Downloading CDEC Snow Depth Data

Created by Hannah Besso, July 2023

This notebook does the following:

• Load a region of interest

• Make a geodataframe of the coordinates of all CDEC stations within the roi

• Make a dataframe of snow depth data from a single CDEC station

• Make a dataframe containing snow depth data from all CDEC stations within the roi

• Convert from inches to meters and plot the timeseries of one station

Import necessary packages

import numpy as np
import geopandas as gpd
import pandas as pd
import matplotlib.pyplot as plt
from datetime import datetime
from shapely.geometry import Point
import metloom
from metloom.pointdata import CDECPointData
from metloom.variables import CdecStationVariables

Load USA states:

states_url = 'http://eric.clst.org/assets/wiki/uploads/Stuff/gz_2010_us_040_00_5m.json'
states_gdf = gpd.read_file(states_url)
california = states_gdf.loc[states_gdf.NAME == 'California']

from shapely.geometry import Polygon

coords = [(-120.03380085114448,38.2593960348446),
          (-120.03380085114448,36.934834847980355),
          (-118.1903872978364,36.934834847980355),
          (-118.1903872978364,38.2593960348446),
          (-120.03380085114448,38.2593960348446)]

polygon = Polygon(coords)

central_sierra = gpd.GeoDataFrame(index=[0], crs='epsg:4326', geometry=[polygon])

Access a geodataframe of CDEC station locations:

Using the California polygon to get a list of all stations, excluding snow courses

vrs = [
#    CdecStationVariables.SWE      # I want snow depth but SWE is also available
    CdecStationVariables.SNOWDEPTH
]
points = CDECPointData.points_from_geometry(central_sierra, vrs, snow_courses=False)
cdec_locations = points.to_dataframe()
cdec_locations.set_crs('epsg:4326')
cdec_locations.head()

	name	id	datasource	geometry
0	HUNTINGTON LAKE (USBR)	HNT	CDEC	POINT Z (-119.22048 37.22757 7000)
1	WHITE WOLF	WHW	CDEC	POINT Z (-119.65161 37.85944 7900)
2	GIN FLAT	GIN	CDEC	POINT Z (-119.77491 37.76689 7050)
3	LOWER RELIEF VALLEY	REL	CDEC	POINT Z (-119.76292 38.24292 8100)
4	MARIPOSA GROVE	MPG	CDEC	POINT Z (-119.6 37.5 6400)

fig, ax = plt.subplots()

california.plot(facecolor='none', edgecolor='k', ax=ax)
central_sierra.plot(facecolor='none', edgecolor='orange', ax=ax)
cdec_locations.plot(ax=ax);

Download snow depth data for a single CDEC station:

start_date = datetime(2020,10,1)
end_date = datetime(2023,7,31)

# Download data for TUM to get the datetime index:
cdec_point = CDECPointData("TUM", "TUM")
tum_depth = cdec_point.get_daily_data(
    start_date, end_date,
    [cdec_point.ALLOWED_VARIABLES.SNOWDEPTH]
)
tum_depth.head()

		geometry	SNOWDEPTH	SNOWDEPTH_units	datasource
datetime	site
2020-10-01 08:00:00+00:00	TUM	POINT Z (-119.34997 37.87622 8500)	0.0	INCHES	CDEC
2020-10-02 08:00:00+00:00	TUM	POINT Z (-119.34997 37.87622 8500)	0.0	INCHES	CDEC
2020-10-03 08:00:00+00:00	TUM	POINT Z (-119.34997 37.87622 8500)	0.0	INCHES	CDEC
2020-10-04 08:00:00+00:00	TUM	POINT Z (-119.34997 37.87622 8500)	1.0	INCHES	CDEC
2020-10-05 08:00:00+00:00	TUM	POINT Z (-119.34997 37.87622 8500)	0.0	INCHES	CDEC

Create a dataframe of snow depth data from every CDEC station within the ROI:

# Reindex from a multiindex so I can easily access the datetime index:
tum_depth = tum_depth.reset_index()
tum_depth = tum_depth.set_index('datetime')

# Create a new empty dataframe with the TUM datetime index:
snwd_df = pd.DataFrame()
snwd_df.index = pd.DatetimeIndex(data=tum_depth.index)

# Iteratively download all the data and add the SNOWDEPTH column to the snwd_df:
# Note that the snow depth unit is inches
for i in range(len(cdec_locations)):
    cdec_point = CDECPointData(cdec_locations.id.values[i], cdec_locations.id.values[i])
    cdec_depth = cdec_point.get_daily_data(
        start_date, end_date,
        [cdec_point.ALLOWED_VARIABLES.SNOWDEPTH]
    )
    if cdec_depth is not None:
        cdec_depth = cdec_depth.reset_index()
        cdec_depth = cdec_depth.set_index('datetime')
        snwd_df[f'{cdec_locations.id.values[i]}'] = cdec_depth['SNOWDEPTH']
    else:
        continue

snwd_df.head()

	HNT	WHW	GIN	REL	MPG	GRV	MHP	KSP	AGP	GRM	...	MLK	KUP	DPO	PDS	HRS	BCB	TMR	SLK	SLI	GEM
datetime
2020-10-01 08:00:00+00:00	2.0	0.0	1.0	0.0	25.208333	1.0	1.0	0.0	0.0	NaN	...	64.750000	0.0	1.0	4.0	NaN	0.0	2.0	10.0	NaN	NaN
2020-10-02 08:00:00+00:00	1.0	0.0	0.0	0.0	11.913043	1.0	0.0	1.0	0.0	NaN	...	65.458333	0.0	1.0	4.0	NaN	0.0	2.0	14.0	NaN	NaN
2020-10-03 08:00:00+00:00	1.0	0.0	0.0	0.0	25.208333	1.0	NaN	1.0	0.0	NaN	...	56.666667	0.0	1.0	1.0	NaN	0.0	2.0	14.0	NaN	NaN
2020-10-04 08:00:00+00:00	1.0	0.0	0.0	0.0	20.541667	1.0	0.0	1.0	0.0	NaN	...	40.208333	0.0	1.0	4.0	NaN	0.0	2.0	13.0	NaN	NaN
2020-10-05 08:00:00+00:00	1.0	0.0	0.0	0.0	11.458333	1.0	0.0	1.0	0.0	NaN	...	47.333333	0.0	1.0	3.0	NaN	0.0	2.0	12.0	NaN	NaN

5 rows × 41 columns

# convert from inches to meters
snwd_df_m = snwd_df * 0.0254

fig, ax = plt.subplots()

snwd_df_m['DAN'].plot(ax=ax)
ax.set_ylabel('Snow Depth (m)')

Text(0, 0.5, 'Snow Depth (m)')