4

I have a NC file (time, lat, lon) and I am trying to extracting time series of multiple stations (lat/lon points. So I tried it this way to read the coordinates and extract the nearest values from the NC file :

import pandas as pd
import xarray as xr
nc_file = r"C:\Users\lab\Desktop\harvey\example.nc"
NC = xr.open_dataset(nc_file)
csv = r"C:\Users\lab\Desktop\harvey\stations.csv"
df = pd.read_csv(csv,delimiter=',')
Newdf = pd.DataFrame([])
# grid point lists
lat = df["Lat"]
lon = df["Lon"]
point_list = zip(lat,lon)
for i, j in point_list:
    dsloc = NC.sel(lat=i,lon=j,method='nearest')
    DT=dsloc.to_dataframe()
    Newdf=Newdf.append(DT,sort=True)

The code works fine and returns this:

                        EVP     lat      lon
time                                        
2019-01-01 19:00:00  0.0546  40.063  -88.313
2019-01-01 23:00:00  0.0049  40.063  -88.313
2019-01-01 19:00:00  0.0052  41.938  -93.688
2019-01-01 23:00:00  0.0029  41.938  -93.688
2019-01-01 19:00:00  0.0101  52.938 -124.938
2019-01-01 23:00:00  0.0200  52.938 -124.938
2019-01-01 19:00:00  0.1644  39.063  -79.438
2019-01-01 23:00:00 -0.0027  39.063  -79.438

However, I need to associate the station-ID (from my original lat/long file) for each of the coordinates like this:

  Station-ID       Lat        Lon            time     EVP     lat      lon
0        Bo1  40.00620  -88.29040  1/1/2019 19:00  0.0546  40.063  -88.313
1                                  1/1/2019 23:00  0.0049  40.063  -88.313
2        Br1  41.97490  -93.69060  1/1/2019 19:00  0.0052  41.938  -93.688
3                                  1/1/2019 23:00  0.0029  41.938  -93.688
4        Brw  71.32250 -156.60917  1/1/2019 19:00  0.0101  52.938 -124.938
5                                  1/1/2019 23:00  0.0200  52.938 -124.938
6        CaV  39.06333  -79.42083  1/1/2019 19:00  0.1644  39.063  -79.438
7                                  1/1/2019 23:00 -0.0027  39.063  -79.438

Any thoughts how can merge my data frames them like the provided example?

1 Answer 1

8

What about if you include the station name in your zip command, and then insert the ID into the pandas dataframe line like this (by the way, I couldn't access your CSV file, so I simplified slightly the example with a dummy list).

import pandas as pd
import xarray as xr
nc_file = "example.nc"
NC = xr.open_dataset(nc_file)

#dummy locations and station id as I can't access the CSV
lat=[40,42,41]
lon=[-100,-105,-99]
name=["a","b","c"]

Newdf = pd.DataFrame([])

for i,j,id in zip(lat,lon,name):
    dsloc = NC.sel(lat=i,lon=j,method='nearest')
    DT=dsloc.to_dataframe()

    # insert the name with your preferred column title:
    DT.insert(loc=0,column="station",value=id)
    Newdf=Newdf.append(DT,sort=True)

print(Newdf)

This gives me:

                        EVP     lat      lon station
time                                                
2019-01-01 19:00:00  0.0527  39.938  -99.938       a
2019-01-01 23:00:00  0.0232  39.938  -99.938       a
2019-01-01 19:00:00  0.0125  41.938 -104.938       b
2019-01-01 23:00:00  0.0055  41.938 -104.938       b
2019-01-01 19:00:00  0.0527  40.938  -98.938       c
2019-01-01 23:00:00  0.0184  40.938  -98.938       c
Sign up to request clarification or add additional context in comments.

4 Comments

Nice idea that works! yes, zipping with an Id and inserting to the dataframe work perfectly. Thank you.
yeah, i'll once I got 15 reputations :) currently I vote but it's not shown till my 15th reputations !
Didn't know that! Welcome to stackexchange by the way 😉
The answer needs to be updated to reflect the fact that append has been deprecated. The second to last line should be ` Newdf=pd.concat([Newdf, DT], ignore_index=True)`

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.