMSE metrics using CVS data.

Full access
Li Fang, Xiwu Zhan, Jifu Yin, Jicheng Liu, Mitchell Schull, Jeffrey P. Walker, Jun Wen, Michael H. Cosh, Tarendra Lakhankar, Chandra Holifield Collins, David D. Bosch, and Patrick J. Starks

Abstract

In the past decade, a variety of algorithms have been introduced to downscale passive microwave soil moisture observations. Some exploit the soil moisture information from optical/thermal sensing of land surface temperature (LST) and vegetation dynamics while others use active microwave (radar) observations. In this study, downscaled soil moisture data at 9- or 1-km resolution from several algorithms are intercompared against in situ soil moisture measurements to determine their reliability in an operational system. The finescale satellite data used here for downscaling the coarse-scale SMAP data are observations of LST from the Geostationary Operational Environmental Satellite (GOES) and vegetation index (VI) from the NASA Moderate Resolution Imaging Spectroradiometer (MODIS) for the warm seasons in 2015 and 2016. Three recently developed downscaling algorithms are evaluated and compared: a simple regression algorithm based on 9-km thermal inertial data, a data mining approach called regression tree based on 9- and 1-km LST and VI, and the NASA SMAP enhanced 9-km soil moisture product algorithm. Seven sets of in situ soil moisture data from intensive networks were used for validation, including 1) the CREST-SMART network in Millbrook, New York; 2) Walnut Gulch Watershed in Arizona; 3) Little Washita Watershed in Oklahoma; 4) Fort Cobb Reservoir Experimental Watersheds in Oklahoma; 5) Little River Watershed in Georgia; 6) the Tibetan Plateau network in China, and 7) the OzNet in Australia. Soil moisture measurements of the in situ networks were upscaled to the corresponding SMAP reference pixels at 9 km and used to assess the accuracy of downscaled products at a 9-km scale. Results revealed that the downscaled 9-km soil moisture products generally outperform the 36-km product for most in situ datasets. The linear regression algorithm using the thermal sensing based evaporative stress index (ESI) had the best agreement with the in situ measurements from networks in the contiguous United States according to the site-by-site comparison. In addition, the inertial thermal linear regression method demonstrated the lowest unbiased RMSE when comparing to the matched-up in situ datasets as well. In general, this method is promising for operational generation of fine-resolution soil moisture data product.

Open access
Rolf H. Reichle, Gabrielle J. M. De Lannoy, Qing Liu, Joseph V. Ardizzone, Andreas Colliander, Austin Conaty, Wade Crow, Thomas J. Jackson, Lucas A. Jones, John S. Kimball, Randal D. Koster, Sarith P. Mahanama, Edmond B. Smith, Aaron Berg, Simone Bircher, David Bosch, Todd G. Caldwell, Michael Cosh, Ángel González-Zamora, Chandra D. Holifield Collins, Karsten H. Jensen, Stan Livingston, Ernesto Lopez-Baeza, José Martínez-Fernández, Heather McNairn, Mahta Moghaddam, Anna Pacheco, Thierry Pellarin, John Prueger, Tracy Rowlandson, Mark Seyfried, Patrick Starks, Zhongbo Su, Marc Thibeault, Rogier van der Velde, Jeffrey Walker, Xiaoling Wu, and Yijian Zeng

Abstract

The Soil Moisture Active Passive (SMAP) mission Level-4 Surface and Root-Zone Soil Moisture (L4_SM) data product is generated by assimilating SMAP L-band brightness temperature observations into the NASA Catchment land surface model. The L4_SM product is available from 31 March 2015 to present (within 3 days from real time) and provides 3-hourly, global, 9-km resolution estimates of surface (0–5 cm) and root-zone (0–100 cm) soil moisture and land surface conditions. This study presents an overview of the L4_SM algorithm, validation approach, and product assessment versus in situ measurements. Core validation sites provide spatially averaged surface (root zone) soil moisture measurements for 43 (17) “reference pixels” at 9- and 36-km gridcell scales located in 17 (7) distinct watersheds. Sparse networks provide point-scale measurements of surface (root zone) soil moisture at 406 (311) locations. Core validation site results indicate that the L4_SM product meets its soil moisture accuracy requirement, specified as an unbiased RMSE (ubRMSE, or standard deviation of the error) of 0.04 m3 m−3 or better. The ubRMSE for L4_SM surface (root zone) soil moisture is 0.038 m3 m−3 (0.030 m3 m−3) at the 9-km scale and 0.035 m3 m−3 (0.026 m3 m−3) at the 36-km scale. The L4_SM estimates improve (significantly at the 5% level for surface soil moisture) over model-only estimates, which do not benefit from the assimilation of SMAP brightness temperature observations and have a 9-km surface (root zone) ubRMSE of 0.042 m3 m−3 (0.032 m3 m−3). Time series correlations exhibit similar relative performance. The sparse network results corroborate these findings over a greater variety of climate and land cover conditions.

Full access