pydaymet 0.11.3

Access the climate data from the Daymet database through its RESTful service.

Package	Description	Status
PyNHD	Navigate and subset NHDPlus (MR and HR) using web services
Py3DEP	Access topographic data through National Map’s 3DEP web service
PyGeoHydro	Access NWIS, NID, HCDN 2009, NLCD, and SSEBop databases
PyDaymet	Access Daymet for daily climate data both single pixel and gridded
AsyncRetriever	High-level API for asynchronous requests with persistent caching
PyGeoOGC	Send queries to any ArcGIS RESTful-, WMS-, and WFS-based services
PyGeoUtils	Convert responses from PyGeoOGC’s supported web services to datasets

PyDaymet: Daily climate data through Daymet

Features

PyDaymet is a part of HyRiver software stack that is designed to aid in watershed analysis through web services. This package provides access to climate data from Daymet V4 database using NetCDF Subset Service (NCSS). Both single pixel (using get_bycoords function) and gridded data (using get_bygeom) are supported which are returned as pandas.DataFrame and xarray.Dataset, respectively. Climate data is available for North America, Hawaii from 1980, and Puerto Rico from 1950 at three time scales: daily, monthly, and annual. Additionally, PyDaymet can compute Potential EvapoTranspiration (PET) using three methods: penman_monteith, priestley_taylor, and hargreaves_samani for both single pixel and gridded data.

To fully utilize the capabilities of the NCSS, under-the-hood, PyDaymet uses AsyncRetriever for retrieving Daymet data asynchronously with persistent caching. This improves the reliability and speed of data retrieval significantly.

You can try using PyDaymet without installing it on you system by clicking on the binder badge below the PyDaymet banner. A Jupyter notebook instance with the stack pre-installed will be launched in your web browser and you can start coding!

Please note that since this project is in early development stages, while the provided functionalities should be stable, changes in APIs are possible in new releases. But we appreciate it if you give this project a try and provide feedback. Contributions are most welcome.

Moreover, requests for additional functionalities can be submitted via issue tracker.

Installation

You can install PyDaymet using pip after installing libgdal on your system (for example, in Ubuntu run sudo apt install libgdal-dev):

$ pip install pydaymet

Alternatively, PyDaymet can be installed from the conda-forge repository using Conda:

$ conda install -c conda-forge pydaymet

Quick start

You can use PyDaymet using command-line or as a Python library. The commanda-line provides access to two functionality:

• Getting gridded climate data: You must create a geopandas.GeoDataFrame that contains the geometries of the target locations. This dataframe must have four columns: id, start, end, geometry. The id column is used as filenames for saving the obtained climate data to a NetCDF (.nc) file. The start and end columns are starting and ending dates of the target period. Then, you must save the dataframe as a shapefile (.shp) or geopackage (.gpkg) with CRS attribute.

• Getting single pixel climate data: You must create a CSV file that contains coordinates of the target locations. This file must have at four columns: id, start, end, lon, and lat. The id column is used as filenames for saving the obtained climate data to a CSV (.csv) file. The start and end columns are the same as the geometry command. The lon and lat columns are the longitude and latitude coordinates of the target locations.

$ pydaymet -h
Usage: pydaymet [OPTIONS] COMMAND [ARGS]...

Command-line interface for PyDaymet.

Options:
-h, --help  Show this message and exit.

Commands:
coords    Retrieve climate data for a list of coordinates.
geometry  Retrieve climate data for a dataframe of geometries.

The coords sub-command is as follows:

$ pydaymet coords -h
Usage: pydaymet coords [OPTIONS] FPATH

Retrieve climate data for a list of coordinates.

FPATH: Path to a csv file with four columns:
    - ``id``: Feature identifiers that daymet uses as the output netcdf filenames.
    - ``start``: Start time.
    - ``end``: End time.
    - ``lon``: Longitude of the points of interest.
    - ``lat``: Latitude of the points of interest.
    - ``time_scale``: (optional) Time scale, either ``daily`` (default), ``monthly`` or ``annual``.
    - ``pet``: (optional) Method to compute PET. Suppoerted methods are:
               ``penman_monteith``, ``hargreaves_samani``, ``priestley_taylor``, and ``none`` (default).
    - ``alpha``: (optional) Alpha parameter for Priestley-Taylor method for computing PET. Defaults to 1.26.

Examples:
    $ cat coords.csv
    id,lon,lat,start,end,pet
    california,-122.2493328,37.8122894,2012-01-01,2014-12-31,hargreaves_samani
    $ pydaymet coords coords.csv -v prcp -v tmin

Options:
-v, --variables TEXT  Target variables. You can pass this flag multiple
                        times for multiple variables.

-s, --save_dir PATH   Path to a directory to save the requested files.
                        Extension for the outputs is .nc for geometry and .csv
                        for coords.

-h, --help            Show this message and exit.

And, the geometry sub-command is as follows:

$ pydaymet geometry -h
Usage: pydaymet geometry [OPTIONS] FPATH

Retrieve climate data for a dataframe of geometries.

FPATH: Path to a shapefile (.shp) or geopackage (.gpkg) file.
This file must have four columns and contain a ``crs`` attribute:
    - ``id``: Feature identifiers that daymet uses as the output netcdf filenames.
    - ``start``: Start time.
    - ``end``: End time.
    - ``geometry``: Target geometries.
    - ``time_scale``: (optional) Time scale, either ``daily`` (default), ``monthly`` or ``annual``.
    - ``pet``: (optional) Method to compute PET. Suppoerted methods are:
               ``penman_monteith``, ``hargreaves_samani``, ``priestley_taylor``, and ``none`` (default).
    - ``alpha``: (optional) Alpha parameter for Priestley-Taylor method for computing PET. Defaults to 1.26.

Examples:
    $ pydaymet geometry geo.gpkg -v prcp -v tmin

Options:
-v, --variables TEXT  Target variables. You can pass this flag multiple
                        times for multiple variables.

-s, --save_dir PATH   Path to a directory to save the requested files.
                        Extension for the outputs is .nc for geometry and .csv
                        for coords.

-h, --help            Show this message and exit.

Now, let’s see how we can use PyDaymet as a library.

PyDaymet offers two functions for getting climate data; get_bycoords and get_bygeom. The arguments of these functions are identical except the first argument where the latter should be polygon and the former should be a coordinate (a tuple of length two as in (x, y)). The input geometry or coordinate can be in any valid CRS (defaults to EPSG:4326). The dates argument can be either a tuple of length two like (start_str, end_str) or a list of years like [2000, 2005]. It is noted that both functions have a pet flag for computing PET. Additionally, we can pass time_scale to get daily, monthly or annual summaries. This flag by default is set to daily.

from pynhd import NLDI
import pydaymet as daymet

geometry = NLDI().get_basins("01031500").geometry[0]

var = ["prcp", "tmin"]
dates = ("2000-01-01", "2000-06-30")

daily = daymet.get_bygeom(geometry, dates, variables=var, pet="priestley_taylor")
monthly = daymet.get_bygeom(geometry, dates, variables=var, time_scale="monthly")

If the input geometry (or coordinate) is in a CRS other than EPSG:4326, we should pass it to the functions.

coords = (-1431147.7928, 318483.4618)
crs = "epsg:3542"
dates = ("2000-01-01", "2006-12-31")
annual = daymet.get_bycoords(coords, dates, variables=var, loc_crs=crs, time_scale="annual")

Also, we can use the potential_et function to compute PET by passing the daily climate data. We can either pass a pandas.DataFrame or a xarray.Dataset. Note that, penman_monteith and priestley_taylor methods have parameters that can be passed via the params argument, if any value other than the default values are needed. For example, default value of alpha for priestley_taylor method is 1.26 (humid regions), we can set it to 1.74 (arid regions) as follows:

pet_hs = daymet.potential_et(daily, methods="priestley_taylor", params={"alpha": 1.74})

Next, let’s get annual total precipitation for Hawaii and Puerto Rico for 2010.

hi_ext = (-160.3055, 17.9539, -154.7715, 23.5186)
pr_ext = (-67.9927, 16.8443, -64.1195, 19.9381)
hi = daymet.get_bygeom(hi_ext, 2010, variables="prcp", region="hi", time_scale="annual")
pr = daymet.get_bygeom(pr_ext, 2010, variables="prcp", region="pr", time_scale="annual")

Some example plots are shown below:

Contributing

Contributions are very welcomed. Please read CONTRIBUTING.rst file for instructions.

Credits

Credits to Koen Hufkens for his implementation of accessing the Daymet RESTful service, daymetpy.