sonu9989 / ESTS

Estimator of Subsidence Trend and Seasonality (ESTS): An open-source integrated framework and model that combines Seasonality and trend decomposition (STD), Geographically Weighted Regression (GWR) & Random Forest (RF) models for identifying and estimating the trend and seasonality in subsidence by only using groundwater (hydraulic head)

ESTS

Estimator of Subsidence Trend and Seasonality (ESTS) is an open-source integrated framework and model that combines Seasonality and trend decomposition (STD), Geographically Weighted Regression (GWR) & Random Forest (RF) models for identifying and estimating the trend and seasonality in subsidence by only using groundwater (hydraulic head) observations as input variable.

1. Seasonality and Trend Decomposition (STD) of Subsidence (s) and Groundwater (h)

Decomposition_main.py: - Conducts Time series decomposition of original hydraulic head (h) & deformation (s) time series dataset using the Seasonal-Trend Decomposition Procedure Based on Loess (STL) approach. Furthermore, this code also computes the Relative Importance (RI) of each decomposed components i.e., Trend, Seasonality, and Residue.

2. Model for the Non-trend (Seasonality) Dataset of Subsidence (s_non-trend) and Groundwater (h_non-trend)

Non-trend Model_main.ipynb: - Estimates the non-trend subsidence (s_non-trend) through non-trend groundwater (h_non-trend) using the Random Forest (RF) model.

s_non-trend = f(h_non-trend)

3. Change Estimator for Subsidence Change (Δs) based on Groundwater Change (Δh)

Change_Estimator_main.m: - Estimates subsidence change (Δs) through hydraulic head change (Δh) using GWR. Also provides spatial Intercepts & Coefficients for each Monitoring Wells.

Δs = f(Δh)

4. Trend Estimation of Subsidence i.e., s_trend

Trend Estimation.xlsx: - Provides calculation to estimate s_trend using the Δs and s_non-trend from "Change_Estimator_main.m" and "Non-trend Model_main.ipynb", respectively in 4 sample wells.

System Environment at the time of Experiment

OS: - Microsoft Windows 10 Home

CPU: - Inter(R) Core(TM) i7 7700HQ 2.80 GHz

RAM: - 16.00 GB

1. The "Decomposition_main.py", "Change_Estimator_main.m", and "Non-trend Model_main.ipynb" were tested on Python Programming Language (Version 3.9)

2. The "Change_Estimator_main.m" was tested on MATLAB (2022a) Environment.

Important References

1. Breiman, L., 2001. Random Forests. Mach. Learn. 45, 5–32. https://doi.org/10.1023/A:1010933404324
2. Cleveland, R.B., Cleveland, W.S., McRae, J.E., Terpenning, I., 1990. STL: A Seasonal-Trend Decomposition Procedure Based on Loess. J. Off. Stat. 6, 3–73.
3. LeSage, J., & Pace, R. K. (2009). Introduction to spatial econometrics. Chapman and Hall/CRC.