The Design Document page provides a description of the algorithms, implementation and planned testing including unit, verification, validation and performance testing. Please read Step 1.3 Performance Expectations that explains feature documentation requirements from the performance group point of view.
...
A module simulating the thermal stratification processes in reservoirs, including the advective heat fluxes into and out of reservoirs, vertical heat exchanges between reservoir surface and atmosphere, and vertical heat fluxes between multiple vertical layers. The success of the code would be satisfactory performance evaluated against the observed reservoir surface temperature and vertical temperature profiles.
Requirements
Requirement: Implementing reservoir heat processes on top of MOSART-heat and MOSART-wm.
Date last modified: July 5, 2020
Contributors: Hongyi Li, Wondie Yigzaw (Unlicensed)
MOSART-reservoir stratification has been developed and documented in Yigzaw et al. (2019). Generally, MOSART-reservoir stratification simulates multi-layer thermal processes within a reservoir with water and heat fluxes between the lakes and atmosphere and upstream/downstream rivers. The parameterization of this module is achieved by a global storage-area-depth dataset developed by Yigzaw et al. (2018).
W Yigzaw, HY Li*, Y Demissie, MI Hejazi, LR Leung, N Voisin, R Payn, 2018. A New Global Storage-Area-Depth Dataset for Modeling Reservoirs in Land Surface and Earth System Models, Water Res. Res., https://doi.org/10.1029/2017WR022040.
Yigzaw, W., Li, H.‐Y.*, Fang, X., Leung, L. R., Voisin, N., Hejazi, M. I., & Demissie, Y. (2019). A multilayer reservoir thermal stratification module for earth system models. Journal of Advances in Modeling Earth Systems, 11, 3265–3283. https://doi.org/10.1029/2019MS001632.
Requirement: Evaluation of MOSART-reservoir stratification
...
Contributors: Hongyi Li, Wondie Yigzaw (Unlicensed)
MOSART-reservoir stratification has been successfully evaluated over the contiguous U.S. against the observed reservoir surface temperature and downstream water temperature. For details please refer to Yigzaw et al. (2019).
Yigzaw, W., Li, H.‐Y.*, Fang, X., Leung, L. R., Voisin, N., Hejazi, M. I., & Demissie, Y. (2019). A multilayer reservoir thermal stratification module for earth system models. Journal of Advances in Modeling Earth Systems, 11, 3265–3283. https://doi.org/10.1029/2019MS001632.
Algorithmic Formulations
Date last modified: July 05, 2020
Contributors: Hongyi Li, Wondie Yigzaw (Unlicensed)
Thermal stratification is modeled using a heat transfer equation, which can be quite complex in two (three)
dimensions or simple in one‐dimensional representation. Two‐ and three‐dimensional models present
technical challenges as they require more input datasets for parametrizations. In contrast, one‐dimensional
models have been used with sufficient accuracy for estimating reservoir temperature at different depths.
The main sources (or sinks) of heat energy in the reservoir are atmospheric radiation and upstream
inflow (outflow downstream) while the only mechanical energy source is wind. Our modeling approach uses
the one‐dimensional heat equation with eddy diffusion (equation (1).
where T is water temperature (°C), t is time (s), z represents specific layer, Az is surface area (km2), αz is the
effective (eddy and molecular) diffusion coefficient (m2/s), ϕ is heat source from atmospheric and terrestrial
radiation (W/m2), vz is layer volume (m3), ρ density of water (kg/m3), c is water specific heat capacity (J/kg/°C),
qi ¯ z and qo ¯ z are inflow and outflow to layer z (m3/s), respectively, and Tr, Tz represent temperature (°C) of
reservoir inflow and outflow from reservoir, respectively. The differential equation is numerically solved to
estimate the layer temperature and its dynamics. The first part in the right‐hand side of equation (1) represents
the temperature change due to heat transfer by diffusion. Temperature change due to heat loss or gain
from atmospheric radiation, and advection is represented by the second and third terms, respectively.
Design and Implementation
Implementation: Implementing reservoir stratification in E3SM
Date last modified: July 05 2020
Contributors: Hongyi Li, Wondie Yigzaw (Unlicensed)
The implementation of MOSART-reservoir stratification within E3SM largely follows the software engineering protocols used in E3SMv2 of MOSART. One major difference comparing to the version in Yigzaw et al. (2019) is that now the reservoir stratification module is integrated online instead of offline, i.e., now the stratification module is exchanging fluxes with the river and atmosphere at each time step via the flux coupler.
Planned Verification and Unit Testing
Verification and Unit Testing: verifying MOSART-reservoir stratification
Date last modified: July 05, 2020
Contributors: Hongyi Li, Wondie Yigzaw (Unlicensed)
Verification is performed by checking the fields that are passed from EAM to MOSART-reservoir stratification through the flux coupler. MOSART-reservoir stratification has been tested and passed the E3SM_developer tests on Compy.
Planned Validation Testing
Validation Testing: validating MOSART-reservoir stratification
Date last modified: July 05, 2020
Contributors: Hongyi Li, Wondie Yigzaw (Unlicensed)
MOSART-reservoir stratification has been successfully evaluated over the contiguous U.S. against the observed reservoir surface temperature and downstream water temperature. For details please refer to Yigzaw et al. (2019).
Yigzaw, W., Li, H.‐Y.*, Fang, X., Leung, L. R., Voisin, N., Hejazi, M. I., & Demissie, Y. (2019). A multilayer reservoir thermal stratification module for earth system models. Journal of Advances in Modeling Earth Systems, 11, 3265–3283. https://doi.org/10.1029/2019MS001632.
Planned Performance Testing
Performance Testing: MOSART-reservoir stratification computing performance
Date last modified: July 05, 2020
Contributors: Hongyi Li, Wondie Yigzaw (Unlicensed)
ELM-MOSART-heat-stratification has been run globally with water management on Compy using 4 nodes (144 cores) during a historical period 1972-2004. The total running time is ~47 ~17 hours.