If you use Matlab you can use the commands ncdisp and ncread. With the command ncdisp you can obtain information about the file (attributes, variables, dimensions, units, etc.). And with the command ncread you can extract data for a given variable.

Otherwise, if you use R, there is a netcdf package (http://cran.r-project.org/web/packages/RNetCDF/index.html). Then, I think the commands to use would be file.inq.nc and var.get.nc. (NB: I haven't used it so I cannot give you advice about it).

I show you an example using Matlab. I downloaded the file 'pr_day_CNRM-CM5_rcp45_r1i1p1_20260101-20301231.nc' from the project CMIP5 (precipitation in 2026-2030 according to the RCP4.5 scenario and CNRM-CM5 model).

I save the file name

>> file = 'pr_day_CNRM-CM5_rcp45_r1i1p1_20260101-20301231.nc';

Then, to display the file properties:

>> ncdisp(file)

Then you can know that the variables you are interested in are coded as lat (latitude), lon (longitude), time (time), and pr (precipitation). lat is a 128x1 vector, lon is a 256x1 vector, and time is a 1826x1 vector and pr is a 256x128x1826 array with the dimensions lon, lat, time.

With the following commands you can extract the longitude, latitude and time values of all the grid points:

>> longitude = ncread(file,'lon');

>> latitude = ncread(file,'lat');

>> time = ncread(file,'time');

You can use the same instruction (with some additional input arguments) to extract precipitation data for a selection of coordinates and times:

vardata = ncread(source,varname,start,count,stride)

source is the file name

varname is the variable name

start is from where you want to start to read data for each dimension (in vector form)

count is how many data points do you want to read for each dimension (in vector form)

stride is the frequency of extraction data, i.e. each day, each two days, etc. (in vector form)

So, if you want to extract precipitation every day for the full period 2026-2030 between latitudes 39.9218ºN and 52.5286ºN, and between longitudes 0º and 11.25ºE you would have to type:

Precip = ncread(file,'pr',[1 93 1], [9 10 Inf], [1 1 1]);

[1 93 1] is start, since longitude(1) = 0, latitude(93) = 39.9218, and you start at the first time measurement

[9 10 Inf] is count, since you want to extract data for 9 longitudes (longitude(9) = 11.25), for 10 latitudes (latitude(102) = 52.5286), and you use Inf to indicate you want all time points

[1 1 1] is stride, since you want to extract each longitude point, each latitude point and each time point

I hope this helps.

If you are working on a Linux/Unix OS then you can use Climate Data Operators (CDO, https://code.zmaw.de/projects/cdo). This is a command line based software for working with netCDF files, it is widely used in climate sciences. It can be used to manipulate climate data in netCDF format and calculate statistics. Let me know if you would like to use CDO and I can give some more detailed instructions.

Thank you so much Jordi Prats, that was exactly what i want :)  .. i am going to use MatLab to extract the data. but from your experience how i should select models from CMIP5 to download . i mean the one that i downloaded was a try to practice on downscaling but for my thesis i have to download more than one CMIP5 models output . so what do u think the basic criteria to select GCMs form CMIP5??

Thank you so much Sebastian Engelstaedter ,no i am not working on a Linux/unix so i can't work on CDO.

but i read about what called NetCDf viewers and manipulators like these in the website http://www.unidata.ucar.edu/software/netcdf/software.html. the most popular one of them is ncview ( http://meteora.ucsd.edu/~pierce/ncview_home_page.html ) is working under Linux but it is said that it could be used under windows too,  after installing other supporting software. so do you have idea if such software will be helpful to extract time series for specific location besides visualizing them or they are just viewers?? . and if they are helpful , which software  you recommend ??  

NCview is a very basic tool. It is good for having a quick look at a netCDF file. But it is not very useful for data analysis.

Another popular netCDF file viewer is call Panoply developed by NASA (http://www.giss.nasa.gov/tools/panoply). It is Java based and should work on all operating systems. But again it does not allow you to actually analyse your data.

If you want to calculate spatial and temporal statistics or extract time series at specific locations you need more sophisticated software such as Matlab or IDL. Both are commercial software packages and involve some code writing.

Hope this helps,

Sebastian

Yes yes that was so helpful and answered my queries about netCDF data analysis .

I am going to use MatLab because I worked with before and use Panoply to  make sense of the CGMs ouputs.

Thank you so much

on top of all the good answers you got - you can use from the prompt in a regular terminal with the right flags to ncdump > file. See the man pages for ncdump, but then you can read the file you created, and you have a nice, efficient, answer! 

All the best, Gidon

The ncdump is a tool in the Netcdf package in unidata  or it is included in the Matlab ??.. can you provide me a link to download the ncdump tool?? .. 

it IS a unidata tool, yes, see https://www.unidata.ucar.edu/software/netcdf/docs/netcdf/ncdump.html for details. matlab has now ncread, which you can also use, a good point!

Thank you very much 

ahalan ya Alham, any time, a pleasure, esp. for a fellow Middle Easterner!!

cool !!, you know some Arabic :) !, its my pleasure too .

Dear Alham,

the question of how to select GCM projections is a difficult one. I am a user of GCM simulations, not really an expert on this, but as far as I know there is not much consense about how to choose them.

The one general piece of advice you find is: take as many as possible in order to include the potential range of climate behaviours in your subset.

Then, it depends also on the objective of your study...

The following references might be helpful:

• http://www.pacificclimate.org/sites/default/files/publications/Murdock.ScenariosGuidance.Dec2011.pdf
• http://onlinelibrary.wiley.com/doi/10.1111/cobi.12163/abstract

regards

Its very simple only if you have ArcGIS Software.

Best Wishes

Dear Dheeraj Kumar , yes i am a user of ArcGIS and that was pretty wonderful and easy , i applied it and worked very well :D. 

Thank you very very much

Alham 

That really great, Thank you so much Dheeraj Kumar

Dear Dheeraj Kumar, iam using the ArcGIS make NetCDF table view to extract cordex dataset for temperature and precipitation as you described above. However, I don't know how I can extract for more values of lat and long. E.g. lat, 11.18, 10.60,7.68, 12.05,7.80 and long 7.58, 7.45, 8.61,8.53,6.73). You suggested that for dimension values---Lat and long (one can give more values). Please let me know how I can extract for more values at once. Thanks

Dear Dheeraj Kumar, iam using the ArcGIS make NetCDF table view to extract cordex dataset for temperature and precipitation as you described above. However, I don't know how I can extract for more values of lat and long. E.g. lat, 11.18, 10.60,7.68, 12.05,7.80 and long 7.58, 7.45, 8.61,8.53,6.73). You suggested that for dimension values---Lat and long (one can give more values). Please let me know how I can extract for more values at once. Thanks

Dear Dheeraj Kumar, iam using the ArcGIS make NetCDF table view to extract cordex dataset for temperature and precipitation as you described above. However, I don't know how I can extract for more values of lat and long. E.g. lat, 11.18, 10.60,7.68, 12.05,7.80 and long 7.58, 7.45, 8.61,8.53,6.73). You suggested that for dimension values---Lat and long (one can give more values). Please let me know how I can extract for more values at once. Thanks

Dheeraj Kumar  what will be the units because the file generated from arcmap gives unknown units

 Thanks, it was very useful

I want to download monthly climate parameters from NCEF project using .nc files.

This is a late response, but I am posting it anyways so further interested people can benefit.

One other way to extract a time series from a netcdf using R is provided by the raster package. Here is an example:

-----------------------------

library(raster)

# Command brick reads all layers (time slices) in the file

b

Thanks

Thiago Veloso dos Santos,

Its really very helpful.

Thanks, Dheeraj Kumar.

You provided detail steps are very useful for ArcGIS users.

Hi Alham,  Hi Guys,

I have a question more or less related to your discussion. Please, where can we download CIMP5 data  (precipitation and temperture) for 2030 at a spatial resolution of 30 second?

Many thanks

Hicham

Hi Hicham,

CORDEX regionalized projections made from CMIP5 data can be downloaded from the ESGF (Earth System Grid Federation) at a resolution of 0.44º. See here: https://esgf-node.ipsl.upmc.fr/projects/esgf-ipsl/.

https://esgf-node.ipsl.upmc.fr/projects/esgf-ipsl/

Hi All,

Great discussion - thanks to everyone for their contributions. I have a related follow up question:

Is there a way to extract a certain region with the downloading script so I don't have to download the entire globe for all the models/time frames I'm interested in? 

Thanks in advance,

Jonas

try this :

cdo -outputtab,date,lon,lat,value -remapnn,lon=-68_lat=-18 input_file.nc output_file.nc

I dont think the remapnn in CDO gives the exact point information. but the R code given by Thiago gives the right value.

well you can use GrADS :

set lat ..

set lon ..

set t 'time'

set gxout print

set prnopts %6.2 f 1 1

d variable

a= sublin(result,2)

b= subwrd(a,1)

res=write('data.txt',time' 'b,append)

and you will obtain your exact point information.

Dear Jordi Prats,

I have used the commands exactly as you suggested above to read radiation data from climate chemistry models but an error occured. Do you have any idea ?

Error using netcdflib

The NetCDF library encountered an error during execution of 'getVarsFloat' function - 'Index exceeds

dimension bound (NC_EINVALCOORDS)'

Hi

You can use Netcdf-Extractor for CMIP5 data there is in this site

https://www.AgriMetSoft.com

Hi

Netcdf-Extractor is a user friendly software that you can use for all Netcdf file but for CMIP5 data you can use coordinate.

http://www.agrimetsoft.com 

if you use Linux OS you can use CDO

to Extract single point near your location

$cdo -remapnn,lon=38.2834_lat=11.7876 input.nc output.nc

when you need to change it to ascii

$cdo -outputkey,year,month,day,value output.nc > new_out.txt

https://www.redmine.org/guide

the same opinion with Prof. Dheeraj Kumar 

If you know QGIS - a free open source  software - you can use climate model data in the form of netcdf and extract it through Netcdf browser- available in Plugins in QGIS. 

I have also similar question.

Currently, I am using ArcGIS to extract climate data from netcdf files. However, I can't extract the data for different grid points (latitude from 7.04 to 8.8 and longitude 37.84 to 39.60) at once.

I also tried to use matlab as Jordi Prats recommended but I couldn't get the answer. I couldn't understand the following part of his recommendation.

[1 93 1] is start, since longitude(1) = 0, latitude(93) = 39.9218, and you start at the first time measurement

[9 10 Inf] is count, since you want to extract data for 9 longitudes (longitude(9) = 11.25), for 10 latitudes (latitude(102) = 52.5286), and you use Inf to indicate you want all time points

[1 1 1] is stride, since you want to extract each longitude point, each latitude point and each time point.

Thank you in advance for your help.

#In matlab we can extract data for our location#

file = ('prcp_daily_1948-1948.nc'); # give u r file name#

lon = ncread(file,'longitude'); # type ncdisp('filename")#

lat = ncread(file,'latitude'); #see variable name for some files longitud#

prcp = ncread(file,'prcp'); #may be as lon if lon means type lon and lat#

latlowlimit=23; #here and prcp also change for some files pr#

latuplimit=28; #will be their u type by seeing variable name#

lonlowlimit=75; #type lat and long limets of u r study area#

lonuplimit=78;

latgrid=[];a=1;

longrid=[];b=1;

for i=1:720 #u have to see how many lons are their and give here#

if lon(i)>lonlowlimit & lon(i)latlowlimit & lat(i)

Hi,

I have tried using matlab as Jordi Prats has described above. It works good if the netcdf file has 3 dimensions. But I am working with 4 dimension relative humidity file having these attributes:

Size: 192x96x15x365

Dimensions: lon,lat,plev,time

Datatype: single

Now I am trying to extract relative humidity values at my study area for only one pressure level (500), plev =4 in this case for all time step by using this command:

rhum = ncread(fname,'hur',[96 45 4 1], [2.4 1.2 4 inf], [1 1 1 1]);

But this gives me rhum values starting from 1st level to 4th level and at each level there are two values of relative humidity. I am unable to understand why matlab shows two values at each pressure level. And how can I extract only one level (4th in this case) by modifying my command above. Can someone help?

Regards,

Raazia

Hi Raazia,

the answers to this question (https://www.researchgate.net/post/How_to_Create_a_Plot_of_CMIP5_Hydrology_Data_Against_Time_in_MATLAB) may help you.

Hey every one,

In r how i extract a range of data from ncdf file like my lat:33-35 and lon: 73-76.

in above mention code by Thiago Veloso dos Santos , we can extract for single point.

can someone help me?

Thanks

Hi Naeem

As I mentioned in past, my software can do it for you.

Please visit the website:

www.agrimetsoft.com

You can easily use it for your desirable region. Also you can watch the help file and video files (for an example) via the following links. If you have further question you can send email to: [email protected]

https://youtu.be/2ipu86ijBjU

https://youtu.be/QLIcBlMIioI

‌ Regards,

Nasrin

Thank you Jordi Prats, the MATLAB code you evaluated works well for files on Regular grids. But what changes should be made in the following line:

Precip = ncread(file,'pr',[1 93 1], [9 10 Inf], [1 1 1]); if I am working on rotated grids. I am attaching a file which is on rotated grids.

Regards,

M Azharuddin

Hi Mohammed,

you need to transform the rotated coordinates to regular coordinates. You can find the Matlab function rotated_grid_transform() by simon Funder here: https://fr.mathworks.com/matlabcentral/fileexchange/43435-rotated-grid-transform

Hi Alham Walid and everyone else.

As mentioned above, I have used Panoply, a user friendly software developed by NASA, available online for free, to extract .nc files. Then I copied them to excel simply and analyse them. Also u can use several options for downscaling. I used the Qmap package of quantile mapping method in R statistical software. There u should define three input files of yours and model baseline and another file for the desired RCP scenario u wish to use it. Then the package simply correct the bias of your RCP data.

Another easy way to use instead of above tasks is to use online MarksimGCM website. U can use the desired GCM and Scenario u wish and download the downscaled data for every individual coordination u wish. The weakness of Marksim is the baseline issue, it has its own baseline derived from Worldclim data series. But again u can correct the bias, if available, using Qmap package or any other tools.

Dear Ahmad Reza Razavi

Please can you send to me the links of the your mention software if possible.

many thanks for all for very useful discussion

Hi

please watch this video:

https://youtu.be/XfYzmbjfmNw

Hi Alham,

In your posts you state you're using Matlab which is cool. If you ever fancy trying some analysis in R take a look at this article:

https://journal.r-project.org/archive/2017/RJ-2017-032/RJ-2017-032.pdf

It might also be useful anyway as it describes quite a bit about the CMIP5 and netCDF formats.

All the best.

Dear Alham,

You try to use either HDF viewer or Panoply to find out the data for your selected station. Coding can be done in ferret or python to extract the data for the station

at a single time.

Hi Alham Al-Shurafat

You can try Ferret, its quite easy to extract time varying location and also save as to any format.

use x.nc

sh data

list/x=long/y=lat parameter

list/clobber/x=/y=/file=x.txt param

Hi, I have attached a python script as an example.

'''

if you are a Python user:

xarray itself is a pure Python package, to read and manipulate NetCDF.

The easiest way to get everything installed is to use conda.

To install xarray with its recommended dependencies on Windows:

in windows command prompt type: conda install -c anaconda xarray

after installation:

'''

# if you have only one file:

import xarray as xr

var = 'soil_moisture' # your variable name

indir = "/g/data"

lat_bound = slice(-23, -23.1)

lon_bound = slice(123, 123.1)

dsref= xr.open_dataset('testFolder/testData.nc').sel(lat=lat_bound,lon=lon_bound)

refarray = dsref[var].values

# if you wants to crete time series for a specific point

import xarray as xr

from datetime import datetime

from datetime import timedelta

var = 'soil_moisture' # your variable name

indir = "/g/data"

lat_bound = slice(-23, -23.1) # point latitude

lon_bound = slice(123, 123.1)

prefix = "myNetcdfData_" #

startdate = datetime(2016,1,1)

ndays = 100 # for Ex 100 days

for di in range(0,ndays+1):

daily.append((startdate+timedelta(days=di)).strftime('%Y%m%d'))

for i in range(len(daily)):

try:

ds= xr.open_dataset(indir + '/' +'{}{}.nc'.format(prefix,daily[i])).sel(lat=lat_bound,lon=lon_bound)

inList=[]

inList.append((ds[var].values))

except IOError as e:

if e.errno == errno.ENOENT:

print "no such file or directory for: %s%s"%(prefix, daily[i])

#inList=[]

#result = np.empty((refarray.shape))

#result.fill(np.nan)

#inList.append((result))

else:

raise

Dear Al-Shurafat

I sure that your problem solved, but I prefer to answer this question since I know this is one of the big issue which many scholars and researchers have problem with it.

For reading and extracting NetCDF files (nc, nc4, and etc) you should work with different programming languages and software, including ArcGIS, NCL, and Matlab. But it is not easy and you need time to learn them.

In this regard, I want to inform you an important user-friendly tool which can assist you for viewing, reading and extracting NetCDF files. By using the mentioned tool you can speed up the time of your project and do your assignments without any concerns. See the following link:

https://agrimetsoft.com/Netcdf-Extractor.aspx

It is useful and applicable. If you want to read about NetCDF format file, you can find an important content in the above link.

Best Regards,

@Nasrin Salehnia

Hi all,

CDO works very good with netCDF

This command can be used for specific locations (long, lat)

cdo -outputtab,date,value -remapnn,lon=X_lat=Y infile.nc > output.txt

here X = longitude and Y = latitude

If you work in Python, a good solution is to use Xarray to convert the netcdf to an annay, which can then be manipulated as a common numpy array. See : http://xarray.pydata.org/en/stable/io.html

Dear all,

Check the beta version of ncextract matlab function (https://se.mathworks.com/matlabcentral/fileexchange/73055-automated-nctools) for an automated extraction of netcdf data at specified coordinates (see example).

If you have any questions or problems, contact: [email protected]

Hi Basem Elshafei ,

CDO (climate data operator) is better tool for NetCDF file analysis.

Concatenate all .nc files into single file using follwing command

$ cdo -cat *.nc outfile.nc

If you want to extract data at specific location use following command

$ cdo -outputtab,date,value -remapnn,lon=x_lat=y infile.nc > output.txt

put Longitude value for x, and latitude value for y

If you want to extract data for specific area, use following command

$ cdo -sellonlatbox,lon1,lon2,lat1,lat2 inputfile.nc outfile.nc

and to take average of specific area, use following

$ cdo -fldmean -sellonlatbox,lon1,lon2,lat1,lat2 inputfile.nc outfile.nc

To convert the output of above command to text, use following command

$ cdo -outputtab,date,name,value outfile.nc > outfile.txt

Now you can import text file to excel or any other software to plot time series graph

Regards

Rashid

1. at ArcMap software

2. Then use the Arc Toolbox: Multidimension Tools- then make NetCDF to Raster layer add the .NC file…. then process

3. Then create a point shapefile to extract the value

4. Extract Multi Values to points

5. Then export the value in Excel format

Sample python script for time series extraction of data sets from NETCDF data:

#This code extract long- lat point over time for a set of locations from NETCDF Data.

import numpy as np

import matplotlib.pyplot as plt

import xarray as xr

import warnings

%matplotlib inline

np.set_printoptions(precision=3, linewidth=80, edgeitems=1) # make numpy less verbose

xr.set_options(display_width=70)

warnings.simplefilter('ignore')

path='xx/xx/data.nc'

stations=[['station1',X,Y],['Station2',X,Y],['Station3',X,Y]] #This is for all the stations you want to extract#

data=xr.open_dataset(path)

for station in stations:

st=data.precip.sel(latitude=station[2],longitude=station[1],method='nearest') #Verify the data arrangement and file name and change appropriately#

name='xx/xx/results/'+station[0]+'.csv'

np.savetxt(name, st ,fmt='%10.2f', delimiter=",")

print(station[0]+" is done")

Hi

#timeseries #analysis

This dataset contains the responses of a gas multisensor device deployed on the field in an Italian city. The dataset contains 9,358 instances of hourly averaged responses from an array of several metal oxide chemical sen- sors embedded in an Air Quality Chemical Multisensor Device. The device was located on the field in a significantly polluted area, at road level, within an Italian city. Data were recorded from March 2004 to April 2005 (one year). Ground Truth hourly averaged concentrations for Total Nitrogen Oxides (NOx) and Nitrogen Dioxide (NO2) are pro- vided, together with information on weather conditions. Missing values are tagged with -200 value.

• Im trying to split the data into two datasets: a training dataset and a test dataset. The training dataset should incude the first 7,111 observations (until the last observation of 2004). The aim will be to use the training dataset to forecast the value of NOx concentration at 9am in January 2005. Therefore, split the original dataset into a training and a test dataset. The test dataset should include the 31 observations at 9am every day in January 2005.
• Fit a statistical model to the training data and use it to forecast the NOx concentra- tion at 9am every day in January 2005

Please guide me, I'm unable to proceed

install.packages("fma")

library(fma)

install.packages("xts")

library(xts)

install.packages("dplyr")

library(dplyr)

airq

Hi,

GCM data are available at the following link:

https://climate4impact.eu/impactportal/data/esgfsearch.jsp

Details on GCM selection and climate data downscaling is mentioned in:

Article Representative General Circulation Models (GCMs) selection a...

Hi

If you can coding in MATLAB or work with Arc GIS, by devoting time and exact check to extract your domain, you can follow the above comments of other researchers have mentioned above. Otherwise, it is not an easy process, since in CMIP5 complex, each model has various specific and distinct files. When you want to extract your domain or several points in your case study, you should pay attention to the scales of files, the number of files, and other tips for an accrue extraction.

But if you don't have enough time and it is boring for you to learn a programming language to this activity, I direct you to see the bellow links; That is so useful.

https://youtu.be/kR4Q59WQ6og

https://youtu.be/FAON53uUlLQ

Regards,

Narges

Simple netcdf command you can use ferret tool

List/x=long/y=lat/z=depth or what variable you want/clobber/file=xyz.txt or xyz.xls

Regards

This paper on "Representative general circulation models selection and downscaling of climate data" might be helpful:

https://rmets.onlinelibrary.wiley.com/doi/full/10.1002/joc.6447

Hi I prefer use Python I show a sample of code that maybe can help you for atmospheric pressure data from CMIP5 :

##############################################

ncfile='psl_Amon_ACCESS1-0_historical_r1i1p1_185001-200512.nc'

file = Dataset(ncfile,mode='r') # abrir el archivo en modo lectura

print(file.variables)

lon=file.variables["lon"][:]

lat=file.variables["lat"][:]

nc_time= file.variables["time"][:]

t_unit= file.variables["time"].units

psl= file.variables["psl"][:]

try :

t_cal= file.variables['time'].calendar

except AttributeError:

t_cal=u"gregorian"

try :

t_cal=file.variables['time'].calendar

except AttributeError:

t_cal=u"gregorian"

################################################

# Conver numbers to dates

datevar=num2date(nc_time[:],units=t_unit,calendar=t_cal)

# Filter by year

years=np.array([idx.year for idx in datevar])

idx_time=(years>=1850)*(years=16.25)*(lat=271.875)*(lon

GCM data download, GCM selection for a catchment and downscaling is mentioned in this paper.

Article Representative General Circulation Models (GCMs) selection a...

You can use CDO or even GrADS, both are excellent tools to do what you are asking for. I hope this can help

You can use nco package using the following command:

ncks -d lat,40 -d lon,40 in.nc out.nc

For the R users

Here a script I did to extract ERA5 time series at different pressure levels and different locations corresponding to :

- Several stations locations (longitudes and latitudes coordinates)

- Climate variables (Temperature, Humidity, Winds)

- Different times of Observations (from the first of Year to last Year of Observations)

Here one of two approaches can be applied : "Simple" and "bilinear"

First of all, you need these two packages:

library(raster)

library(ncdf4)

report_dir

Take help from Met office or GrADS experts

CDO can help!

Dear Alham,

It is so easy with CDO remap command. Since your CMIP5 data are gridded, you have to interpolate them to your specific coordinates. Please check CDO user guide.

cdo remapdis,lon=xx.xx-lat=xx.xx input.nc output.nc

In given example, I use distance-weighted average remapping, but you can either choose nearest neighbor (remapnn).

Best,

Tufan

Use CDO or NCO or you can find prewritten codes for Matlab and python in Github

You can easily do this using R language. You need the package "ncdf4" from CRAN and then follow the code below,

library(ncdf4)

nc = nc_open("filename")

lat=ncvar_get (nc,"lat")

lon=ncvar_get(nc,"lon")

suppose,i want to extract longitude from 80E to 100E and latitude 5N to 25N

lat_rng = c(5,25)

lon_rng=c(80,100)

lon_ind = lon_lim[1] & lon

Dear Khandoker Rakid Hossain. Thank for your code.

It works fine for me with NetCDF from the NOAA-PSL FTP server, except for the line with the object which is not define in your code ?

I guess this particular line is not mandatory if NetCDF is already spatially subseted.

Thank again.

Here are my minor improvements, to read already spatially subsetted NetCDF form that code. Hope it helps. Alex

# open NetCDF format downloads from NOAA-PSL FTP

#https://psl.noaa.gov/data/gridded/data.20thC_ReanV3.html

#https://psl.noaa.gov/cgi-bin/db_search/SearchMenus.pl

# 2021 Mar 2

# code from https://www.researchgate.net/post/How-can-I-extract-a-time-series-of-variable-for-a-specific-location-specific-longitude-and-latitude-from-a-CMIP5-experiment-netCDF-data-set

library(ncdf4)

nc = nc_open("X24.212.19.197.60.9.14.36.nc")

lat=ncvar_get (nc,"lat") ; lon=ncvar_get(nc,"lon")

lat ; lon

nct = ncvar_get(nc, "time")

variable

This is how I extract the time series (and find the annual max data) in python:

#importing libraries

from netCDF4 import Dataset

import numpy as np

import pandas as pd

import glob, os

#Paste the directory of the folder that your nc files are in it, instead of "XXXXX"

mydir=r"XXXXX"

os.chdir(mydir)

#List your files that are in the path (directory)

names=[]

for file in os.listdir(mydir):

if file.endswith(".nc"):

names.append(file)

#Get the variables that the NC files have

datavariables=Dataset(names[1])

###Next step is: By Printing the variable names you can choose the title of the column you want to use it. Lets assume the variable that you want, is called "precipitation__map_rcp85_daily".

print(datavariables.variables.keys())

#Extractin time series (called rain here) and then pick the max:

maxrain=[]

for name in names:

rain=[]

data = Dataset(name)

pr=data.variables['precipitation__map_rcp85_daily'][:]

time=data.variables['time'][:]

for j in range(0, len(time)):

#choosing the location you want to looking at.

value=pr[j,1,1]

rain.append(value)

maxis=max(rain)

maxrain.append(maxis)

# saving the max data into a table

dfj = pd.DataFrame(maxrain, index =names,columns =['Names'])

#writting that table in "mydir"

dfj.to_csv('rcp85.csv', index=True)

There are several tools for this aim and those can save your time. https://agrimetsoft.com/netcdf