Correcting Coarse-Resolution Weather and Climate Models by Machine
Learning from Global Storm-Resolving Simulations
Abstract
Global atmospheric ‘storm-resolving’ models with horizontal grid spacing
of less than 5~km resolve deep cumulus convection and
flow in complex terrain. They promise to be reference models that could
be used to improve computationally affordable coarse-grid global climate
models across a range of climates, reducing uncertainties in regional
precipitation and temperature trends. Here, machine learning of nudging
tendencies as functions of column state is used to correct the physical
parameterization tendencies of temperature, humidity, and optionally
winds, in a real-geography coarse-grid model (FV3GFS with a 200 km grid)
to be closer to those of a 40-day reference simulation using X-SHiELD, a
modified version of FV3GFS with a 3 km grid. Both simulations specify
the same historical sea-surface temperature fields. This methodology
builds on a prior study using a global observational analysis as the
reference. The coarse-grid model without machine learning corrections
has too little cloud, causing too much daytime heating of land surfaces
that creates excessive surface latent heat flux and rainfall. This bias
is avoided by learning downwelling radiative flux from the fine-grid
model. The best configuration uses learned nudging tendencies for
temperature and humidity but not winds. Neural nets slightly outperform
random forests. Forecasts of 850 hPa temperature gain 18 hours of skill
at 3-7 day leads and time-mean precipitation patterns are improved 30%
by applying the ML correction. Adding machine-learned wind tendencies
improves 500 hPa height skill for the first five days of forecasts but
degrades time-mean upper tropospheric temperature and zonal wind
patterns thereafter. The figure shows maps of 30-day time-mean
precipitation pattern difference from the fine-grid reference for
prognostic simulations: (a) 200 km baseline(no machine learning
correction) (b) Using random forest correction and (c) neural net
correction for temperature. humidity and surface radiation corrections.
RMSE is the root mean squared precipitation difference from the
reference, which is 30% less for the two machine-learning corrected
simulations compared to the baseline. (d) Bar charts of the land-mean,
ocean-mean and global-mean precipitation biases for these three
configurations, showing the machine-learning corrected simulations
remove a high bias of land surface precipitation in the baseline
simulation.