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. 

Design Document

In the table below 4.Equ means Equations and Algorithms, 5.Ver means Verification, 6.Perf - Performance, 7. Val - Validation,    - competed,  - in progress,  - not done

Overview table for the owner and an approver of this feature

1.Description	Variable soil consistency in CTC
2.Owner	Michael Brunke
3.Created	18 March 2020
4.Equ
5.Ver
6.Perf
7.Val
8.Approver
9.Approved Date

Title: Variable soil thickness in CTC

Requirements and Design

E3SM BGC  Group

Date: 18 March 2020  

Summary

Variable soil thickness was already implemented as an option in v1.  In order to properly run variable soil thickness with BGC, CTC needs to be made consistent with variable soil thickness if turned on.

Requirements

Requirement: Add variable soil thickness consistency in CTC

Date last modified: 18 March 2020  
Contributors: Michael Brunke

In order to properly run variable soil thickness with BGC, carbon, nitrogen, and phosphorus pools should be consistent with the number of soil layers defined by our soil thickness dataset (Pelletier et al. 2016).

Algorithmic Formulations

Design solution: Variable soil thickness consistency in CTC

Date last modified: 18 March 2020 
Contributors: Michael Brunke

Variable soil thickness has already been implemented into E3SMv1 as described in Brunke et al. (2016).  In order to properly run variable soil thickness with BGC, carbon, nitrogen, and phosphorus pools need to go down only to the number of layers defined per grid cell or topounit based upon our soil thickness dataset (Pelletier et al. 2016).

Design and Implementation

Implementation: Adding variable soil thickness consistency in CTC

Date last modified: 18 March 2020 
Contributors: Michael Brunke

Carbon, nitrogen, and phosphorus pools need to go down only to the number of layers defined per grid cell or topounit based upon our soil thickness dataset (Pelletier et al. 2016).  Also, checks will be implemented to prevent people from trying to run variable soil thickness for anything other than satellite phenology or CTC.  Grid cell mean soil thicknesses also need to be added to the surface input data.

Planned Verification and Unit Testing 

Verification and Unit Testing: Verifying soil thickness in CTC

Date last modified:  18 March 2020
Contributors: Michael Brunke

Unit testing will be done by running the E3SM land developers suite.

Carbon, nitrogen, and phosphorus pools need to go down only to the number of layers defined per grid cell or topounit based upon our soil thickness dataset (Pelletier et al. 2016).  Also, checks will be implemented to prevent people from trying to run variable soil thickness for anything other than satellite phenology or CTC.  Grid cell mean soil thicknesses also need to be added to the surface input data.

Point simulations have already been done.  The soil moisture results are what I expect from my work of implementing variable soil thickness in CLM4.5. At US-UMB, US-Ha1, and BR-Sa1, the soil moisture profiles are largely unchanged so that the carbon, nitrogen, and phosphorus pools are also unaffected.  However, as expected from my CLM4.5 results, the soil moisture profile at US-SRM in the Santa Rita Range near here is drier, especially from ~1880-1960 resulting in smaller vegetation and soil organic pools.  At AT-Neu, we see the expected larger soil moisture variations with a general moister profile which results in a larger soil organic pool.

Figure 1.  Time series of differences in the volumetric soil moisture profiles between simulations with (the standard runs with a constant 10 layers down to 3.8 m) and without variable soil thickness derived from our depth-to-bedrock dataset for point simulations at five flux tower locations in the 20th century transient.

Figure 2.  Time series of carbon (C), nitrogen (N), and phosphorus (P) pools for vegetation (solid lines), litter (dashed), and soil organic matter (dotted) from the standard runs with a constant 10 layers (down to 3.8 m, green lines) and the runs with variable soil thickness derived from our depth-to-bedrock dataset (red lines) in the 20th century transient.

Planned Validation Testing 

Validation Testing: Validating soil thickness in CTC

Date last modified: 18 March 2020
Contributors: Michael Brunke

Global simulations will be performed with only 1 topounit (grid cell average soil thickness) and multiple topounits (different soil thicknesses per topounit) as part of the implementation of elevation classes.  We expect soil moisture to be better represented in multiple topounits.

Planned Performance Testing 

Performance Testing: Soil thickness in CTC testing

Date last modified: 18 March 2020
Contributors: Michael Brunke

Variable soil thickness should have a minimal impact on the performance of the model.  This will be verified from the model’s timing tables.