# -*- coding: utf-8 -*-
"""In-built QC frameworks to apply to rainfall data to create quality controlled data."""
from rainfallqc.checks import (
comparison_checks,
gauge_checks,
neighbourhood_checks,
pypwsqc_filters,
subhourlyqc_checks,
timeseries_checks,
)
INTENSE_QC = {
"QC1": {"function": gauge_checks.check_years_where_nth_percentile_is_zero},
"QC2": {"function": gauge_checks.check_years_where_annual_kth_largest_value_is_zero},
"QC3": {"function": gauge_checks.check_day_of_week},
"QC4": {"function": gauge_checks.check_hour_of_day},
"QC5": {"function": gauge_checks.check_intermittency},
"QC6": {"function": gauge_checks.check_breakpoints},
"QC7": {"function": gauge_checks.check_min_val_change},
"QC8": {"function": comparison_checks.check_annual_exceedance_etccdi_r99p},
"QC9": {"function": comparison_checks.check_annual_exceedance_etccdi_prcptot},
"QC10": {"function": comparison_checks.check_exceedance_of_rainfall_world_record},
"QC11": {"function": comparison_checks.check_hourly_exceedance_etccdi_rx1day},
"QC12": {"function": timeseries_checks.check_dry_period_cdd},
"QC13": {"function": timeseries_checks.check_daily_accumulations},
"QC14": {"function": timeseries_checks.check_monthly_accumulations},
"QC15": {"function": timeseries_checks.check_streaks},
"QC16": {"function": neighbourhood_checks.check_wet_neighbours_daily},
"QC17": {"function": neighbourhood_checks.check_wet_neighbours_hourly},
"QC18": {"function": neighbourhood_checks.check_dry_neighbours_daily},
"QC19": {"function": neighbourhood_checks.check_dry_neighbours_hourly},
"QC20": {"function": neighbourhood_checks.check_monthly_neighbours},
"QC21": {"function": neighbourhood_checks.check_timing_offset},
"QC22": {"function": neighbourhood_checks.check_neighbour_affinity_index},
"QC23": {"function": neighbourhood_checks.check_neighbour_correlation},
"QC24": {"function": neighbourhood_checks.check_daily_factor},
"QC25": {"function": neighbourhood_checks.check_monthly_factor},
}
INTENSE_RULEBASE_QC = {
"QC2": {
"function": gauge_checks.check_years_where_annual_kth_largest_value_is_zero,
},
"QC10": {
"function": comparison_checks.check_exceedance_of_rainfall_world_record,
},
"QC11": {
"function": comparison_checks.check_hourly_exceedance_etccdi_rx1day,
},
"QC12": {
"function": timeseries_checks.check_dry_period_cdd,
},
"QC13": {
"function": timeseries_checks.check_daily_accumulations,
},
"QC14": {
"function": timeseries_checks.check_monthly_accumulations,
},
"QC15": {
"function": timeseries_checks.check_streaks,
},
"QC17": {
"function": neighbourhood_checks.check_wet_neighbours_hourly,
},
"QC19": {
"function": neighbourhood_checks.check_dry_neighbours_hourly,
},
"QC20": {
"function": neighbourhood_checks.check_monthly_neighbours,
},
}
# SubHourlyQC from Villalobos-Herrera et al. 2022 (https://doi.org/10.1002/qj.4357). Also see Table S1 of that paper.
UK_SUBHOURLY_QC = {
# From IntenseQC (12)
"HQC_QC13": {"function": timeseries_checks.check_daily_accumulations},
"HQC_QC14": {"function": timeseries_checks.check_monthly_accumulations},
"HQC_QC5": {"function": gauge_checks.check_intermittency},
"HQC_QC1": {"function": gauge_checks.check_years_where_nth_percentile_is_zero},
"HQC_QC2": {"function": gauge_checks.check_years_where_annual_kth_largest_value_is_zero},
"HQC_QC3": {"function": gauge_checks.check_day_of_week},
"HQC_QC4": {"function": gauge_checks.check_hour_of_day},
"HQC_QC6": {"function": gauge_checks.check_breakpoints},
"HQC_QC17": {"function": neighbourhood_checks.check_wet_neighbours_hourly},
"HQC_QC19": {"function": neighbourhood_checks.check_dry_neighbours_hourly},
"HQC_QC8": {"function": comparison_checks.check_annual_exceedance_etccdi_r99p},
"HQC_QC9": {"function": comparison_checks.check_annual_exceedance_etccdi_prcptot},
# Modified part of IntenseQC (4)
"HQC_UK1hr": {"function": subhourlyqc_checks.check_exceedance_of_UK_1hr_record},
"HQC_UK24hr": {"function": subhourlyqc_checks.check_exceedance_of_UK_24hr_record},
"HQC_UK24hr_rolling": {"function": subhourlyqc_checks.check_daily_exceedance_of_UK_24hr_record},
"HQC_streaks_20mm": {"function": subhourlyqc_checks.check_streaks_20mm},
# New checks (3)
# "SHQC_QC_spike_check": {"function": subhourlyqc_checks.spike_check}, # awaiting chat with Roberto
"SHQC_freqResChecker": {"function": subhourlyqc_checks.check_freq_is_subhourly},
"SHQC_subH_checkr": {"function": subhourlyqc_checks.check_subhourly_thresholds},
}
INTENSEQC_W_SUBHOURLYQC_RULEBASE = {
"QC2": {
"function": gauge_checks.check_years_where_annual_kth_largest_value_is_zero,
},
"QC10": {
"function": comparison_checks.check_exceedance_of_rainfall_world_record,
},
"QC11": {
"function": comparison_checks.check_hourly_exceedance_etccdi_rx1day,
},
"QC12": {
"function": timeseries_checks.check_dry_period_cdd,
},
"QC13": {
"function": timeseries_checks.check_daily_accumulations,
},
"QC14": {
"function": timeseries_checks.check_monthly_accumulations,
},
"QC15": {
"function": timeseries_checks.check_streaks,
},
"QC17": {
"function": neighbourhood_checks.check_wet_neighbours_hourly,
},
"QC19": {
"function": neighbourhood_checks.check_dry_neighbours_hourly,
},
"QC20": {
"function": neighbourhood_checks.check_monthly_neighbours,
},
"HQC_UK1hr": {"function": subhourlyqc_checks.check_exceedance_of_UK_1hr_record},
"HQC_UK24hr": {"function": subhourlyqc_checks.check_exceedance_of_UK_24hr_record},
"HQC_UK24hr_rolling": {"function": subhourlyqc_checks.check_daily_exceedance_of_UK_24hr_record},
"HQC_streaks_20mm": {"function": subhourlyqc_checks.check_streaks_20mm},
"SHQC_freqResChecker": {"function": subhourlyqc_checks.check_freq_is_subhourly},
"SHQC_subH_checkr": {"function": subhourlyqc_checks.check_subhourly_thresholds},
}
PYPWSQC = {
"BC": {"function": pypwsqc_filters.run_bias_correction},
"EBF": {"function": pypwsqc_filters.run_event_based_filter},
"IC": {"function": pypwsqc_filters.run_indicator_correlation},
"FZ": {"function": pypwsqc_filters.check_faulty_zeros},
"HI": {"function": pypwsqc_filters.check_high_influx_filter},
"PRF": {"function": pypwsqc_filters.run_peak_removal},
"SO": {"function": pypwsqc_filters.check_station_outlier},
}
INBUILT_QC_FRAMEWORKS = {
"intenseqc": INTENSE_QC,
"intenseqc_rulebase_only": INTENSE_RULEBASE_QC,
"uk_subhourlyqc": UK_SUBHOURLY_QC,
"pypwsqc": PYPWSQC,
"intenseqc_w_subhourlyqc_rulebase": INTENSEQC_W_SUBHOURLYQC_RULEBASE,
}
# all checks that are computed as summary of overall data or once per year values
NON_ROWWISE_QC_CHECKS = ["QC1", "QC2", "QC3", "QC4", "QC5", "QC7", "QC8", "QC9", "QC21", "QC22", "QC23"]
NON_ROWWISE_QC_CHECKS_NAMES = [
"percentiles_zero",
"k-largest_zero",
"days_of_week",
"hours_of_day",
"intermittency",
"min_val_change",
"r99p",
"prcptot",
"timings_offset",
"affinity_index",
"pearson_corr",
]
NON_ROWWISE_QC_CONVERTER = dict(zip(NON_ROWWISE_QC_CHECKS, NON_ROWWISE_QC_CHECKS_NAMES, strict=True))
[docs]
def get_qc_name_from_qc_key(qc_key: str) -> str:
"""
Return the name of a QC check.
Parameters
----------
qc_key:
QC check ID
Returns
-------
qc_name:
QC check name
"""
if qc_key in NON_ROWWISE_QC_CONVERTER.keys():
return NON_ROWWISE_QC_CONVERTER[qc_key]
