Streamflow data anomaly

Introduction

To get the anomaly of the MAX streamflow data in a hydrobasin, we will continue working on the csv file we get during our raster to plygon conversion step

Goal

Python Scripting

Setup

We only need the pandas library to work with .csv files

import pandas as pd

nc_path = r"./hybas_lev05_Streamflow_tmax.csv"

data = pd.read_csv(nc_path)

Step 1

data["StdTime"] = pd.to_datetime(data["StdTime"])
data.set_index('StdTime', inplace=True)

data['month'] = data.index.month
data['year'] = data.index.year
# Filter the data for the years 2001 to 2024 (or beyond your current maximum date)
data = data[data.index.year <= 2024]

Step 2

# Calculate climatology (mean temperature for each month over years 2001 to 2020 for each PFAF_ID)
climatology_2001_to_2020 = data[data.index.year.isin(range(2001, 2021))].groupby(['PFAF_ID', 'month'])['MAX'].mean().rename('climatology_mean')

1. data[data.index.year.isin(range(2001, 2021))]: This part filters our data frame to include only the rows where the year in the index is within the range from 2001 to 2020. The isin(range(2001, 2021)) function checks if the year part of the index is within this range.

2. .groupby(['PFAF_ID', 'month']): After filtering, we grouped by the columns PFAF_ID and month. This means that the subsequent operations will be performed on each group of rows that have the same PFAF_ID and month.

3. ['MAX'].mean(): For each group, the code calculates the mean of the values in the MAX column.

4. .rename('climatology_mean'): The resulting Series, which contains the mean values, is renamed to climatology_mean.

Step 3

Now, let's merge the climatology we just calculated back to the object <data>. Then to get the anomaly, simply subtract the monthly mean from the column of cells that has the maximum stream flow.

# Merge climatology mean with the original data
data = data.merge(climatology_2001_to_2020, left_on=['PFAF_ID', 'month'], right_index=True, suffixes=('', '_clim'))

# Calculate anomaly
data['anomaly'] = data['MAX'] - data['climatology_mean']

Now the dataset should look as the following,

             PFAF_ID      Variable          MAX         climatology_mean  anomaly
2001-01-31     61564  ... Streamflow_tavg    53.298950         66.873812 -13.574861
2002-01-31     61564  ... Streamflow_tavg    69.457275         66.873812   2.583464
2003-01-31     61564  ... Streamflow_tavg    61.097664         66.873812  -5.776148
2004-01-31     61564  ... Streamflow_tavg    57.528519         66.873812  -9.345293
2005-01-31     61564  ... Streamflow_tavg    64.865562         66.873812  -2.008249
2019-12-31     67208  ... Streamflow_tavg  1349.599243       1814.002460 -464.403217
2020-12-31     67208  ... Streamflow_tavg  1546.011719       1814.002460 -267.990741
2021-12-31     67208  ... Streamflow_tavg  2432.778564       1814.002460  618.776105
2022-12-31     67208  ... Streamflow_tavg  1480.862183       1814.002460 -333.140277
2023-12-31     67208  ... Streamflow_tavg  1363.781982       1814.002460 -450.220477

[41676 rows x 10 columns]

And now we could move on appending it to our polygon files layer.