pyswi package

Submodules

pyswi.iterative_storage module

pyswi.iterative_swi module

pyswi.swi_calc_routines module

pyswi.swi_calc_routines.swi_calc_cy(juldate, ssm, ctime, swi_jd, nom, denom, last_jd_var, norm_factor, nan)
pyswi.swi_calc_routines.swi_calc_cy_noise(juldate, ssm, ctime, swi_jd, nom, denom, last_jd_var, ssm_noise, nom_noise)

pyswi.swi_ts module

pyswi.swi_ts.calc_swi_noise_rec(ssm_ts, t_value, last_den=1, last_nom=0)[source]

Recursive calculation of Soil Water Index (SWI) noise.

Parameters:

  • ssm_ts (numpy.ndarray) – Surface soil moisture time series with fields: sm_jd, sm, sm_noise

  • t_value (numpy.ndarray) – Characteristic time length.

  • denom (float) – denom value of the last calculation and starting point for the calculation.

  • nom (float) – nom value of the last calculation and starting point for the calculation.

Returns:

swi_noise_ts – Soil Water Index noise time series.

Return type:

numpy.ndarray

pyswi.swi_ts.calc_swi_ts(ssm_ts, swi_jd, gain_in=None, t_value=[1, 5, 10, 15, 20], nom_init=0, denom_init=0, nan=-9999.0)[source]

Time series calculation of the Soil Water Index.

Parameters:

  • ssm_ts (numpy.ndarray or dict) – Surface soil moisture time series with fields: sm_jd, sm, sm_noise

  • swi_jd (numpy.ndarray) – Julian date time stamps of the SWI time series.

  • gain_in (dict, optional) – Gain parameters of last calculation. Dictionary with fields: last_jd, nom, denom

  • t_value (list, optional) – Characteristic time length (default: 1, 5, 10, 15, 20).

  • nom_init (float64, optional) – Initial value of nom in the SWI calculation (default: 0).

  • denom_init (float64, optional) – Initial value of denom in the SWI calculation (default: 0).

  • nan (float64, optional) – NaN value to be masked in the SWI retrieval.

Returns:

  • swi_ts (numpy.ndarray) – Soil Water Index (SWI) time series.

  • gain_out (dict) – Gain parameters of last calculation. fields gpi, last_jd, nom, denom, nom_ns

pyswi.swi_ts.swi_error_prop(ssm, t_value, t_noise, swi_error, gain_in=None, nan=-9999.0)[source]

Recursive SWI calculation and error propagation function based on DeSantis and Biondi (2018; https://doi.org/10.29007/kvhb) Translated from MatLab code obtained from the authors.

Parameters:

  • ssm (numpy.ndarray) – Surface soil moisture time series with fields ‘sm’, ‘sm_uncertainty’ and ‘sm_jd’

  • t_value (numpy.ndarray) – Exponential filter characteristic T-value parameter

  • t_noise (numpy.ndarray) – T-value standard error. 10% of T for calibrated T-values.

  • swi_error (numpy.ndarray) – Exponential filter model structural error. ubMSE(ISMNswi, ISMNrzsm), based on empirical experiments.

  • gain_in (dict) – stored parameters of the last iteration.

  • nan (float) – nan value of the input ssm dataset

Returns:

  • swi (numpy.ndarray) – Soil water index time series

  • swi_noise (numpy.ndarray) – Soil water index noise time series

Module contents