Use Case 4: Rain-on-snow events
N.B. For this exercise, one might need to install the bottleneck package
$ conda install bottleneck
In this example we will consider more than one variable in the statistical calculation, by estimating the frequency of rainfall days when there is snow on the ground – so called rain-on-snow events.
Typical climate model output variables will be used, namely total precipitation (pr), solid precipitation (prsn) and surface snow amount (water equivalent; snw). Other variables (and temporal resolutions) could be used, for example the liquid precipitation. However, we use this set of variables to test certain processing functions in RCAT.
STEP 1: Data input
Similarly as for the other use cases, specify the model data to use. We will not use any observations in this example.
model_1 = {
'fpath': '<path-to-folder-1>',
'grid type': 'reg', 'grid name': '<grid-name>',
'start year': <start year>, 'end year': <end year>, 'months': [10,11,12,1,2,3]
'date interval start': '<yyyy>-<mm>', 'date interval end': '<yyyy>-<mm>',
'chunks_time': {'time': -1}, 'chunks_x': {'x': -1}, 'chunks_y': {'y': -1},
}
model_2 = {
'fpath': '<path-to-folder-2>',
'grid type': 'reg', 'grid name': '<grid-name>',
'start year': <start year>, 'end year': <end year>, 'months': [10,11,12,1,2,3]
'date interval start': '<yyyy>-<mm>', 'date interval end': '<yyyy>-<mm>',
'chunks_time': {'time': -1}, 'chunks_x': {'x': -1}, 'chunks_y': {'y': -1},
}
STEP 2: Variables
Set output directory under the SETTINGS section.
In the key variables we specify the three variables. Precipitation variables are available with daily resolution while snw is available on 3hr resolution.
We skip regional estimates, i.e. set ``regions = ``.
output dir = <path-to-output-directory>
variables = {
'pr': {'freq': 'day',
'units': 'mm',
'scale factor': 86400,
'offset factor': None,
'accumulated': False,
'obs': None,
'obs scale factor': None,
'obs freq': None,
'var names': None,
'regrid to': None,
'regrid method': 'bilinear'},
'prsn': {'freq': 'day',
'units': 'mm',
'scale factor': 86400,
'offset factor': None,
'accumulated': False,
'obs': None,
'obs scale factor': None,
'obs freq': None,
'var names': None,
'regrid to': None,
'regrid method': 'bilinear'},
'snw': {'freq': '3hr',
'units': 'mm',
'scale factor': 86400,
'offset factor': None,
'accumulated': False,
'obs': None,
'obs scale factor': None,
'obs freq': None,
'var names': None,
'regrid to': None,
'regrid method': 'bilinear'},
}
regions =
STEP 3: Set the variable modification option
We will test the variable modification option here by calculating the rainfall from total and solid precipitation. It is set as a dictionary with the key as the name of the new created variable:
variable modification = {
'prrain': {
'models': 'all',
'obs': None,
'input': {'x': 'pr', 'y': 'prsn'},
'expression': 'x - y',
'replace': True},
}
In the dictionary, you should specify for which ‘models’ and ‘obs’ the new variable will be created. Set specific names, ‘all’ to do it for all models/obs or None to not do it for any model/obs.
In the ‘input’ key you will specify algebraic notations for the variables included in the creation of the new variable, and in the ‘expression’ key one will provide the mathematical expression (using the algebraic notation). Since we want to get the liquid precipitation we simply take the difference between total and solid precipitation, and call the new variable prrain. The replace switch (boolean) is set to True if the new variable will replace the variables used to create the new one. If set to False, these will be kept.
STEP 4: Select statistics
Use the moment stat in the STATISTICS section. Its default options are found in the default_stats_config function in RCAT Statistics.
The statistical calculation will be performed for the new prrain variable only using conditional selection based on snow amounts (which will be resampled to daily mean values). The criteria for rain-on-snow events are here set to daily rainfall amounts of 5 mm or more occurring over gridpoints with snow on the ground corresponding to 10 mm of snow water equivalent:
'moment stat': {'vars': ['prrain'], 'moment stat': ['all', 'count'], 'thr': {'prrain': 5},
'cond analysis': {'prrain': {
'cvar': 'snw',
'resample resolution': ['D', 'mean'],
'type': 'static',
'operator': '>=',
'value': 10}}
},
Note that moment stat is set to ['all', 'count'] which here means
summation of all the true values (where conditions are fulfilled) per gridpoint.
STEP 5: Plotting
Under PLOTTING, validation plot should be set to True to enable plotting. Plot a map of the results by setting
validation plot = True
moments plot config = {'plot type': 'map'}
map projection = 'LambertConformal'
map configuration = {
'central_longitude': 10,
'central_latitude': 60.6,
'standard_parallels': (60.6, 60.6),
}
map extent = [3, 16, 42, 51] # Extent of the map; [lon_start, lon_end, lat_start, lat_end]
map gridlines = False
map grid config = {'axes_pad': 0.3, 'cbar_mode': 'each', 'cbar_location': 'right',
'cbar_size': '4%%', 'cbar_pad': 0.05}
map plot kwargs = {'filled': True, 'mesh': True}
map model domain =
STEP 6: Configure cluster
cluster type = slurm
nodes = 10
cluster kwargs = {'walltime': '02:00:00'}
STEP 7: Run RCAT
To run the analysis run from terminal (see Run RCAT in RCAT Configuration):
python <path-to-RCAT>/src/rcatool/runtime/RCAT_main.py -c config_main.ini
Output statistics and image files will be located under the user-defined output directory in the stats and imgs sub-folders respectively