A Python positive degree day model for glacier surface mass balance.
This module provides a simple model to compute accumulation and melt on a glacier using near-surface air temperature and precipitation time series. The model assumes that melt is proportional to the number of positive degree-days, which corresponds to the integral of temperature above 0°C. Temperature variability is included by assuming a normal temperature distribution around the mean. The model optionally includes refreezing of melted snow and ice at the glacier surface.
PyPDD can be used as a module within Python to operate on Numpy arrays. In addition, it reads and writes netCDF files directly from the command line, and provides a raster module for GRASS GIS. The PDD model is based on an algorithm that was initially developed for the Parallel Ice Sheet Model and adopted here with very few changes.
Installation:
pip install pypdd
A PDD model instance can be created by:
from pypdd import PDDModel pdd = PDDModel()
Several model parameters can be set at initialization. See help(PDDModel)
for a list. Provided two arrays temp
and prec
of shape (t, x, y)
containing temperature and precipitation data, the PDD model can be called
with:
pdd(temp, prec)
This will return a dictionary containing a number of two- and three-dimensional
arrays, including the number of positive degree day 'pdd'
and total surface
mass balance 'smb'
. Temperature variability can be included in a third array
stdv
containing temperature standard deviation values:
pdd(temp, prec, stdv)
If any of temp
, prec
, or stdv
has shape (x, y)
, it will be
interpreted as constant in time and expanded along the time dimension. Floats
with be interpreted as constant in time and space and expanded along all
dimensions.
Requires: netCDF4-Python.
The PDDModel class holds a netCDF operator, which can be called by:
pdd.nco('input.nc', 'output.nc')
The file 'input.nc'
should contain temperatures and precipitation in
variables 'temp'
and 'prec'
. The calculated number of positive degree
days and total surface mass balance are stored in variables 'pdd'
and
'smb'
of 'output.nc'
. Keyword argument output_size
or
output_variables
can be used to produce more output.
The netCDF interface can be used directly from the command line by executing the module as a script:
python pypdd.py -i 'input.nc' -o 'output.nc'
If no input file is provided, an artificial climate will be generated under
atm.nc
and used by the model. By default, output is saved as smb.nc
.
Many more command-line options are available. For an overview type:
python pypdd.py --help
Requires: GRASS GIS.
PyPDD can also operate on GRASS raster maps using the attached module r.pdd
.
Temperature, precipitation and standard deviation maps should be provided as
comma-separated lists:
r.pdd.py temp=list,of,temp,maps prec=list,of,prec,maps pdd=pdd_map smb=smb_map
All time-independent PyPDD output variables can currently be exported as raster
maps. Alike any other GRASS module, a graphical prompt can be invoked by calling
r.pdd
without arguments, and a list of options can be obtained with:
r.pdd.py --help
Applications of PyPDD:
- N. Gandy, L. J. Gregoire, J. C. Ely, C. D. Clark, D. M. Hodgson, V. Lee, T. Bradwell, and R. F. Ivanovic. Marine ice sheet instability and ice shelf buttressing of the Minch Ice Stream, northwest Scotland. The Cryosphere, 12(11):3635--3651, https://doi.org/10.5194/tc-12-3635-2018, 2018.
- A. Plach, K. H. Nisancioglu, S. Le clec'h, A. Born, P. M. Langebroek, C. Guo, M. Imhof, and T. F. Stocker. Eemian Greenland SMB strongly sensitive to model choice. Clim. Past, 14:1463--1485, https://doi.org/10.5194/cp-14-1463-2018, 2018.
- J. Seguinot. Spatial and seasonal effects of temperature variability in a positive degree-day glacier surface mass-balance model. J. Glaciol., 59(218):1202--1204, http://doi.org/10.3189/2013JoG13J081, 2013.