Closed edoddridge closed 7 years ago
Merging #120 into master will increase coverage by
<.01%
. The diff coverage is100%
.
@@ Coverage Diff @@
## master #120 +/- ##
==========================================
+ Coverage 90.16% 90.17% +<.01%
==========================================
Files 5 5
Lines 1119 1120 +1
Branches 56 57 +1
==========================================
+ Hits 1009 1010 +1
Misses 90 90
Partials 20 20
Impacted Files | Coverage Δ | |
---|---|---|
aronnax.f90 | 92.67% <100%> (ø) |
:arrow_up: |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update e419283...cfd44c1. Read the comment docs.
The rather embarrassingly named
pysics-test
branch is now at the point where it provides some useful diagnostics and is a worthwhile addition to the master branch.However, because I'm still learning about all of this stuff it includes the commit history of the
external-matrix-solver
branch, and so is currently blocked on PR #89. Sorry.@axch I'm putting this in as a PR so that you can see what's been done. I think it's a useful addition as it stands, but there is definitely still work to be done on it.
Notably:
physics_tests.py
runs a number of simulations with varying values ofdx
in which the model is initialised with flat motionless layers. The wind forcing acts on a single grid cell in the centre of the domain and so the expected momentum can be easily calculated. This expected momentum is compared with the simulated momentum after one year of model time for the n+1/2 layer cases and 90 days in the n-layer cases. The error is then plotted as a function of resolution.This PR includes code for running tests with non-zero initial conditions to monitor the evolution of momentum in the absence of forcing. Those tests are not currently included in the documentation, but should be in the future. After some initial period of time, in which waves propagate and are dissipated by the boundary, the model conserves momentum very nicely.