Skip to content

climakitae

BiasAdjustToStation

cal-adapt/climakitae

Processor: BiasAdjustModelToStation

Registry key: bias_adjust_model_to_station | Priority: 60 | Category: Data Refinement

Apply quantile delta mapping (QDM) bias correction to gridded climate model data using HadISD weather station observations as the training reference. Output is bias-corrected at the requested station locations, with one data variable per station.

Algorithm

flowchart TD
    Start([execute]) --> DictCheck{result is dict?}
    DictCheck -->|Yes| ExecDict[_execute_dict<br/>per-key recursion]
    DictCheck -->|No| LoadObs[_load_station_data<br/>HadISD reference]
    LoadObs --> TypeCheck{Dataset / DataArray?}
    TypeCheck -->|Yes| ProcSingle[_process_single_dataset]
    TypeCheck -->|No| TypeErr[raise TypeError]

    ExecDict --> ProcSingle
    ProcSingle --> Preproc[_preprocess_hadisd<br/>rename, K to C, attrs]
    Preproc --> BiasCorrect[_bias_correct_model_data<br/>QDM via xclim]
    BiasCorrect --> UpdateCtx[update_context]
    UpdateCtx --> End([Output: Dataset with one variable per station])

    click Start "https://github.com/cal-adapt/climakitae/blob/main/climakitae/new_core/processors/bias_adjust_model_to_station.py#L790" "execute"
    click ExecDict "https://github.com/cal-adapt/climakitae/blob/main/climakitae/new_core/processors/bias_adjust_model_to_station.py#L723" "_execute_dict"
    click LoadObs "https://github.com/cal-adapt/climakitae/blob/main/climakitae/new_core/processors/bias_adjust_model_to_station.py#L243" "_load_station_data"
    click ProcSingle "https://github.com/cal-adapt/climakitae/blob/main/climakitae/new_core/processors/bias_adjust_model_to_station.py#L510" "_process_single_dataset"
    click Preproc "https://github.com/cal-adapt/climakitae/blob/main/climakitae/new_core/processors/bias_adjust_model_to_station.py#L185" "_preprocess_hadisd"
    click BiasCorrect "https://github.com/cal-adapt/climakitae/blob/main/climakitae/new_core/processors/bias_adjust_model_to_station.py#L309" "_bias_correct_model_data"
    click UpdateCtx "https://github.com/cal-adapt/climakitae/blob/main/climakitae/new_core/processors/bias_adjust_model_to_station.py#L864" "update_context"

Parameters

The processor takes a dict:

Key	Type	Default	Description
`stations`	`list[str]`	`[]`	Station names to bias-correct against. Required for non-trivial use.
`historical_slice`	`tuple[int, int]`	`(1980, 2014)`	Years used as the training period.
`window`	`int`	`90`	Window size (days) for seasonal grouping in QDM.
`nquantiles`	`int`	`20`	Number of quantiles for the QDM mapping.
`group`	`str`	`"time.dayofyear"`	Temporal grouping passed to `xclim` QDM.
`kind`	`str`	`"+"`	Adjustment kind: `"+"` (additive) or `"*"` (multiplicative).

Requirements

Activity ID: WRF (dynamical downscaling). LOCA2 already includes statistical bias correction at the watershed level and is not the intended input.
Variable: Designed for hourly temperature (t2); the underlying station observations are HadISD temperature.
Time coverage: Input must include the historical training period (default 1980–2014). HadISD reference data is available through 2014‑08‑31.
Calendar: All inputs are converted to a noleap calendar internally for consistency.

Example

from climakitae.new_core.user_interface import ClimateData

data = (ClimateData()
    .catalog("cadcat")
    .activity_id("WRF")
    .institution_id("UCLA")
    .variable("t2")
    .table_id("1hr")
    .grid_label("d03")
    .processes({
        "time_slice": ("1980-01-01", "2050-12-31"),
        "bias_adjust_model_to_station": {"stations": ["KSAC", "KSFO"]},
    })
    .get())

# Result: one data variable per station, time-sliced to requested range.

Code References

Method	Lines	Purpose
`__init__`	148–183	Read configuration dict with defaults
`_preprocess_hadisd`	185–241	Rename / unit-convert / attribute the raw HadISD slice
`_load_station_data`	243–307	Resolve station IDs, load HadISD subset, return reference Dataset
`_bias_correct_model_data`	309–508	Build QDM (`xclim`) train/adjust per station
`_process_single_dataset`	510–721	Load reference, run QDM via `xarray.map`, return per-station vars
`_execute_dict`	723–788	Recursive dict path
`execute`	790–862	Dispatcher: dict / Dataset / DataArray
`update_context`	864–891	Record stations and QDM parameters in `new_attrs`
`set_data_accessor`	893–	Receive `DataCatalog` reference (used by `_load_station_data`)

Earlier docs implied an early "activity_id == WRF" guard with line numbers in the 80–145 range. Those line numbers do not exist in the source — the file's first method (__init__) starts at line 148. The activity-id constraint is enforced by data availability rather than an explicit early-return check.

See also