Using Daymet data from ORNL webservice

Daymet is a a daily gridded weather dataset distributed through the ORNL DAAC https://daymet.ornl.gov/

As part of their tools for obtaining data they have created a webservice that allows one to download a timeseries of temperature and precipitation for any location in the US.

Functions in Daymetpy allow for painless use of this service DaymetPy package developed by Koen Hufkens https://bitbucket.org/khufkens/daymetpy



In [1]:

    
import sys
sys.path.append(r"..")

import daymetpy



In [2]:

    
ornl_lat, ornl_long = 35.9313167, -84.3104124
df = daymetpy.daymet_timeseries(lon=ornl_long, lat=ornl_lat, start_year=2012, end_year=2013)

Which gives us a nice data frame with weather data for the Oak Ridge National Lab



In [3]:

    
df.head()

Which we can visualize using matplotlib and seaborn



In [4]:

    
import pandas as pd

import seaborn as sns
import matplotlib.pyplot as plt



In [5]:

    
%matplotlib inline
fig, (ax1, ax2) = plt.subplots(2, figsize=(18, 10), sharex=True)
rolling3day = pd.rolling_mean(df, 15)
ax1.fill_between(rolling3day.index, rolling3day.tmin, rolling3day.tmax, alpha=0.5, lw=0)
ax1.plot(df.index, df[['tmax', 'tmin']].mean(axis=1), lw=2, alpha=0.5)

ax1.set_title('Daymet temp at ORNL', fontsize=20)
ax1.set_ylabel(u'Temp. (°C)', fontsize=20)

monthlysum = df.resample("M", how='sum')
ax2.bar(monthlysum.index, monthlysum.prcp, width=20,)
ax2.set_title('Daymet precip at ORNL', fontsize=20)
ax2.set_ylabel(u'Precip. (mm)', fontsize=20)

fig.tight_layout()









    



C:\Anaconda\lib\site-packages\ipykernel\__main__.py:3: FutureWarning: pd.rolling_mean is deprecated for DataFrame and will be removed in a future version, replace with 
	DataFrame.rolling(window=15,center=False).mean()
  app.launch_new_instance()
C:\Anaconda\lib\site-packages\ipykernel\__main__.py:10: FutureWarning: how in .resample() is deprecated
the new syntax is .resample(...).sum()

And ask questions such as: "What does the difference in temperature between Denver and Miami look like?"



In [6]:

    
denver_loc = (-104.9903, 39.7392)
miami_loc = (-80.2089, 25.7753)

denver = daymetpy.daymet_timeseries(lon=denver_loc[0], lat=denver_loc[1], start_year=2012, end_year=2014)
miami = daymetpy.daymet_timeseries(lon=miami_loc[0], lat=miami_loc[1], start_year=2012, end_year=2014)



In [7]:

    
fig, ax1 = plt.subplots(1, figsize=(18, 10))
rolling3day = pd.rolling_mean(denver, 15)
ax1.fill_between(rolling3day.index, rolling3day.tmin, rolling3day.tmax, 
                 alpha=0.4, lw=0, label='Denver', color=sns.xkcd_palette(['faded green'])[0])

ax1.set_title('Denver vs Miami temps (15 day mean)', fontsize=20)

rolling3day = pd.rolling_mean(miami, 15)
ax1.fill_between(rolling3day.index, rolling3day.tmin, rolling3day.tmax, 
                 alpha=0.4, lw=0, label='Miami', color=sns.xkcd_palette(['dusty purple'])[0])

ax1.set_ylabel(u'Temp. (°C)', fontsize=20)
fig.tight_layout()
plt.legend(fontsize=20)









    



C:\Anaconda\lib\site-packages\ipykernel\__main__.py:2: FutureWarning: pd.rolling_mean is deprecated for DataFrame and will be removed in a future version, replace with 
	DataFrame.rolling(window=15,center=False).mean()
  from ipykernel import kernelapp as app
C:\Anaconda\lib\site-packages\ipykernel\__main__.py:8: FutureWarning: pd.rolling_mean is deprecated for DataFrame and will be removed in a future version, replace with 
	DataFrame.rolling(window=15,center=False).mean()






    Out[7]:





<matplotlib.legend.Legend at 0xb5f22e8>

	year	yday	dayl	prcp	srad	tmax	tmin	vp
2012-01-01	2012	1	34560	3.0	204.8	15.0	0.5	640.0
2012-01-02	2012	2	34560	2.0	278.4	11.5	-3.0	480.0
2012-01-03	2012	3	34560	0.0	204.8	1.5	-7.0	360.0
2012-01-04	2012	4	34560	0.0	252.8	3.0	-8.5	320.0
2012-01-05	2012	5	34906	0.0	288.0	9.5	-5.5	400.0