rcabell
commented
3 years ago CONUS regression tests on Cheyenne all pass with the default channel_loss_option = 0 setting. Turning on that channel loss setting causes tests to fail as expected: that there is a new output variable in the candidate files qloss that doesn't exist in the reference.
Addition of University of Arizona Channel Infiltration Function
TYPE: new feature
KEYWORDS: Channel Infiltration, Ephemeral Channels, Semi-Arid Environments
SOURCE: Timothy Lahmers; NASA-GSFC Hydrological Sciences Laboratory (HSL; Code 617), Previous affiliation: University of Arizona Department of Hydrology and Atmospheric Sciences (UA-HAS).
DESCRIPTION OF CHANGES:
Feature Summary: In semi-arid environments, it is common for low-order (and sometimes even high-order) stream channels to lose water due to low soil moisture and high depth to groundwater. In these ephemeral channels, this loss of water is another sink term; however, WRF-Hydro does not account for this. This addition to the code permits WRF-Hydro to simulate channel transmission losses (as an added sink-term), when the Muskingum-Cunge (hydro.namelist channel_option = 2) is chosen and user-defined mapping is activated (udmp = 1). Channel infiltration is activated when the Kchan variable in the RouteLink.nc file is greater than zero. Otherwise, channel infiltration is zero, and the model behaves as it would in previous versions of WRF-Hydro. Instantaneous channel infiltration (ms-1) for a given reach is output in the CHRTOUT.YYYYMMDDHHMM files as the variable qloss. Channel infiltration has been demonstrated to add value in two southern Arizona catchments in Lahmers et al. (2019; J. Hydrometeor.) and in 54 different semi-arid environments in Lahmers et al. (2020; manuscript submitted to J. Hydrometeor.).
Detailed Description: In semi-arid environments, it is common for low-order (and sometimes even high-order) stream channels to lose water due to low soil moisture and high depth to groundwater (e.g. Blasch et al. 2004). In these ephemeral channels, this loss of water is another sink term; however, WRF-Hydro does not account for this. WRF-Hydro is now able to simulate channel transmission losses (as an added sink-term), when the Muskingum-Cunge (hydro.namelist channel_option = 2) is chosen and user-defined mapping is activated (udmp = 1). As channel transmission losses are an additional sink-term. Channel infiltration is activated when the Kchan variable in the RouteLink.nc file is greater than zero. Otherwise, channel infiltration is zero, and the model behaves as it would in previous versions of WRF-Hydro. Instantaneous channel infiltration (ms-1) for a given reach is output in the CHRTOUT.YYYYMMDDHHMM files as the variable qloss. Channel infiltration has been demonstrated to add value in two southern Arizona catchments in Lahmers et al. (2019) and in 54 different semi-arid environments in Lahmers et al. (2020; manuscript submitted to J. Hydrometeor.).
The channel loss function computes channel infiltration from the wetted area in a channel reach (length times modified wetted perimeter) multiplied by the prescribed channel conductivity (Kchan), a tunable model parameter (equivalent to ChannK in Lahmers et al. 2019). In equation (1), channel infiltration I (m3 s-1) may be derived as: I = klpe. (1).
Here, k (ms-1) is the saturated conductivity of the channel bed (the Kchan parameter), and l (m) is the length of the reach. In the Muskingum-Cunge routing scheme of WRF-Hydro, channel infiltration is accounted for as an added sink in the iterative calculation, by deriving the effective flow height and wetted perimeter for a reach volume and with an assigned saturated conductivity (Kchan) for the channel bed below. To account for the fact that water in a trapezoidal channel will tend to follow a preferred path in low flow conditions and thus will not cover the whole perimeter of the trapezoidal channel, p is reduced during periods of low flow using a conceptual model similar to what is used in KINEROS2 (Goodrich et al. 2004; Woolhiser et al. 1990). KINEROS2 computes a corrected wetted perimeter (pe) for a channel using the function shown in equation (2): p_e=min[h/(b√w),1]p (2)
In the model source code, b is assumed to be a constant value of 0.15 when computing the wetted perimeter, the same as is used in previous KINEROS2 literature and in Lahmers et al. (2019). All water that drains out of the channels through the loss function is removed from the model (i.e. as a sink) not returned to the LSM, channel, or baseflow bucket model. A detailed description of the channel infiltration function may be found in Lahmers et al. (2019).
References for Detailed Description: Blasch, K., T. P. Ferré, J. Hoffmann, D. Poll, M. Baily, and J. Cordova, 2004: Processes Controlling Recharge Beneath Ephemeral Streams in Southern Arizona, Groundwater Recharge in a Desert Environment: The Southwestern United States, J. F. Hogan, F. M. Phillips and B. R. Scanlon, Eds., American Geophysical Union, 69-76, doi:https://doi.org/10.1029/009WSA05.
Goodrich, D. C., D. G. Williams, C. L. Unkrich, J. F. Hogan, R. L. Scott, K. R. Hultine, D. Pool, A. L. Coes, and S. Miller 2004: Comparison of Methods to Estimate Ephemeral Channel Recharge, Walnut Gulch San Pedro River Basin, Arizona. Recharge and Valdose Zone Processes: Alluvia Basins of the Southwestern United States, Eds., F. M. Phillips, J. F. Hogan, and B. Scanlon, 77-99.
Lahmers T. M., H. V. Gupta, C. L. Castro, D. J. Gochis, D. N. Yates, A. L. Dugger, D. C. Goodrich, and P. Hazenberg, 2019: Enhancing the Structure of the WRF-Hydro Hydrologic Model for Semi-Arid Environments, J. Hydrometeor., 20, 691-714, doi: https://doi.org/10.1175/JHM-D-18-0064.1.
Woolhiser, D. A., R. E. Smith, and D. C. Goodrich, 1990: A kinematic runoff and erosion manual: Documentation and user manual, ARS 77, US Department of Agriculture.
ISSUE: 515
TESTS CONDUCTED: Preliminary versions of this function were tested by the developers, and results are documented in a manuscript that was submitted to J. Hydrometeor. in August 2020. Prior to calibration, this function was tested in one of the NWM v2.1 calibration basins (USGS 09505350) in central Arizona. Additional testing by the developers is available upon request.
NOTES: Additional tests will likely be needed. I defer to my contact on which tests will be required.
Checklist
Merging the PR depends on following checklist being completed. Add
Xbetween each of the square brackets if they are completed in the PR itself. If a bullet is not relevant to you, please comment on why below the bullet.NEWS.md