RainfallQC - Quality control for rainfall data

Provides methods for running rainfall quality control.

NOTEBOOK DEMO AVAILABLE HERE

Please email [email protected] if you have any questions.

Project To Do:

• There is a plan to make a ReadTheDocs for the package!
• add flag descriptions
• Use timestep str maker to the checks

Installation

RainfallQC can be installed from PyPi:

pip install rainfallqc

Example use

Example 1. - Individual quality checks on single rain gauge

# Load two types of QC'ing modules from RainfallQC
from rainfallqc import gauge_checks, comparison_checks

# 1. Simple 1 gauge QC
# 1.1. Run 1 qc method for 1 gauge
intermittency_flag = gauge_checks.check_intermittency(data, target_gauge_col="rain_mm")

# 1.2. Run 1 qc method for 1 gauge using in-built comparison dataset
wr_flags = comparison_checks.check_exceedance_of_rainfall_world_record(data, target_gauge_col="rain_mm", time_res='hourly')

# 1.3. Run 1 qc method for 1 gauge using in-built comparison dataset and location of gauge
rx1day_flags = comparison_checks.check_annual_exceedance_etccdi_rx1day(data, target_gauge_col="rain_mm", gauge_lon=1.0, gauge_lat=55.0)

Example 2. - Individual quality checks on networks of rain gauges

# 2. Run neighbour/network checks on a subset of a rain gauge network
from rainfallqc import neighbourhood_checks
from rainfallqc.utils import data_readers

# 2.1. Read in GDSR gauge network metadata
gdsr_obj = data_readers.GDSRNetworkReader(path_to_gdsr_dir="./tests/data/GDSR/")

# 2.2. subset by max 10 gauges within 50 km and with at least 500 days of overlap
nearby_ids = list(
    gdsr_obj.get_nearest_overlapping_neighbours_to_target(
        target_id="DE_00310", distance_threshold=50, n_closest=10, min_overlap_days=500
    )
)
nearby_ids.append(target_id)
nearby_data_paths = gdsr_obj.metadata.filter(pl.col("station_id").is_in(nearby_ids))["path"]

# 2.3. Load those nearest gauges from network metadata
gdsr_network = gdsr_obj.load_network_data(data_paths=nearby_data_paths)

# 2.4 Run a neighbourhood check (checking if suspiciously large rainfall values were seen in neighbours)
extreme_wet_flags = neighbourhood_checks.check_wet_neighbours(
    gdsr_network,
    target_gauge_col="rain_mm_DE_02483",
    neighbouring_gauge_cols=gdsr_network.columns[1:],  # exclude time
    time_res="hourly",
    wet_threshold=1.0, # threshold for rainfall intensity to be considered
    min_n_neighbours=5, # number of neighbours needed for comparison
    n_neighbours_ignored=0, # ignore no neighbours and include all
)

Example 3. - Applying a framework of QC methods (e.g. IntenseQC)

# 3. Applying multiple QC methods in framework (e.g. IntenseQC)
from rainfallqc.qc_frameworks import apply_qc_framework

# 3.1. Decide which QC methods of IntenseQC will be run
qc_framework = "IntenseQC"
qc_methods_to_run = ["QC1", "QC8", "QC9", "QC10", "QC11", "QC12", "QC14", "QC15", "QC16"]

# 3.2 Decide which parameters for QC
qc_kwargs = {
    "QC1": {"quantile": 5},
    "QC14": {"wet_day_threshold": 1.0, "accumulation_multiplying_factor": 2.0},
    "QC16": {
        "neighbouring_gauge_cols": daily_gpcc_network.columns[2:],
        "wet_threshold": 1.0,
        "min_n_neighbours": 5,
        "n_neighbours_ignored": 0,
    },
    # Shared defaults applied to all
    "shared": {
        "target_gauge_col": "rain_mm_DE_02483",
        "gauge_lat": gpcc_metadata["latitude"],
        "gauge_lon": gpcc_metadata["longitude"],
        "time_res": "daily",
        "data_resolution": 0.1,
    },
}

# 3.3. Run QC methods on network data
qc_result = apply_qc_framework.run_qc_framework(
    daily_gpcc_network, qc_framework=qc_framework, qc_methods_to_run=qc_methods_to_run, qc_kwargs=qc_kwargs
)

Other examples

Also see example Jupyter Notebooks here: https://github.com/Thomasjkeel/RainfallQC-notebooks/tree/main

Documents

• Free software: GNU General Public License v3
• Documentation: https://rainfallqc.readthedocs.io.

Features

• 25 rainfall QC methods (all from IntenseQC)
• editable parameters so you can tweak thresholds, streak or accumulation lengths, and distances to neighbouring gauges

Credits

Based on the IntenseQC: https://github.com/nclwater/intense-qc/tree/master

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.