• Anderson, M. C., , Norman J. M. , , Diak G. R. , , Kustas W. P. , , and Mecikalski J. R. , 1997: A two-source time-integrated model for estimating surface fluxes using thermal infrared remote sensing. Remote Sens. Environ., 60, 195216, doi:10.1016/S0034-4257(96)00215-5.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., , Neale C. M. U. , , Li F. , , Norman J. M. , , Kustas W. P. , , Jayanthi H. , , and Chavez J. , 2004a: Upscaling ground observations of vegetation water content, canopy height, and leaf area index during SMEX02 using aircraft and Landsat imagery. Remote Sens. Environ., 92, 447464, doi:10.1016/j.rse.2004.03.019.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., , Norman J. M. , , Mecikalski J. R. , , Torn R. D. , , Kustas W. P. , , and Basara J. B. , 2004b: A multi-scale remote sensing model for disaggregating regional flues to micrometeorological scales. J. Hydrometeor., 5, 343363, doi:10.1175/1525-7541(2004)005<0343:AMRSMF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., , Norman J. M. , , Kustas W. P. , , Li F. , , Prueger J. H. , , and Mecikalski J. R. , 2005: Effects of vegetation clumping on two-source model estimates of surface energy fluxes from an agricultural landscape using SMACEX. J. Hydrometeor., 6, 892909, doi:10.1175/JHM465.1.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., , Kustas W. P. , , and Norman J. M. , 2007a: Upscaling tower and aircraft fluxes from local to continental scales using thermal remote sensing. Agron. J., 99, 240254, doi:10.2134/agronj2005.0096S.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., , Norman J. M. , , Kustas W. P. , , Li F. , , Prueger J. H. , , and Mecikalski J. R. , 2007b: A climatological study of evapotranspiration and moisture stress across the continental United States: 1. Model formulation. J. Geophys. Res., 112, D11112, doi:10.1029/2006JD007507.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., and Coauthors, 2012: Mapping daily evapotranspiration at Landsat spatial scales during the BEAREX’08 field campaign. Adv. Water Resour., 50, 162177, doi:10.1016/j.advwatres.2012.06.005.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., , Hain C. , , Otkin J. , , Zhan X. , , Mo K. , , Svoboda M. , , Wardlow B. , , and Pimstein A. , 2013a: An intercomparison of drought indicators based on thermal remote sensing and NLDAS-2 simulations with U.S. Drought Monitor classifications. J. Hydrometeor., 14, 10351056, doi:10.1175/JHM-D-12-0140.1.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., , Kustas W. P. , , and Hain C. R. , 2013b: Mapping surface fluxes and moisture conditions from field to global scales using ALEXI/DisALEXI. Remote Sensing of Energy Fluxes and Soil Moisture Content, G. P. Petropoulos, Ed., CRC Press, 207–232, doi:10.1201/b15610-11.

  • Cammalleri, C., , Anderson M. C. , , Ciraolo G. , , D'Urso G. , , Kustas W. P. , , La Loggia G. , , and Minacapilli M. , 2012: Applications of a remote sensing–based two-source energy balance algorithm for mapping surface fluxes without in situ air temperature observations. Remote Sens. Environ., 124, 502515, doi:10.1016/j.rse.2012.06.009.

    • Search Google Scholar
    • Export Citation
  • Cammalleri, C., , Anderson M. C. , , Gao F. , , Hain C. R. , , and Kustas W. P. , 2013: A data fusion approach for mapping daily evapotranspiration at field scale. Water Resour. Res., 49, 46724686, doi:10.1002/wrcr.20349.

    • Search Google Scholar
    • Export Citation
  • Cammalleri, C., , Anderson M. C. , , Gao F. , , Hain C. R. , , and Kustas W. P. , 2014: Mapping daily Evapotranspiration at field scales over rainfed and irrigated agricultural areas using remote sensing data fusion. Agric. For. Meteor., 186, 111, doi:10.1016/j.agrformet.2013.11.001.

    • Search Google Scholar
    • Export Citation
  • Chen, F., , and Dudhia J. , 2001: Coupling an advanced land surface hydrology model with the Penn State/NCAR MM5 modeling system. Part 1: Model description and implementation. Mon. Wea. Rev., 129, 569586, doi:10.1175/1520-0493(2001)129<0569:CAALSH>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Chen, F., and Coauthors, 1996: Modeling of land-surface evaporation by four schemes and comparison with FIFE observations. J. Geophys. Res., 101, 72517268, doi:10.1029/95JD02165.

    • Search Google Scholar
    • Export Citation
  • Cosgrove, B. A., and Coauthors, 2003: Real-time and retrospective forcing in the North American Land Data Assimilation System (NLDAS) project. J. Geophys. Res., 108, 8842, doi:10.1029/2002JD003118.

    • Search Google Scholar
    • Export Citation
  • Dickinson, R. E., , Henderson-Sellers A. , , and Kennedy P. J. , 1993: Biosphere–Atmosphere Transfer Scheme (BATS) Version 1e as coupled to the NCAR Community Climate Model. NCAR Tech. Note NCAR-TN-387+STR, 88 pp., doi:10.5065/D67W6959.

  • Dirmeyer, P. A., and Coauthors, 2012: Evidence for enhanced land–atmosphere feedback in a warming climate. J. Hydrometeor., 13, 981995, doi:10.1175/JHM-D-11-0104.1.

    • Search Google Scholar
    • Export Citation
  • Ek, M. B., , Mitchell K. E. , , Lin Y. , , Rogers E. , , Grummann P. , , Koren V. , , Gayno G. , , and Tarpley J. D. , 2003: Implementation of Noah land surface model advances in the National Centers for Environmental Prediction operational mesoscale Eta model. J. Geophys. Res., 108, 8851, doi:10.1029/2002JD003296.

    • Search Google Scholar
    • Export Citation
  • Fan, Y., , and Miguez-Macho G. , 2011: A simple hydrologic framework for simulating wetlands in climate and Earth system models. Climate Dyn., 37, 253278, doi:10.1007/s00382-010-0829-8.

    • Search Google Scholar
    • Export Citation
  • Fan, Y., , Miguez-Macho G. , , Weaver C. P. , , Walko R. , , and Robock A. , 2007: Incorporating water table dynamics in climate modeling: 1. Water table observations and equilibrium water table simulations. J. Geophys. Res., 112, D10125, doi:10.1029/2006JD008111.

    • Search Google Scholar
    • Export Citation
  • Fan, Y., , Li H. , , and Miguez-Macho G. , 2013: Global patterns of groundwater table depth. Science, 339, 940943, doi:10.1126/science.1229881.

    • Search Google Scholar
    • Export Citation
  • Fry, J., and Coauthors, 2011: Completion of the 2006 National Land Cover Database for the conterminous United States. Photogramm. Eng. Remote Sens.,77, 858–864.

  • Gedney, N., , and Cox P. M. , 2003: The sensitivity of global climate model simulations to the representation of soil moisture heterogeneity. J. Hydrometeor., 4, 12651275, doi:10.1175/1525-7541(2003)004<1265:TSOGCM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Gedney, N., , Cox P. M. , , Betts R. A. , , Boucher O. , , Huntingford C. , , and Stott P. A. , 2006: Detection of a direct carbon dioxide effect in continental river runoff records. Nature, 439, 835–838, doi:10.1038/nature04504.

    • Search Google Scholar
    • Export Citation
  • Haddeland, I., , Skaugen T. , , and Lettenmaier D. P. , 2006: Anthropogenic impacts on continental surface water fluxes. Geophys. Res. Lett.,33, L08406, doi:10.1029/2006GL026047.

  • Hain, C. R., , Mecikalski J. R. , , and Anderson M. C. , 2009: Retrieval of an available water-based soil moisture proxy from thermal infrared remote sensing. Part I: Methodology and validation. J. Hydrometeor., 10, 665683, doi:10.1175/2008JHM1024.1.

    • Search Google Scholar
    • Export Citation
  • Hain, C. R., , Crow W. T. , , Mecikalski J. R. , , Anderson M. C. , , and Holmes T. , 2011: An intercomparison of available soil moisture estimates from thermal infrared and passive microwave remote sensing and land surface modeling. J. Geophys. Res., 116, D15107, doi:10.1029/2011JD015633.

    • Search Google Scholar
    • Export Citation
  • Hansen, M. C., , DeFries R. S. , , Townshend J. R. G. , , and Sohlberg R. , 2000: Global land cover classification at 1 km spatial resolution using a classification tree approach. Int. J. Remote Sens., 21, 13311364, doi:10.1080/014311600210209.

    • Search Google Scholar
    • Export Citation
  • Harding, K. J., , and Snyder P. K. , 2012: Modeling the atmospheric response to irrigation in the Great Plains. Part II: The precipitation or irrigated water and changes in precipitation recycling. J. Hydrometeor., 13, 16871703, doi:10.1175/JHM-D-11-099.1.

    • Search Google Scholar
    • Export Citation
  • Jedlovec, G. J., , Haines S. L. , , and LaFontaine F. J. , 2008: Spatial and temporal varying thresholds for cloud detection in GOES imager. IEEE Trans. Geosci. Remote Sens., 46, 17051717, doi:10.1109/TGRS.2008.916208.

    • Search Google Scholar
    • Export Citation
  • Jung, M., and Coauthors, 2010: Recent decline in the global land evapotranspiration trend due to limited moisture supply. Nature, 467, 951954, doi:10.1038/nature09396.

    • Search Google Scholar
    • Export Citation
  • Koch, S., , Bauwe A. , , and Lennartz B. , 2013: Application of the SWAT model for a tile-drained lowland catchment in north-eastern Germany on subbasin scale. Water Resour. Manage., 27, 791805, doi:10.1007/s11269-012-0215-x.

    • Search Google Scholar
    • Export Citation
  • Kumar, S. V., and Coauthors, 2006: Land information system: An interoperable framework for high resolution land surface modeling. Environ. Modell. Software, 21, 14021415, doi:10.1016/j.envsoft.2005.07.004.

    • Search Google Scholar
    • Export Citation
  • Liang, X., , Xie Z. , , and Huang M. , 2003: A new parameterization for surface and groundwater interactions and its impact on water budgets with the variable infiltration capacity (VIC) land surface model. J. Geophys. Res., 108, 8613, doi:10.1029/2002JD003090.

    • Search Google Scholar
    • Export Citation
  • Long, D., , Longuevergne L. , , and Scanlon B. , 2014: Uncertainty in evapotranspiration from land surface modeling, remote sensing and GRACE satellites. Water Resour. Res., 50, 11311151, doi:10.1002/2013WR014581.

    • Search Google Scholar
    • Export Citation
  • Maxwell, R. M., , and Miller N. L. , 2005: On the development of a coupled land surface and groundwater model. J. Hydrometeor., 6, 233247, doi:10.1175/JHM422.1.

    • Search Google Scholar
    • Export Citation
  • McNaughton, K. J., , and Spriggs T. W. , 1986: A mixed-layer model for regional evaporation. Bound.-Layer Meteor., 34, 243262, doi:10.1007/BF00122381.

    • Search Google Scholar
    • Export Citation
  • Miguez-Macho, G., , Fan Y. , , Weaver C. P. , , Walko R. , , and Robock A. , 2007: Incorporating water table dynamics in climate modeling: 2. Formulation, validation, and soil moisture simulation. J. Geophys. Res., 112, D13108, doi:10.1029/2006JD008112.

    • Search Google Scholar
    • Export Citation
  • Miguez-Macho, G., , Li H. , , and Fan Y. , 2008: Simulated water table and soil moisture climatology over North America. Bull. Amer. Meteor. Soc., 89, 663672, doi:10.1175/BAMS-89-5-663.

    • Search Google Scholar
    • Export Citation
  • Moody, E. G., , King M. D. , , Platnick S. , , Schaaf C. B. , , and Gao F. , 2005: Spatially complete global spectral surface albedos: Value-added datasets derived from Terra MODIS land products. IEEE Trans. Geosci. Remote Sens., 43, 144158, doi:10.1109/TGRS.2004.838359.

    • Search Google Scholar
    • Export Citation
  • Mu, Q., , Heinsch F. A. , , Zhao M. , , and Running S. W. , 2007: Development of a global evapotranspiration algorithm based on MODIS and global meteorology data. Remote Sens. Environ., 111, 519536, doi:10.1016/j.rse.2007.04.015.

    • Search Google Scholar
    • Export Citation
  • Mu, Q., , Zhao M. , , Kimball J. S. , , McDowell N. G. , , and Running S. W. , 2011: Improvements to a MODIS global terrestrial evapotranspiration algorithm. Remote Sens. Environ., 115, 17811800, doi:10.1016/j.rse.2011.02.019.

    • Search Google Scholar
    • Export Citation
  • Myneni, R. B., and Coauthors, 2002: Global products of vegetation leaf area and fraction absorbed PAR from year one of MODIS data. Remote Sens. Environ., 83, 214231, doi:10.1016/S0034-4257(02)00074-3.

    • Search Google Scholar
    • Export Citation
  • 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.

    • Search Google Scholar
    • Export Citation
  • Niu, G.-Y., and Coauthors, 2011: The community Noah land surface model with multiparameterization options (Noah-MP): 1. Model description and evaluation with local-scale measurements. J. Geophys. Res., 116, D12109, doi:10.1029/2010JD015139.

    • Search Google Scholar
    • Export Citation
  • Norman, J. M., , Kustas W. P. , , and Humes K. S. , 1995: A two-source approach for estimating soil and vegetation energy fluxes from observations of directional radiometric surface temperatures. Agric. For. Meteor., 77, 263293, doi:10.1016/0168-1923(95)02265-Y.

    • Search Google Scholar
    • Export Citation
  • Ozdogan, M., , and Gutman G. , 2008: A new methodology to map irrigated areas using multi-temporal MODIS and ancillary data: An application example in the continental US. Remote Sens. Environ., 112, 35203537, doi:10.1016/j.rse.2008.04.010.

    • Search Google Scholar
    • Export Citation
  • Ozdogan, M., , Rodell M. , , Beaudoing H. K. , , and Toll D. L. , 2010: Simulating the effects of irrigation over the United States in a land surface model based on satellite-derived agricultural data. J. Hydrometeor., 11, 171184, doi:10.1175/2009JHM1116.1.

    • Search Google Scholar
    • Export Citation
  • Pervez, M. S., , and Brown J. F. , 2010: Mapping irrigated lands at 250-m scale by merging MODIS data and national agricultural statistics. Remote Sens., 2, 23882412, doi:10.3390/rs2102388.

    • Search Google Scholar
    • Export Citation
  • Peters-Lidard, C. D., , Kumar S. V. , , Mocko D. M. , , and Tian Y. , 2011: Estimating evapotranspiration with land data assimilation systems. Hydrol. Processes, 25, 39793992, doi:10.1002/hyp.8387.

    • Search Google Scholar
    • Export Citation
  • Price, J. C., 1983: Estimating surface temperatures from satellite thermal infrared data—A simple formulation for the atmospheric effect. Remote Sens. Environ., 13, 353361, doi:10.1016/0034-4257(83)90036-6.

    • Search Google Scholar
    • Export Citation
  • Rover, J., , Wright C. K. , , Euliss N. H. , , Mushet D. M. , , and Wylie B. K. , 2011: Classifying the hydrologic function of prairie potholes with remote sensing and GIS. Wetlands, 31, 319327, doi:10.1007/s13157-011-0146-y.

    • Search Google Scholar
    • Export Citation
  • Sutanudjaja, E. H., , van Beek L. P. H. , , de Jong S. M. , , van Geer F. C. , , and Bierkens M. F. P. , 2014: Calibrating a large-extent high-resolution coupled groundwater–land surface model using soil moisture and discharge data. Water Resour. Res., 50, 687–705, doi:10.1002/2013WR013807.

    • Search Google Scholar
    • Export Citation
  • Wei, H., , Xia Y. , , Mitchell K. , , and Ek M. , 2013: Improvement of the Noah land surface model for warm season processes: Evaluation of water and energy flux simulation. Hydrol. Processes, 27, 297303, doi:10.1002/hyp.9214.

    • Search Google Scholar
    • Export Citation
  • Xia, Y., and Coauthors, 2012a: Continental-scale water and energy flux analysis and validation for the North American Land Data Assimilation System project phase 2 (NLDAS-2): 1. Intercomparison and application of model products. J. Geophys. Res., 117, D03109, doi:10.1029/2011JD016048.

    • Search Google Scholar
    • Export Citation
  • Xia, Y., and Coauthors, 2012b: Continental-scale water and energy flux analysis and validation for North American Land Data Assimilation System project phase 2 (NLDAS-2): 2. Validation of model-simulated streamflow. J. Geophys. Res., 117, D03110, doi:10.1029/2011JD016051.

    • Search Google Scholar
    • Export Citation
  • Yang, Z.-L., and Coauthors, 2011: The community Noah land surface model with multiparameterization options (Noah-MP): 2. Evaluation over global river basins. J. Geophys. Res., 116, D12110, doi:10.1029/2010JD015140.

    • Search Google Scholar
    • Export Citation
  • Yeh, P. J. F., , and Eltahir E. A. B. , 2005: Representation of water table dynamics in a land surface scheme. Part I: Model development. J. Climate, 18, 18611880, doi:10.1175/JCLI3330.1.

    • Search Google Scholar
    • Export Citation
  • Yilmaz, M. T., , Anderson M. C. , , Zaitchik B. , , Hain C. R. , , Crow W. T. , , Ozdogan M. , , Chun J. A. , , and Evans J. , 2014: Comparison of prognostic and diagnostic surface flux modeling approaches over the Nile River basin. Water Resour. Res., 50, 386408, doi:10.1002/2013WR014194.

    • Search Google Scholar
    • Export Citation
  • York, J. P., , Person M. , , Gutowski W. J. , , and Winter T. C. , 2002: Putting aquifers into atmospheric simulation models: An example from the Mill Creek watershed, northeastern Kansas. Adv. Water Resour., 25, 221238, doi:10.1016/S0309-1708(01)00021-5.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    CONUS (a) LEALEXI (W m−2), (b) LENoah (W m−2), (c) VARALEXI (W m−2)2, and (d) VARNoah (W m−2)2 maps computed for clear-sky JJA conditions between 2000 and 2012. The color scale for (a),(b) runs from 0 to 500 in increments of 50 and for (c),(d) runs from 0 to 5000 in increments of 500.

  • View in gallery

    (a) ASSET computed from an average of 13 JJA composites [2000–12; positive (negative) values indicate regions where ALEXI clear-sky LE was greater (less) than Noah clear-sky LE, collocated with low annual variability of ALEXI clear-sky LE], (b) MODIS irrigation percentage (Ozdogan and Gutman 2008), and (c) simulated water-table depth (m; Fan et al. 2007; Miguez-Macho et al. 2008). The roman numeral–labeled red rectangles are referenced in subsequent figures.

  • View in gallery

    Composite of ASSET and irrigation and groundwater proxy datasets over the western CONUS region [i.e., domain (i) in Fig. 2a].

  • View in gallery

    Composite of ASSET and irrigation and groundwater proxy datasets over the south-central CONUS region [i.e., domain (ii) in Fig. 2a].

  • View in gallery

    (a) ASSET (color scale is from −3000 to +3000 in increments of 600), (b) MODIS irrigation percentage (%; color scale is from 0% to 100% in increments of 10%), (c) simulated water-table depth (m; color scale is from 0 to 2 m in increments of 0.25 m), and (d) NLCD wetland percentage (%; color scale is from 0% to 100% in increments of 10%) maps for the southeast CONUS region [i.e., domain (iv) in Fig. 2a].

  • View in gallery

    Composite of ASSET and irrigation and groundwater proxy datasets over the southeast CONUS region [i.e., domain (iv) in Fig. 2a].

  • View in gallery

    As in Fig. 5, but for the north-central CONUS region [i.e., domain (iii) in Fig. 2a].

  • View in gallery

    Composite of ASSET and irrigation and groundwater proxy datasets over the north-central CONUS region [i.e., domain (iii) in Fig. 2a].

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 4 4 4
PDF Downloads 0 0 0

Diagnosing Neglected Soil Moisture Source–Sink Processes via a Thermal Infrared–Based Two-Source Energy Balance Model

View More View Less
  • 1 Earth System Science Interdisciplinary Center, University of Maryland College Park, College Park, Maryland
  • 2 Hydrology and Remote Sensing Laboratory, Agricultural Research Service, USDA, Beltsville, Maryland
  • 3 Water Resources Division, Civil Engineering Department, Middle East Technical University, Ankara, Turkey
© Get Permissions
Full access

Abstract

In recent years, increased attention has been paid to the role of previously neglected water source (e.g., irrigation, direct groundwater extraction, and inland water bodies) and sink (e.g., tile drainage) processes on the surface energy balance. However, efforts to parameterize these processes within land surface models (LSMs) have generally been hampered by a lack of appropriate observational tools for directly observing the impact(s) of such processes on surface energy fluxes. One potential strategy for quantifying these impacts are direct comparisons between bottom-up surface energy flux predictions from a one-dimensional, free-drainage LSM with top-down energy flux estimates derived via thermal infrared remote sensing. The neglect of water source (and/or sink) processes in the bottom-up LSM can be potentially diagnosed through the presence of systematic energy flux biases relative to the top-down remote sensing retrieval. Based on this concept, the authors introduce the Atmosphere–Land Exchange Inverse (ALEXI) Source–Sink for Evapotranspiration (ASSET) index derived from comparisons between ALEXI remote sensing latent heat flux retrievals and comparable estimates obtained from the Noah LSM, version 3.2. Comparisons between ASSET index values and known spatial variations of groundwater depth, irrigation extent, inland water bodies, and tile drainage density within the contiguous United States verify the ability of ASSET to identify regions where neglected soil water source–sink processes may be impacting modeled surface energy fluxes. Consequently, ASSET appears to provide valuable information for ongoing efforts to improve the parameterization of new water source–sink processes within modern LSMs.

Corresponding author address: Christopher R. Hain, Earth System Science Interdisciplinary Center, University of Maryland, College Park, 5825 University Research Ct., Suite 4001, College Park, MD 20740. E-mail: chris.hain@noaa.gov

Abstract

In recent years, increased attention has been paid to the role of previously neglected water source (e.g., irrigation, direct groundwater extraction, and inland water bodies) and sink (e.g., tile drainage) processes on the surface energy balance. However, efforts to parameterize these processes within land surface models (LSMs) have generally been hampered by a lack of appropriate observational tools for directly observing the impact(s) of such processes on surface energy fluxes. One potential strategy for quantifying these impacts are direct comparisons between bottom-up surface energy flux predictions from a one-dimensional, free-drainage LSM with top-down energy flux estimates derived via thermal infrared remote sensing. The neglect of water source (and/or sink) processes in the bottom-up LSM can be potentially diagnosed through the presence of systematic energy flux biases relative to the top-down remote sensing retrieval. Based on this concept, the authors introduce the Atmosphere–Land Exchange Inverse (ALEXI) Source–Sink for Evapotranspiration (ASSET) index derived from comparisons between ALEXI remote sensing latent heat flux retrievals and comparable estimates obtained from the Noah LSM, version 3.2. Comparisons between ASSET index values and known spatial variations of groundwater depth, irrigation extent, inland water bodies, and tile drainage density within the contiguous United States verify the ability of ASSET to identify regions where neglected soil water source–sink processes may be impacting modeled surface energy fluxes. Consequently, ASSET appears to provide valuable information for ongoing efforts to improve the parameterization of new water source–sink processes within modern LSMs.

Corresponding author address: Christopher R. Hain, Earth System Science Interdisciplinary Center, University of Maryland, College Park, 5825 University Research Ct., Suite 4001, College Park, MD 20740. E-mail: chris.hain@noaa.gov

1. Introduction

Traditional soil water balance modeling is based on one-dimensional (vertical only) water flow, free drainage at the bottom of the soil column, and neglecting ancillary water inputs due to processes such as irrigation (Dickinson et al. 1993). As a consequence, the vertical infiltration of local precipitation represents the only source of soil water available for surface evapotranspiration ET. However, recent work has also highlighted the importance of secondary water source (e.g., irrigation, groundwater extraction, inland wetlands, and the lateral redistribution of water by topography) and sink (e.g., tile drainage in agricultural areas) processes on the partitioning of evaporative and sensible heat fluxes at the land surface (Yilmaz et al. 2014). While attempts have been made to incorporate irrigation (Gedney et al. 2006; Haddeland et al. 2006; Ozdogan et al. 2010; Harding and Snyder 2012), tile drainage (Koch et al. 2013), and groundwater (York et al. 2002; Liang et al. 2003; Gedney and Cox 2003; Maxwell and Miller 2005; Yeh and Eltahir 2005; Fan et al. 2007, 2013; Miguez-Macho et al. 2007; Niu et al. 2007; Fan and Miguez-Macho 2011) processes into LSMs, these efforts generally require parameters that are difficult to measure (e.g., the volume of water applied in irrigation or groundwater recharge rates) and modification of LSM outputs in ways that are challenging to validate. As a result, new strategies are required to parameterize and/or evaluate LSM representation of these processes (Sutanudjaja et al. 2014).

In response to this need, recent work has focused on the development of products that map the spatial extent of irrigation and climatological groundwater conditions. For example, Ozdogan and Gutman (2008) used MODIS imagery, along with an algorithm to estimate “irrigation potential” based on precipitation and temperature climatologies, to delineate irrigated pixels at 500-m resolution over the contiguous United States (CONUS). Likewise, Pervez and Brown (2010) merged 250-m MODIS imagery and irrigation statistics from the National Agricultural Statistics Service (NASS) to map irrigated areas over the CONUS. Satellite-based observations of water-table depth are not as readily available, but mapping methods have been developed that use a combination of limited ground-based observations and hydrological modeling systems. For instance, Fan et al. (2007) and Miguez-Macho et al. (2008) estimated climatological mean water-table depth by using a two-dimensional groundwater model to calculate the long-term hydrologic balance between recharge, the lateral movement of groundwater, and drainage.

While these datasets can be used as proxy information for describing the impact of nonprecipitation-based water sources on the surface energy balance, they offer only an indirect bottom-up assessment. For example, while satellite-based irrigation maps can be useful in determining the spatial extent of irrigated agricultural lands, accurate estimates of actual water added during typical irrigation events are largely unavailable. Likewise, estimates of climatological water-table heights, while useful, do not directly describe the real-time impact of groundwater on the surface energy balance, nor do they adequately capture the impact of extensive subsurface tile drainage in agricultural landscapes. In addition to providing only a quantitative assessment, the proxy maps (of, e.g., irrigation extent, climatological water-table heights, and surface water extent) have a large number of significant shortcomings. First, the proxy data are not truly “observations” but rather estimates with their own unique set of spatial uncertainties. In addition, even if the proxy data provide a perfect static spatial representation, it still 1) is lacking in seasonal dynamics, 2) is largely limited to only data-rich portions of the globe, 3) is only indirectly related to surface energy fluxes, and 4) fails to assess the aggregate impact of all processes on the surface energy balance. Therefore, new independent tools are required to address these shortcomings.

Surface energy balance models based on thermal infrared remote sensing offer a top-down opportunity to directly observe the impact of nonprecipitation water sources and anthropogenic water sinks on the land surface energy balance (Yilmaz et al. 2014). For example, the Atmosphere–Land Exchange Inverse (ALEXI) model uses time-differential measurements of morning land surface temperature LST rise to diagnose the partitioning of available energy into sensible, latent, and ground heat flux components (Anderson et al. 1997, 2007b). In contrast to prognostic LSMs, ALEXI does not employ a water balance model to predict ET and soil water availability. Instead, water availability and its subsequent impact on surface energy fluxes is diagnosed directly from the observed pre-noon rise in LST. As a result, the model requires no a priori parameterization of water source and/or sink processes. Additionally, ALEXI has a limited reliance on ground-based meteorological forcing, needing only an estimate of the early morning lapse rate and wind speed. ALEXI has been extensively tested and validated against a number of available tower-based flux observation sites over a wide range of climatic and vegetation types, including rainfed and irrigated fields (Anderson et al. 2004a,b, 2005, 2007a, 2012, 2013b; Cammalleri et al. 2012, 2013, 2014).

Several studies have demonstrated that LST conveys valuable and spatially detailed information regarding the surface moisture status and its modulating effect on evaporative fluxes (Anderson et al. 2007b; Hain et al. 2009, 2011). Methods for retrieving ET using remote sensing that do not employ LST are, in many cases, less effective in determining reductions in ET due to moisture stress (Anderson et al. 2012). For example, the MODIS ET (MOD16) dataset developed by Mu et al. (2007, 2011) does not use LST as an input and instead relies on a model analysis product of surface vapor pressure deficit VPD as a proxy for soil moisture stress. Unless ancillary moisture sources such as groundwater and irrigation are represented a priori in the land surface model component of the analysis system generating the VPD input fields, stress constraints on localized fluxes associated with such features may not be reasonably captured by this approach and nonequilibrium effects (e.g., horizontal dry air advection) will be neglected (Yilmaz et al. 2014).

As a result, LST-based methods such as ALEXI provide a unique opportunity to augment the development of LSMs by providing an independent, top-down method for directly diagnosing the impact of nonprecipitation sources (and anthropogenic sinks like tile drainage) on surface energy fluxes. In this study, we develop and evaluate an index for mapping land surface areas where soil water source–sink processes may have a significant impact on the surface energy balance. The index is based on comparisons between ALEXI-derived diagnostic latent heat flux LE estimates and LE predicted by a free-drainage prognostic LSM lacking irrigation, groundwater, and tile drainage modules. Sections 2 and 3 describe the basis of the index, and section 4 describes the application of the index over CONUS. In particular, the index is compared to existing maps of water table, open water fraction, and irrigated area to provide a preliminary evaluation of the index as a diagnostic tool for evaluating the impact of soil water source–sink processes not captured by traditional one-dimensional water balance models. The index also captures regions where unattributed biases between modeling approaches persist, highlighting areas that require further investigation. As such, the proposed index is of potential value for ongoing efforts to improve and evaluate the parameterization of groundwater and/or anthropogenic source–sink processes in LSMs (Niu et al. 2011; Yang et al. 2011).

2. Model methodology

As described above, our approach is based on the comparison of top-down LE derived from a diagnostic remote sensing approach (ALEXI) with bottom-up LE results obtained from a prognostic LSM (employing a soil water balance approach). These two contrasting LE modeling strategies are described below.

a. ALEXI

ALEXI was formulated as an extension to the two-source energy balance (TSEB) model of Norman et al. (1995), which in turn was developed to address many of the documented challenges in monitoring surface energy fluxes using thermal remote sensing data. The two-source approximation treats the radiometric temperature TRAD of a vegetated surface (i.e., the LST) as the weighted average of the individual temperatures of soil Ts and vegetation Tc subcomponents, partitioned by the fractional vegetation cover fc(φ) apparent from the thermal sensor view angle φ. The TSEB separately balances energy budgets for the soil and vegetation components of the system and also solves for total system fluxes of net radiation RN, LE, sensible heat H, and soil heat conduction G. In ALEXI, regional application is achieved by coupling the TSEB with an atmospheric boundary layer (ABL) model to internally simulate land–atmosphere feedbacks on near-surface air temperature (Anderson et al. 1997, 2007a,b). In this coupled mode, ALEXI simulates air temperature Ta at the blending height internally within the ABL model, ensuring that Ta is consistent with the modeled surface fluxes from the TSEB. The TSEB is applied at two times during the morning hours, at approximately 1.5 h after local sunrise t1 and 1.5 h before local noon t2. The ABL component of ALEXI relates the rise in Ta within the mixed layer over the time interval (from t1 to t2) to the time-integrated influx of H from the surface, thus providing a means for surface energy closure (McNaughton and Spriggs 1986; Anderson et al. 1997). By using a time-differential temperature signal as input (the change in TRAD between t1 and t2), sensitivity to errors in absolute temperature retrieval is significantly reduced (Anderson et al. 1997). Here, TRAD was determined from 11-μm brightness temperature observations from the GOES-East (at 75°W) and GOES-West (at 105°W) imager instruments at a spatial resolution of 4-km (nadir). The raw brightness temperature observations were atmospherically corrected using atmospheric profiles of temperature and mixing ratio following the single channel algorithm of Price (1983). A complete description of ALEXI can be found in Anderson et al. (1997, 2007b).

In the construction of the nonprecipitation moisture source index, we use the system LE (direct soil evaporation plus canopy transpiration/evaporation) from ALEXI (hereafter referred to as LEALEXI). In particular, we focus on clear-sky estimates at t2, shortly before local noon. Anderson et al. (2013a) demonstrated that clear-sky LE is better related to surface moisture conditions than all-sky LE because it allows for a separation of evaporation constraints due to moisture stress from those due to insolation forcing and cloud conditions. Cloud-free pixels are determined by using the bispectral composite threshold technique that determines cloudy pixels from a series of threshold tests using GOES 3.7- and 11.0-μm brightness temperature measurements (Jedlovec et al. 2008).

b. Noah land surface model

The Noah LSM (version 3.2; Chen and Dudhia 2001; Ek et al. 2003) was used within the NASA Land Information System (Kumar et al. 2006) to generate prognostic model estimates of surface energy fluxes for comparison with ALEXI. This version of Noah contains several improvements from the current operational version (version 2.7), including improvements to the snow model and to the parameterization of roughness lengths for heat and moisture. Note that Noah, version 3.2, was used specifically because it lacks physics to parameterize irrigation, tile drains, or the impact of shallow groundwater on the surface energy balance. Noah is a one-dimensional soil–vegetation–atmosphere transfer (SVAT) model that physically represents energy exchange between the land surface and the atmosphere, as well as the subsurface vertical transport of water and heat. The model configuration employed here uses four soil layers representing depths between 0 and 10 cm, 10 and 40 cm, 40 and 100 cm, and 100 and 200 cm and adopts a free-drainage assumption with no representation of groundwater physics. Meteorological forcing was provided from the North American Land Data Assimilation System (NLDAS) retrospective forcing dataset (Cosgrove et al. 2003), and the model was integrated forward every 30 min. A full description of the standard model physics can be found in Chen et al. (1996) and Chen and Dudhia (2001).

Noah has been extensively validated across the CONUS by a number of recent studies (Peters-Lidard et al. 2011; Xia et al. 2012a,b; Wei et al. 2013; Long et al. 2014). In particular, Peters-Lidard et al. (2011) compared ET from NLDAS with output from two other ET estimation methods [one remote sensing (MOD16; Mu et al. 2007, 2011) and one regression-tree method based on flux tower observations (FLUXNET; Jung et al. 2010)]. That study found generally good agreement between Noah, version 2.7, and Noah, version 3.2, (used here) across the CONUS, but it did find generally higher ET from Noah, version 3.2, over much of the eastern United States. None of the Noah simulations used in that study accounted for irrigation or groundwater influences. A recent study by Long et al. (2014) assessed uncertainty in ET over several southern U.S. basins using LSMs (including Noah) and remote sensing–based ET and recommended that future ET estimation should be built on a hybrid approach that integrates strengths of LSMs and satellite-based products to constrain uncertainties.

For comparison with LEALEXI, clear-sky Noah estimates of total LE—including direct soil evaporation, canopy transpiration, and evaporation from canopy interception—were extracted at time t2 (hereafter referred to as LENoah). Estimates of LENoah were only used on days when LEALEXI is determined to be cloud-free, as explained in section 2a. However, an additional step is needed as Noah uses a model-based (NLDAS-2/NARR) incoming shortwave radiation input that may not accurately represent clear-sky conditions coincident with observed clear-sky conditions in ALEXI. Therefore, clear-sky conditions for LENoah were determined from NLDAS-2/NARR when the incoming shortwave radiation is within 10% of the average clear-sky value valid at 1.5 h before local noon. Days in which both LEALEXI and LENoah estimates were not available were excluded from the analyses.

3. Methodology

a. Model runs

Daily maps of LEALEXI and LENoah were generated across the CONUS domain (24°–50°N, 125°–65°W) at a spatial resolution of 4 km, with cloudy pixels flagged. The temporal analysis period for each dataset consists of three growing-season months [June–August (JJA)] over the period from 2000 to 2012. To facilitate a consistent comparison between the two modeling systems that minimize differences in available net radiation and potential LE, both ALEXI and Noah simulations used identical input datasets to describe the fraction of green vegetation cover, albedo, incoming short- and longwave radiation, and land cover classification. Observed vegetation cover information was obtained from the 8-day MODIS leaf area index composites (MOD15A; Myneni et al. 2002), albedo information from 16-day MODIS albedo composites (MOD43C; Moody et al. 2005), and land cover classification from the University of Maryland (UMD) global land cover datasets based on observations from the Advanced Very High Resolution Radiometer (AVHRR; Hansen et al. 2000). Hourly incoming short- and longwave radiation were provided from the NLDAS-2 meteorological forcing dataset at a 0.125° spatial resolution. Consistency between incoming radiation datasets is necessary to ensure that differences in potential LE are not contributing to observed differences in actual ET.

To create realistic initial states in Noah, soil moisture and temperature profiles were uniformly initialized and spun up by running the period from 1 January 1998 to 31 December 1999. The final states from this spinup run were then used to initialize another cycle starting on 1 January 1998, in which the model was run continuously through the analysis period ending on 31 December 2012. Starting 1 January 2000, hourly LE time series were generated for use in these analyses. Because ALEXI retrieves daily instantaneous LE over clear-sky pixels valid at ALEXI t2 (1.5 h before local noon), hourly estimates of LENoah were temporally interpolated to match this time.

An initial evaluation of LEALEXI and LENoah maps showed large differences over areas of complex terrain (not shown). These differences are likely attributed to a set of deficiencies in both models that are unique to areas of high topographic relief. These deficiencies include 1) a lack of correction for terrain shading in the current implementation of ALEXI, which can lead to anomalously high LE (low LST due to shadows is misinterpreted as a moisture signal); 2) an incomplete representation of exposed rocky terrain in ALEXI and Noah, which can lead to anomalously low ground heat flux (through an inaccurate specification of the differences in thermal conductivity between soil and rock) and an underestimation of roughness length, which in turn leads to anomalously high LEALEXI; and 3) a poor representation of deep soil moisture in Noah in regions of low vegetation and complex terrain. As a result, we chose to mask out pixels where the terrain variability is greater than 100 m (sampled from a 1-km digital elevation model where terrain variability is computed within a 3 km × 3 km box centered at each pixel) and focus our analysis exclusively on areas of low topographic relief.

Average LE values for each year were computed by compositing over the JJA time period using only clear-sky estimates of LE. Long-term averages for LEALEXI and LENoah were then computed as a nonweighted average of all JJA composites obtained during the years 2000–12. Finally, a map of the variance of LEALEXI VARALEXI was sampled across all 13 yearly (2000–12) JJA composites.

b. ASSET

Since the standard Noah formulation lacks the ability to represent any water source except local precipitation, areas with large irrigation and/or groundwater-based water inputs should demonstrate a negative bias in LENoah relative to LEALEXI. Likewise, these same areas should also be associated with relatively low VARALEXI (since effects of interannual precipitation variability will be largely muted by irrigation and/or groundwater extraction). Conversely, areas in which bottom-up Noah simulations neglect the impact of tile drainage (i.e., anthropogenic sinks) should be associated with a positive bias in LENoah relative to LEALEXI. Therefore, in an attempt to define a unitless index with large positive values in the presence of large nonprecipitation sources of water and negative values in the presence of neglected soil water sinks, we define the ALEXI Source–Sink for Evapotranspiration (ASSET) index as
e1
This unitless index can be interpreted as a qualitative indicator of the impact of nonprecipitation water sources on the diagnostic energy balance model output. In particular, large positive ASSET values reflect regions where the neglect of nonparameterized moisture sources (e.g., irrigation, extraction of shallow groundwater by phreatophytic plants or soil wicking, or direct evaporation from surface water) introduces a negative bias in growing-season LENoah results. Conversely, negative ASSET values will reflect regions where sinks in soil moisture (e.g., tile drainage) are not accurately represented in the Noah LSM, leading to a positive bias in seasonal LENoah.

Independent datasets describing irrigation percentage, open water–wetland percentage, and simulated groundwater depths are used to assess spatial maps of ASSET. In particular, estimates of irrigation percentage were obtained from Ozdogan and Gutman (2008) at a spatial resolution of 500 m. Simulated water-table depth was determined from Fan et al. (2007) and Miguez-Macho et al. (2008), with maps of climatological mean water-table depth available at a spatial resolution of 30 arc s. Both data products were spatially aggregated from their native resolutions to the 4-km spatial domain used by ALEXI and Noah. The percentage of surface wetlands in each 4-km ALEXI–Noah pixel was quantified by aggregating the 30-m resolution National Land Cover Database 2006 (NLCD; Fry et al. 2011). The NLCD is a 16-class classification based on Landsat Enhanced Thematic Mapper Plus (ETM+) and includes classes for open water, woody wetlands, and emergent herbaceous wetlands. These three classes were combined and used to calculate a percentage over each 4-km ALEXI–Noah pixel. Digital maps of tile drainage extent are difficult to obtain; however, qualitative information can be obtained from land managers and agricultural drainage specialists in various agricultural regions.

As an initial assessment of the ASSET index, we will conduct a qualitative comparison with independent irrigation, water table, and surface water proxies. Because of the nature of the problem, where ASSET represents, in many cases, the net impact of several nonprecipitation water sources, the comparisons are intended as an initial proof-of-concept and will not provide a fully quantitative assessment. They will, however, clarify the degree to which spatial patterns in ASSET index values can be credibly attributed to missing soil water balance processes attributed to irrigation and/or high water-table depths.

4. Results

a. Mean and seasonal variability of JJA clear-sky LE

Figure 1 shows CONUS-wide patterns of LEALEXI, LENoah, VARALEXI, and VARNoah computed for JJA over the period 2000–12. At the CONUS scale, spatial patterns of mean LEALEXI and LENoah are generally consistent. In general, both datasets show lower LE in the western United States with a transition zone across the central United States and higher LE over the eastern United States. As expected, the observed patterns in LEALEXI and LENoah closely match mean annual precipitation patterns (not shown). CONUS-averaged values for LEALEXI are 267.3 (±103.4) W m−2 and for LENoah are 224.4 (±111.9) W m−2. However, some differences do emerge at the regional scale. The value of LEALEXI is generally higher over most of the southeastern United States, especially along the Gulf Coast and throughout much of eastern Texas. Also, the slightly higher LEALEXI values are found in the central United States along an axis from western Texas to the western Dakotas and eastern Montana.

Fig. 1.
Fig. 1.

CONUS (a) LEALEXI (W m−2), (b) LENoah (W m−2), (c) VARALEXI (W m−2)2, and (d) VARNoah (W m−2)2 maps computed for clear-sky JJA conditions between 2000 and 2012. The color scale for (a),(b) runs from 0 to 500 in increments of 50 and for (c),(d) runs from 0 to 5000 in increments of 500.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0017.1

Maps of VARALEXI and VARNoah in Fig. 1 both show low LE variability over arid and semiarid (water limited) portions of the western United States and over the upper midwestern and northeastern United States, where changes in LE are more related to the available energy as opposed to available soil water (Figs. 1c,d). Both models show the greatest temporal variability in LE over the central Great Plains, a region where changes in clear-sky LE are closely coupled with interannual variations in available soil water (Dirmeyer et al. 2012). In general, VARNoah is higher than VARALEXI, with CONUS-averaged values of 1691.1 (W m−2)2 and 1247.0 (W m−2)2, respectively. Areas of enhanced VARNoah include eastern Texas and portions of the Mississippi River valley, collocated with higher LE values diagnosed by ALEXI. In addition, Noah predicts higher variability over southern Alabama and Georgia and all of Florida, possibly due to the sandy soil types and higher water tables prevalent in the region. The free-drainage assumption in the standard Noah configuration will be violated in regions where the water table is shallower than the bottom boundary of the soil model, leading to overestimation of near-surface moisture variability and underestimation of profile-averaged soil moisture content and evaporative flux.

b. Assessing the impact of nonprecipitation water sources on LE

Using the definition in Eq. (1), Fig. 2a maps unitless 4-km ASSET index values over the CONUS domain. Pixels where nonprecipitation moisture inputs may have a significant effect on clear-sky LE (areas of elevated LEALEXI as compared to LENoah and low VARALEXI) are denoted as positive values (green and blue tones). Conversely, negative ASSET values (red tones) potentially reflect the neglect of anthropogenic sinks in Noah surface energy balance predictions. As discussed above, areas of complex terrain are masked and shaded in gray. Spatial variations in positive ASSET values appear to accurately map the extent and the magnitude of nonprecipitation water sources (e.g., irrigation, shallow water-table depths, and surface wetlands) on the surface energy balance. Labeled domains in Fig. 2a correspond to areas examined in detail below: (i) irrigated agricultural regions of the western CONUS (e.g., the Central Valley of California, Snake River valley of southern Idaho, central Washington, and south of the Salton Sea in extreme Southern California); (ii) irrigated agricultural regions in the south-central United States (e.g., regions in the panhandle of Texas, western Kansas, and large portions of Nebraska); (iii) the north-central United States exhibiting numerous soil water sources [e.g., shallow water table and surface wetlands (prairie potholes) and sinks (e.g., extensive agricultural tile drainage)]; and (iv) the southeastern United States, a region with extensive irrigation and shallow water tables in the Mississippi River valley, extensive wetlands along the Gulf Coast, and shallow water table throughout much of Florida. Large positive ASSET values in many of these areas can generally be explained via comparisons to independent irrigation and/or groundwater depth maps shown in Figs. 2b and 2c. Likewise, the occurrence of negative ASSET values in the north-central United States may be partially attributed to the known impact of agricultural tile drains (see discussion below).

Fig. 2.
Fig. 2.

(a) ASSET computed from an average of 13 JJA composites [2000–12; positive (negative) values indicate regions where ALEXI clear-sky LE was greater (less) than Noah clear-sky LE, collocated with low annual variability of ALEXI clear-sky LE], (b) MODIS irrigation percentage (Ozdogan and Gutman 2008), and (c) simulated water-table depth (m; Fan et al. 2007; Miguez-Macho et al. 2008). The roman numeral–labeled red rectangles are referenced in subsequent figures.

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0017.1

Detailed comparisons between ASSET and known patterns of water table, irrigation, and tile drainage for each of these regions are provided below. Composites showing the relationships between elevated ASSET values (defined as greater than 500) and the proxy datasets are shown for each subregion. We also identify areas where elevated ASSET values either do not match any of the proxy datasets used in the study or, conversely, areas with negative ASSET values where a proxy dataset either shows the presence of a “positive” nonprecipitation water source (e.g., water table and irrigation) or no positive nonprecipitation water source. Note that the occurrence of negative ASSET values with a “negative” nonprecipitation water source such as tile drainage is not properly assessed because of a lack of a quantitative dataset describing current tile drainage practices.

Some ASSET–proxy mismatches are likely attributable to systematic errors in Noah observations that are not directly associated with the neglect of water source–sink processes described by the particular set of proxies we selected for comparison. For example, modest negative ASSET values (usually on the order of from −100 to −500) are observed outside of regions with significant nonprecipitation sources–sinks along the East Coast of the United States. However, Peters-Lidard et al. (2011) found that Noah, version 3.2, showed higher warm season ET over the eastern United States than Noah, version 2.7, and two reference datasets (MOD16 and FLUXNET). They hypothesized that differences may have been a result of changes in the aerodynamic conductance formulation used in Noah, version 3.2. If, in fact, Noah, version 3.2, has a wet bias over much of the eastern and southeastern United States, the modest negative ASSET values observed would be consistent with such an assumption.

Additionally, higher ASSET values are observed over the forested regions of the northeastern United States, yet no nonprecipitation water source can be identified to explain the differences in Noah and ALEXI associated with the elevated ASSET values. These unattributed ASSET signals also have value, providing an avenue for identification and potential refinement of other (currently undefined) problems existing in either prognostic (e.g., Noah) and diagnostic (e.g., ALEXI) ET estimates. These include potential deficiencies in model physics, crude representation of model parameters, and errors in forcing. However, as discussed below, other ASSET–proxy differences appear to be associated with spatial errors in the proxy itself and not ALEXI and/or Noah LE estimates used to generated the ASSET index.

1) Irrigated regions in the western United States

Large ASSET values are observed over portions of the western United States supporting significant agricultural production. Because of a prevailing arid and semiarid climate, irrigation is widely used across the western CONUS, including productions systems in the Central Valley of California, the Snake River valley of Idaho, areas of central Washington, and a large agricultural region south of the Salton Sea in Southern California. In these regions, high ASSET values are strongly spatially correlated with the MODIS irrigation percentage maps (see Fig. 2c).

The lack of an irrigation parameterization in Noah leads to average JJA midday clear-sky LE values computed over actively irrigated pixels in the western United States of only 10–30 W m−2 versus 300–400 W m−2 for ALEXI. Shallow water-table depths are also observed in the Central Valley of California, which may further contribute to relatively elevated ALEXI LE predictions. However, it is difficult for ASSET to distinguish between root extraction of groundwater and irrigation when both possibilities exist.

Figure 3 identifies areas of the western CONUS where positive ASSET values (greater than 500) are collocated with one or more nonprecipitation water sources as mapped in the proxy datasets. Additionally, it shows areas where positive ASSET values are not collocated with features in one of the nonprecipitation proxy datasets (gray), and where a nonprecipitation source is collocated with ASSET values less than zero (yellow). ASSET values greater than 500 cover 26.8% of the unmasked subdomain of which 60.8% of pixels are collocated with either the irrigation and/or groundwater proxy dataset. Of the remaining 39.2%, 2.2% percent of these unattributed high ASSET value pixels occur in coastal Southern California, generally spanning from Los Angeles to San Diego, and may be associated with urban vegetation that is irrigated during the dry season months used in this study (JJA) and therefore poorly represented in Noah. Another 3.1% lies in a patch in northeast California and southern Oregon. A visual inspection of high-resolution true-color imagery in these areas reveals several small-scale irrigated regions that are not represented in the MODIS irrigation percentage dataset.

Fig. 3.
Fig. 3.

Composite of ASSET and irrigation and groundwater proxy datasets over the western CONUS region [i.e., domain (i) in Fig. 2a].

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0017.1

The largest source of mismatch at lower values of ASSET appears to be associated with complex topographic terrain features below the masking threshold described in section 2a, potentially leading to uncertainties in the Noah and ALEXI LE estimates (i.e., model error may be a significant contributing factor in these areas). In general, there are only a few pixels with an ASSET value less than zero that are also collocated with a possible nonprecipitation moisture source (2.1% of all pixels, indicated in yellow).

2) South-central United States

Large positive ASSET values observed over the central CONUS are also generally associated with either known irrigated agricultural production or shallow groundwater (Fig. 2). For example, green pixels in Fig. 4 highlight several regions of intensive irrigation (e.g., northeastern Colorado along the South Platte River; south-central Colorado in the Rio Grande basin, and the panhandle of Texas and southeastern Kansas). ALEXI performance in retrieving LE and other energy balance components in this region was investigated in detail using flux measurements collected during the Bushland Evapotranspiration and Agricultural Remote Sensing Experiment (BEAREX) 2008 field campaign in Bushland, Texas. During BEAREX 2008, ALEXI performed well in comparison with flux observations in both irrigated and rainfed fields, with estimated errors of approximately 10% at daily time steps and 5% for seasonal cumulative ET (Anderson et al. 2012; Cammalleri et al. 2013).

Fig. 4.
Fig. 4.

Composite of ASSET and irrigation and groundwater proxy datasets over the south-central CONUS region [i.e., domain (ii) in Fig. 2a].

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0017.1

Positive ASSET values are also observed collocated with areas of irrigation and shallow groundwater within a large section of central and southwestern Nebraska; however, the magnitude of ASSET values in Nebraska is somewhat lower (by approximately 300–600) than those associated with irrigated areas in the west and southwest CONUS. This is consistent with the findings of Cammalleri et al. (2013), who compared seasonal ET curves over rainfed and irrigated agriculture in Bushland, Texas, and in Mead, Nebraska. In both cases, ALEXI reproduced observed cumulative ET curves well. The difference in ET dynamics between sites was striking, with rainfed and irrigated ET curves for cotton deviating early in the season (in July) at the semiarid Bushland site, whereas at the more humid Mead site, water use in rainfed and irrigated corn remained relatively consistent until late August—near the end of the JJA compositing interval considered in this study. At this site, the rainfed assumption in Noah may work reasonably well in many years through much of the growing season, leading to lower ASSET values.

The composite map of ASSET and the proxy datasets for the south-central CONUS subdomain show results are similar to those observed in the western CONUS subdomain, with 60.5% of pixels where ASSET was greater than 500 collocated with a known nonprecipitation proxy dataset. Less than 1% of pixels in the subdomain were found to have ASSET values less than zero when a nonprecipitation proxy was observed.

3) Southeastern United States

As in the other regions discussed above, large positive ASSET values in eastern Texas and the lower Mississippi River basin (Fig. 5) may be explained by the juxtaposition of intensive irrigation and shallow groundwater (usually under 1-m depth; see Fig. 5c). Over the agricultural portions of the region (northern Louisiana, western Mississippi, and eastern Arkansas), ASSET is likely detecting differences dominated by the influence of irrigation. However, large ASSET values are also observed along coastal sections of eastern Texas and Louisiana, characterized by a shallow simulated water-table depth and extensive coastal wetlands. While the largest ASSET values in this region can be associated with a known nonprecipitation water input, slightly elevated values are also observed over all of eastern Texas (see Fig. 2a), a region lacking an obvious nonprecipitation source of water. However, the highest ASSET values in eastern Texas (outside of coastal regions associated with lower simulated groundwater depths or coastal wetlands) are highly correlated with the density of inland lakes and water bodies that may not be adequately resolved in Noah, but are captured in the LST signal used in ALEXI.

Fig. 5.
Fig. 5.

(a) ASSET (color scale is from −3000 to +3000 in increments of 600), (b) MODIS irrigation percentage (%; color scale is from 0% to 100% in increments of 10%), (c) simulated water-table depth (m; color scale is from 0 to 2 m in increments of 0.25 m), and (d) NLCD wetland percentage (%; color scale is from 0% to 100% in increments of 10%) maps for the southeast CONUS region [i.e., domain (iv) in Fig. 2a].

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0017.1

Strong positive ASSET signals cover much of central and southern Florida (Fig. 5), regions with very shallow groundwater depth and large coverage of wetlands. In addition, the intensively irrigated agricultural production region south of Lake Okeechobee is also collocated with large positive ASSET values. As discussed earlier, the free-drainage model assumption in Noah and the existence of predominantly sandy soils may be contributing to model underestimation of evaporative flux in these areas.

Figure 6 shows the composite matching ASSET and the proxy datasets used. The southeastern United States subdomain exhibited the highest percentage of pixels with ASSET values greater than 50 and a proxy dataset match; 89.1% of all such pixels were associated with the presence of a potential nonprecipitation proxy. A much smaller number of pixels (4.3%) showed negative ASSET values that are collocated with a moisture source–sink in a proxy dataset or negative ASSET values with no moisture source–sink (2.7% of pixels).

Fig. 6.
Fig. 6.

Composite of ASSET and irrigation and groundwater proxy datasets over the southeast CONUS region [i.e., domain (iv) in Fig. 2a].

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0017.1

4) North-central United States

Significant ASSET signals are observed across regions of North Dakota, South Dakota, Minnesota, and Wisconsin. While some of the positive ASSET values can potentially be explained by groundwater influences (Fig. 7), especially in North Dakota and northern Minnesota, or irrigation, these do not seem likely to be the major explanatory factors in this region. For example, concentrated areas of large positive values are observed in eastern South Dakota and North Dakota that are not associated with either irrigation or groundwater influences. These patches are coincident with regions containing a high density of small (subpixel) shallow wetlands, also known as “prairie potholes” (Rover et al. 2011). These seasonal small-scale wetlands are not adequately parameterized in Noah, as each pixel is assigned a dominant land class and the influence of subgrid wetlands is not considered. However, the cooling influence of such wetlands is inherently captured in the LST measurement used in ALEXI.

Fig. 7.
Fig. 7.

As in Fig. 5, but for the north-central CONUS region [i.e., domain (iii) in Fig. 2a].

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0017.1

Localized large positive values of ASSET are also observed in central Minnesota and a small region of northern Wisconsin and are also clearly associated with regions of extensive small-scale lake and wetland cover. Figure 8 shows the composite map of ASSET and the proxy datasets for the subdomain. It was found that 72.3% of all pixels with an ASSET value greater than 500 are coincident with a nonprecipitation moisture source as indicated in the proxy datasets. Interestingly, this domain had the largest percentage of pixels with negative ASSET values that mapped to a moisture source in a proxy dataset. Nearly 10% of all pixels in the domain exhibited negative ASSET values that were mainly associated with shallow simulated water-table depth (see orange-shaded areas in Fig. 8), giving the visual impression of a strong spatial anticorrelation between ASSET and shallow water-table depths. This is contrary to the behavior in other subregions considered in this study where shallow water tables enhanced evaporative fluxes in ALEXI relative to Noah (and thus produced positive ASSET values). The strongest anticorrelation patterns are observed in the Red River valley along the border between Minnesota and North Dakota, in the James River valley in eastern South Dakota, and in sections of southern Minnesota, Wisconsin, and north-central Iowa.

Fig. 8.
Fig. 8.

Composite of ASSET and irrigation and groundwater proxy datasets over the north-central CONUS region [i.e., domain (iii) in Fig. 2a].

Citation: Journal of Hydrometeorology 16, 3; 10.1175/JHM-D-14-0017.1

A potential explanation for these patterns is the intensive expansion of subsurface agricultural tile drainage installation in these regions over the past decade (R. Finocchiaro, USGS Northern Prairie Wildlife Research Center, 2013, personal communication). Tile drainage is commonly employed in areas with poor natural drainage, which can adversely affect agricultural production during wet periods. As has been demonstrated in prior sections, ALEXI is able to capture impacts of nonprecipitation water sources by using changes in LST to estimate surface fluxes. It could be expected that the same methods may also be sensitive to anthropogenic removal of soil moisture that would not be adequately represented in most land surface models, including the version of Noah used in this study. While gridded datasets recording percentage of acreage drained in these regions are limited, it is likely that tile drainage may be at least partially responsible for the negative ASSET values observed in this subdomain. However, a more detailed analysis of drain installation history and concomitant changes in the ALEXI ET record over the past decade is underway to better understand this phenomenon.

5. Discussion and conclusions

This study compares multiyear (2000–12) JJA clear-sky latent heat flux from a bottom-up land surface model (Noah, version 3.2) and a top-down diagnostic two-source energy balance model forced with observations of LST change (ALEXI) to attempt to diagnose regions where nonprecipitation water inputs (and/or anthropogenic water sinks) have an observable impact on terrestrial LE. These comparisons lead to the development of the ALEXI Source–Sink for Evapotranspiration (ASSET) index, which is computed from three quantities related to nonprecipitation water inputs: 1) positive differences between JJA ALEXI and Noah clear-sky latent heat flux, 2) large JJA clear-sky latent heat flux in ALEXI, and 3) low variance in ALEXI JJA clear-sky latent heat flux. Positive ASSET values reflect the neglect of soil water source processes by Noah. Likewise, negative (or small positive) values of the index can be used to diagnose the neglected impact of soil water sinks on Noah ET simulations. In this study Noah, version 3.2, was specifically chosen as the baseline model because it does not incorporate irrigation, tile drainage, or groundwater physics and does not consider the subpixel effects of wetlands and/or open water bodies. As such it provides an appropriate point of comparison for ALEXI to diagnose and spatially map the impact of nonprecipitation water inputs (or nonresolved outputs) on the surface energy balance.

Both ALEXI and Noah were forced with identical incoming short- and longwave information to maintain consistency with regards to the estimation of energy available for LE. Comparisons against independent irrigation and groundwater depth maps suggest that ASSET can reliably map areas in which nonprecipitation-based water sources have a significant impact on the surface energy balance. In addition, ASSET is also potentially highlighting deficiencies in the proxy datasets used in the study. It is also important to consider that, while the agreement between ASSET and the proxy datasets are relatively strong, each proxy dataset contains uncertainties regarding the physical quantity it attempts to represent. An interesting example of this is over regions of the upper Midwest, where negative ASSET values were found in areas where a groundwater analysis predicts shallow water-table depths. This apparent inconsistency likely indicates the impact of extensive tile drainage (i.e., anthropogenic soil water sinks) on the surface energy balance—a process not explicitly considered in the water balance analysis underlying the water-table depth proxy.

While ASSET almost certainly responds to Noah-specific biases and errors (which are not associated with missing processes), we observe enough of a qualitative relationship between ASSET and the proxy datasets to suggest that ASSET can be reliably interpreted as a proxy for the impact of these missing processes on Noah energy balance estimates. ALEXI observations underlying the ASSET index may also contain errors that may not be related to the missing processes in Noah; however, the high degree of spatial correspondence observed in the assessment suggests that there errors do not preclude the physical interpretation of ASSET index values as a robust diagnostic for the presence of missing water balance process in Noah simulations. At the very least, this supports the idea that ASSET can be used as an independent tool to refine and/or update proxy maps. In addition, when large systematic differences between ASSET and the proxies are noted, it can often be credibly attributed to processes (e.g., tile drainage) not considered in a particular proxy. Therefore, systematic qualitative differences between ASSET and the proxy maps can be used to identify systematic shortcoming in the proxies. Even if their quantitative attributes have not been fully validated, ASSET maps provide new information in a number of ways. In particular, they 1) provide an independent source of information that can be used to filter random spatial errors existing in the proxy products, 2) reveal systematic errors that may be present in the proxies, 3) convey an integrated assessment of impact on the surface water balance as opposed to proxy maps for individual processes, and 4) can be (potentially) scaled up to provide global analysis. In addition, ASSET can potentially be used to provide seasonally varying information. However, additional assessment will be required to validate this potential.

ASSET may also play an important role in ongoing improvements to LSMs via the introduction of new irrigation and/or groundwater physics. The success of these efforts will depend in part on the availability of sufficient independent information to accurately parameterize and/or evaluate these new processes and their subsequent impact on the surface water and energy balance. This is especially true in data-poor areas of water where ancillary information regarding water-table heights and irrigation inputs is difficult to obtain (see Yilmaz et al. 2014; Sutanudjaja et al. 2014). It should also be noted that while agreement between ASSET and the proxy datasets leads to an inferred credibility in both ASSET and the proxy datasets, only ASSET (or ALEXI) is capable of directly quantifying the impact of these nonprecipitation inputs on the surface energy balance without reliance on extensive ground-based information. Therefore, results presented here establish the potential role of remotely sensed surface energy flux maps in general, and ALEXI in particular, in improving the parameterization of nonprecipitation water sources (and anthropogenic water sinks) in LSM simulations.

This first attempt at using a diagnostic method such as ALEXI to assess the role of missing nonprecipitation water sources–sinks in a prognostic land surface model (Noah) will serve as the framework for identifying regions that need more sophisticated methodologies (e.g., water balance analysis) to truly quantify the role of such sources–sinks.

Acknowledgments

This work was supported by NASA Applied Sciences Water Resources Program Award NNX12AK90G, “Development of a Multi-Scale Remote-Sensing Based Framework for Mapping Drought over North America,” and NOAA Climate Program Office Award NA11OAR4310139, “Dual Assimilation of Microwave and Thermal-Infrared Satellite Observations of Soil Moisture into NLDAS for Improved Drought Monitoring.” We wish to thank the anonymous reviewers for helping to significantly improve the quality of the manuscript.

REFERENCES

  • Anderson, M. C., , Norman J. M. , , Diak G. R. , , Kustas W. P. , , and Mecikalski J. R. , 1997: A two-source time-integrated model for estimating surface fluxes using thermal infrared remote sensing. Remote Sens. Environ., 60, 195216, doi:10.1016/S0034-4257(96)00215-5.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., , Neale C. M. U. , , Li F. , , Norman J. M. , , Kustas W. P. , , Jayanthi H. , , and Chavez J. , 2004a: Upscaling ground observations of vegetation water content, canopy height, and leaf area index during SMEX02 using aircraft and Landsat imagery. Remote Sens. Environ., 92, 447464, doi:10.1016/j.rse.2004.03.019.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., , Norman J. M. , , Mecikalski J. R. , , Torn R. D. , , Kustas W. P. , , and Basara J. B. , 2004b: A multi-scale remote sensing model for disaggregating regional flues to micrometeorological scales. J. Hydrometeor., 5, 343363, doi:10.1175/1525-7541(2004)005<0343:AMRSMF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., , Norman J. M. , , Kustas W. P. , , Li F. , , Prueger J. H. , , and Mecikalski J. R. , 2005: Effects of vegetation clumping on two-source model estimates of surface energy fluxes from an agricultural landscape using SMACEX. J. Hydrometeor., 6, 892909, doi:10.1175/JHM465.1.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., , Kustas W. P. , , and Norman J. M. , 2007a: Upscaling tower and aircraft fluxes from local to continental scales using thermal remote sensing. Agron. J., 99, 240254, doi:10.2134/agronj2005.0096S.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., , Norman J. M. , , Kustas W. P. , , Li F. , , Prueger J. H. , , and Mecikalski J. R. , 2007b: A climatological study of evapotranspiration and moisture stress across the continental United States: 1. Model formulation. J. Geophys. Res., 112, D11112, doi:10.1029/2006JD007507.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., and Coauthors, 2012: Mapping daily evapotranspiration at Landsat spatial scales during the BEAREX’08 field campaign. Adv. Water Resour., 50, 162177, doi:10.1016/j.advwatres.2012.06.005.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., , Hain C. , , Otkin J. , , Zhan X. , , Mo K. , , Svoboda M. , , Wardlow B. , , and Pimstein A. , 2013a: An intercomparison of drought indicators based on thermal remote sensing and NLDAS-2 simulations with U.S. Drought Monitor classifications. J. Hydrometeor., 14, 10351056, doi:10.1175/JHM-D-12-0140.1.

    • Search Google Scholar
    • Export Citation
  • Anderson, M. C., , Kustas W. P. , , and Hain C. R. , 2013b: Mapping surface fluxes and moisture conditions from field to global scales using ALEXI/DisALEXI. Remote Sensing of Energy Fluxes and Soil Moisture Content, G. P. Petropoulos, Ed., CRC Press, 207–232, doi:10.1201/b15610-11.

  • Cammalleri, C., , Anderson M. C. , , Ciraolo G. , , D'Urso G. , , Kustas W. P. , , La Loggia G. , , and Minacapilli M. , 2012: Applications of a remote sensing–based two-source energy balance algorithm for mapping surface fluxes without in situ air temperature observations. Remote Sens. Environ., 124, 502515, doi:10.1016/j.rse.2012.06.009.

    • Search Google Scholar
    • Export Citation
  • Cammalleri, C., , Anderson M. C. , , Gao F. , , Hain C. R. , , and Kustas W. P. , 2013: A data fusion approach for mapping daily evapotranspiration at field scale. Water Resour. Res., 49, 46724686, doi:10.1002/wrcr.20349.

    • Search Google Scholar
    • Export Citation
  • Cammalleri, C., , Anderson M. C. , , Gao F. , , Hain C. R. , , and Kustas W. P. , 2014: Mapping daily Evapotranspiration at field scales over rainfed and irrigated agricultural areas using remote sensing data fusion. Agric. For. Meteor., 186, 111, doi:10.1016/j.agrformet.2013.11.001.

    • Search Google Scholar
    • Export Citation
  • Chen, F., , and Dudhia J. , 2001: Coupling an advanced land surface hydrology model with the Penn State/NCAR MM5 modeling system. Part 1: Model description and implementation. Mon. Wea. Rev., 129, 569586, doi:10.1175/1520-0493(2001)129<0569:CAALSH>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Chen, F., and Coauthors, 1996: Modeling of land-surface evaporation by four schemes and comparison with FIFE observations. J. Geophys. Res., 101, 72517268, doi:10.1029/95JD02165.

    • Search Google Scholar
    • Export Citation
  • Cosgrove, B. A., and Coauthors, 2003: Real-time and retrospective forcing in the North American Land Data Assimilation System (NLDAS) project. J. Geophys. Res., 108, 8842, doi:10.1029/2002JD003118.

    • Search Google Scholar
    • Export Citation
  • Dickinson, R. E., , Henderson-Sellers A. , , and Kennedy P. J. , 1993: Biosphere–Atmosphere Transfer Scheme (BATS) Version 1e as coupled to the NCAR Community Climate Model. NCAR Tech. Note NCAR-TN-387+STR, 88 pp., doi:10.5065/D67W6959.

  • Dirmeyer, P. A., and Coauthors, 2012: Evidence for enhanced land–atmosphere feedback in a warming climate. J. Hydrometeor., 13, 981995, doi:10.1175/JHM-D-11-0104.1.

    • Search Google Scholar
    • Export Citation
  • Ek, M. B., , Mitchell K. E. , , Lin Y. , , Rogers E. , , Grummann P. , , Koren V. , , Gayno G. , , and Tarpley J. D. , 2003: Implementation of Noah land surface model advances in the National Centers for Environmental Prediction operational mesoscale Eta model. J. Geophys. Res., 108, 8851, doi:10.1029/2002JD003296.

    • Search Google Scholar
    • Export Citation
  • Fan, Y., , and Miguez-Macho G. , 2011: A simple hydrologic framework for simulating wetlands in climate and Earth system models. Climate Dyn., 37, 253278, doi:10.1007/s00382-010-0829-8.

    • Search Google Scholar
    • Export Citation
  • Fan, Y., , Miguez-Macho G. , , Weaver C. P. , , Walko R. , , and Robock A. , 2007: Incorporating water table dynamics in climate modeling: 1. Water table observations and equilibrium water table simulations. J. Geophys. Res., 112, D10125, doi:10.1029/2006JD008111.

    • Search Google Scholar
    • Export Citation
  • Fan, Y., , Li H. , , and Miguez-Macho G. , 2013: Global patterns of groundwater table depth. Science, 339, 940943, doi:10.1126/science.1229881.

    • Search Google Scholar
    • Export Citation
  • Fry, J., and Coauthors, 2011: Completion of the 2006 National Land Cover Database for the conterminous United States. Photogramm. Eng. Remote Sens.,77, 858–864.

  • Gedney, N., , and Cox P. M. , 2003: The sensitivity of global climate model simulations to the representation of soil moisture heterogeneity. J. Hydrometeor., 4, 12651275, doi:10.1175/1525-7541(2003)004<1265:TSOGCM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Gedney, N., , Cox P. M. , , Betts R. A. , , Boucher O. , , Huntingford C. , , and Stott P. A. , 2006: Detection of a direct carbon dioxide effect in continental river runoff records. Nature, 439, 835–838, doi:10.1038/nature04504.

    • Search Google Scholar
    • Export Citation
  • Haddeland, I., , Skaugen T. , , and Lettenmaier D. P. , 2006: Anthropogenic impacts on continental surface water fluxes. Geophys. Res. Lett.,33, L08406, doi:10.1029/2006GL026047.

  • Hain, C. R., , Mecikalski J. R. , , and Anderson M. C. , 2009: Retrieval of an available water-based soil moisture proxy from thermal infrared remote sensing. Part I: Methodology and validation. J. Hydrometeor., 10, 665683, doi:10.1175/2008JHM1024.1.

    • Search Google Scholar
    • Export Citation
  • Hain, C. R., , Crow W. T. , , Mecikalski J. R. , , Anderson M. C. , , and Holmes T. , 2011: An intercomparison of available soil moisture estimates from thermal infrared and passive microwave remote sensing and land surface modeling. J. Geophys. Res., 116, D15107, doi:10.1029/2011JD015633.

    • Search Google Scholar
    • Export Citation
  • Hansen, M. C., , DeFries R. S. , , Townshend J. R. G. , , and Sohlberg R. , 2000: Global land cover classification at 1 km spatial resolution using a classification tree approach. Int. J. Remote Sens., 21, 13311364, doi:10.1080/014311600210209.

    • Search Google Scholar
    • Export Citation
  • Harding, K. J., , and Snyder P. K. , 2012: Modeling the atmospheric response to irrigation in the Great Plains. Part II: The precipitation or irrigated water and changes in precipitation recycling. J. Hydrometeor., 13, 16871703, doi:10.1175/JHM-D-11-099.1.

    • Search Google Scholar
    • Export Citation
  • Jedlovec, G. J., , Haines S. L. , , and LaFontaine F. J. , 2008: Spatial and temporal varying thresholds for cloud detection in GOES imager. IEEE Trans. Geosci. Remote Sens., 46, 17051717, doi:10.1109/TGRS.2008.916208.

    • Search Google Scholar
    • Export Citation
  • Jung, M., and Coauthors, 2010: Recent decline in the global land evapotranspiration trend due to limited moisture supply. Nature, 467, 951954, doi:10.1038/nature09396.

    • Search Google Scholar
    • Export Citation
  • Koch, S., , Bauwe A. , , and Lennartz B. , 2013: Application of the SWAT model for a tile-drained lowland catchment in north-eastern Germany on subbasin scale. Water Resour. Manage., 27, 791805, doi:10.1007/s11269-012-0215-x.

    • Search Google Scholar
    • Export Citation
  • Kumar, S. V., and Coauthors, 2006: Land information system: An interoperable framework for high resolution land surface modeling. Environ. Modell. Software, 21, 14021415, doi:10.1016/j.envsoft.2005.07.004.

    • Search Google Scholar
    • Export Citation
  • Liang, X., , Xie Z. , , and Huang M. , 2003: A new parameterization for surface and groundwater interactions and its impact on water budgets with the variable infiltration capacity (VIC) land surface model. J. Geophys. Res., 108, 8613, doi:10.1029/2002JD003090.

    • Search Google Scholar
    • Export Citation
  • Long, D., , Longuevergne L. , , and Scanlon B. , 2014: Uncertainty in evapotranspiration from land surface modeling, remote sensing and GRACE satellites. Water Resour. Res., 50, 11311151, doi:10.1002/2013WR014581.

    • Search Google Scholar
    • Export Citation
  • Maxwell, R. M., , and Miller N. L. , 2005: On the development of a coupled land surface and groundwater model. J. Hydrometeor., 6, 233247, doi:10.1175/JHM422.1.

    • Search Google Scholar
    • Export Citation
  • McNaughton, K. J., , and Spriggs T. W. , 1986: A mixed-layer model for regional evaporation. Bound.-Layer Meteor., 34, 243262, doi:10.1007/BF00122381.

    • Search Google Scholar
    • Export Citation
  • Miguez-Macho, G., , Fan Y. , , Weaver C. P. , , Walko R. , , and Robock A. , 2007: Incorporating water table dynamics in climate modeling: 2. Formulation, validation, and soil moisture simulation. J. Geophys. Res., 112, D13108, doi:10.1029/2006JD008112.

    • Search Google Scholar
    • Export Citation
  • Miguez-Macho, G., , Li H. , , and Fan Y. , 2008: Simulated water table and soil moisture climatology over North America. Bull. Amer. Meteor. Soc., 89, 663672, doi:10.1175/BAMS-89-5-663.

    • Search Google Scholar
    • Export Citation
  • Moody, E. G., , King M. D. , , Platnick S. , , Schaaf C. B. , , and Gao F. , 2005: Spatially complete global spectral surface albedos: Value-added datasets derived from Terra MODIS land products. IEEE Trans. Geosci. Remote Sens., 43, 144158, doi:10.1109/TGRS.2004.838359.

    • Search Google Scholar
    • Export Citation
  • Mu, Q., , Heinsch F. A. , , Zhao M. , , and Running S. W. , 2007: Development of a global evapotranspiration algorithm based on MODIS and global meteorology data. Remote Sens. Environ., 111, 519536, doi:10.1016/j.rse.2007.04.015.

    • Search Google Scholar
    • Export Citation
  • Mu, Q., , Zhao M. , , Kimball J. S. , , McDowell N. G. , , and Running S. W. , 2011: Improvements to a MODIS global terrestrial evapotranspiration algorithm. Remote Sens. Environ., 115, 17811800, doi:10.1016/j.rse.2011.02.019.

    • Search Google Scholar
    • Export Citation
  • Myneni, R. B., and Coauthors, 2002: Global products of vegetation leaf area and fraction absorbed PAR from year one of MODIS data. Remote Sens. Environ., 83, 214231, doi:10.1016/S0034-4257(02)00074-3.

    • Search Google Scholar
    • Export Citation
  • 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.

    • Search Google Scholar
    • Export Citation
  • Niu, G.-Y., and Coauthors, 2011: The community Noah land surface model with multiparameterization options (Noah-MP): 1. Model description and evaluation with local-scale measurements. J. Geophys. Res., 116, D12109, doi:10.1029/2010JD015139.

    • Search Google Scholar
    • Export Citation
  • Norman, J. M., , Kustas W. P. , , and Humes K. S. , 1995: A two-source approach for estimating soil and vegetation energy fluxes from observations of directional radiometric surface temperatures. Agric. For. Meteor., 77, 263293, doi:10.1016/0168-1923(95)02265-Y.

    • Search Google Scholar
    • Export Citation
  • Ozdogan, M., , and Gutman G. , 2008: A new methodology to map irrigated areas using multi-temporal MODIS and ancillary data: An application example in the continental US. Remote Sens. Environ., 112, 35203537, doi:10.1016/j.rse.2008.04.010.

    • Search Google Scholar
    • Export Citation
  • Ozdogan, M., , Rodell M. , , Beaudoing H. K. , , and Toll D. L. , 2010: Simulating the effects of irrigation over the United States in a land surface model based on satellite-derived agricultural data. J. Hydrometeor., 11, 171184, doi:10.1175/2009JHM1116.1.

    • Search Google Scholar
    • Export Citation
  • Pervez, M. S., , and Brown J. F. , 2010: Mapping irrigated lands at 250-m scale by merging MODIS data and national agricultural statistics. Remote Sens., 2, 23882412, doi:10.3390/rs2102388.

    • Search Google Scholar
    • Export Citation
  • Peters-Lidard, C. D., , Kumar S. V. , , Mocko D. M. , , and Tian Y. , 2011: Estimating evapotranspiration with land data assimilation systems. Hydrol. Processes, 25, 39793992, doi:10.1002/hyp.8387.

    • Search Google Scholar
    • Export Citation
  • Price, J. C., 1983: Estimating surface temperatures from satellite thermal infrared data—A simple formulation for the atmospheric effect. Remote Sens. Environ., 13, 353361, doi:10.1016/0034-4257(83)90036-6.

    • Search Google Scholar
    • Export Citation
  • Rover, J., , Wright C. K. , , Euliss N. H. , , Mushet D. M. , , and Wylie B. K. , 2011: Classifying the hydrologic function of prairie potholes with remote sensing and GIS. Wetlands, 31, 319327, doi:10.1007/s13157-011-0146-y.

    • Search Google Scholar
    • Export Citation
  • Sutanudjaja, E. H., , van Beek L. P. H. , , de Jong S. M. , , van Geer F. C. , , and Bierkens M. F. P. , 2014: Calibrating a large-extent high-resolution coupled groundwater–land surface model using soil moisture and discharge data. Water Resour. Res., 50, 687–705, doi:10.1002/2013WR013807.

    • Search Google Scholar
    • Export Citation
  • Wei, H., , Xia Y. , , Mitchell K. , , and Ek M. , 2013: Improvement of the Noah land surface model for warm season processes: Evaluation of water and energy flux simulation. Hydrol. Processes, 27, 297303, doi:10.1002/hyp.9214.

    • Search Google Scholar
    • Export Citation
  • Xia, Y., and Coauthors, 2012a: Continental-scale water and energy flux analysis and validation for the North American Land Data Assimilation System project phase 2 (NLDAS-2): 1. Intercomparison and application of model products. J. Geophys. Res., 117, D03109, doi:10.1029/2011JD016048.

    • Search Google Scholar
    • Export Citation
  • Xia, Y., and Coauthors, 2012b: Continental-scale water and energy flux analysis and validation for North American Land Data Assimilation System project phase 2 (NLDAS-2): 2. Validation of model-simulated streamflow. J. Geophys. Res., 117, D03110, doi:10.1029/2011JD016051.

    • Search Google Scholar
    • Export Citation
  • Yang, Z.-L., and Coauthors, 2011: The community Noah land surface model with multiparameterization options (Noah-MP): 2. Evaluation over global river basins. J. Geophys. Res., 116, D12110, doi:10.1029/2010JD015140.

    • Search Google Scholar
    • Export Citation
  • Yeh, P. J. F., , and Eltahir E. A. B. , 2005: Representation of water table dynamics in a land surface scheme. Part I: Model development. J. Climate, 18, 18611880, doi:10.1175/JCLI3330.1.

    • Search Google Scholar
    • Export Citation
  • Yilmaz, M. T., , Anderson M. C. , , Zaitchik B. , , Hain C. R. , , Crow W. T. , , Ozdogan M. , , Chun J. A. , , and Evans J. , 2014: Comparison of prognostic and diagnostic surface flux modeling approaches over the Nile River basin. Water Resour. Res., 50, 386408, doi:10.1002/2013WR014194.

    • Search Google Scholar
    • Export Citation
  • York, J. P., , Person M. , , Gutowski W. J. , , and Winter T. C. , 2002: Putting aquifers into atmospheric simulation models: An example from the Mill Creek watershed, northeastern Kansas. Adv. Water Resour., 25, 221238, doi:10.1016/S0309-1708(01)00021-5.

    • Search Google Scholar
    • Export Citation
Save