RP.62 Dataset
=============

Description
-----------
* Target Soil Properties: SOC, pH, Clay
* Groups of Features: ERa, Gamma, NIR, pH-ISE, VI 
* Sample size: 62
* Number of Features: 1,410
* Coordinates: Without coordinates because of privacy concerns
* Location: Rhineland-Palatinate, Germany
* Sampling Design: Regular grid sampling
* Study Area Size: 3.3 ha
* Geological Setting: Pleistocene periglacial slope deposits consisting of Weichselian loess with variable amounts of weathered Devonian sand-, silt-, and claystones and scattered Tertiary basalt bombs
* Previous Data Publication: None
* Contact Information:
    * Stefan Paetzold (s.paetzold@uni-bonn.de), University of Bonn
    * Hamed Tavakoli (HTavakoli@atb-potsdam.de), Leibniz Institute for Agricultural Engineering and Bioeconomy (ATB)
* License: CC BY-SA 4.0
* Publication/Modification Date (d/m/y): 28.02.25, version 1.0
* Changelog:
    * Version 1.0 (28.02.25): Initial release

Details
-------

Dataset
^^^^^^^
The dataset contains the following target soil properties and features:

Target Soil Properties:
""""""""""""""""""""""

SOC - Soil Organic Carbon
    * Code: ``SOC_target``
    * Unit: %
    * Protocol: Measured through titration after oxidization of the organic carbon (Walkley & Black 1934)
    * Sampling Date: October 2017
    * Sampling Depth: 0 - 30 cm

pH
    * Code: ``pH_target``
    * Unit: Unitless
    * Protocol: Measured in a water suspension with a glass electrode with a 5:1 liquid:soil volumetric ratio (DIN ISO 10390)
    * Sampling Date: May 2020
    * Sampling Depth: 0 - 30 cm

Clay
    * Code: ``Clay_target``
    * Unit: %
    * Protocol: Sieve-Pipette method, measured through fractioning the soil into the sand fractions by sieving, and the silt and clay fractions by sedimentation in water, German adaptation (DIN ISO 11277)
    * Sampling Date: October 2017
    * Sampling Depth: 0 - 30 cm

Groups of Features:
"""""""""""""""""

ERa – Apparent Electrical Resistivity
    * Number of Features: 1
    * Code(s): ``ERa``
    * Unit: Ω m
    * Sensing: Array of multiple rolling electrodes (Geophilus company, Caputh, Germany) on RapidMapper platform with exploration depth of 0 - 50 cm, in-situ
    * Processing: Ordinary Kriging to align sensing- with soil sampling locations
    * Sampling Date: September 2020

Gamma
    * Number of Features: 5
    * Code(s): ``G_Total_Counts``, ``G_K``, ``G_U``, ``G_Th``, ``G_Cs``
    * Unit: Unitless
    * Sensing: Passive gamma sensor (MS-2000-CsI-MTS, Medusa Radiometrics BV, Groningen, Netherlands) on RapidMapper platform, in-situ
    * Processing: Ordinary Kriging to align sensing- with soil sampling locations
    * Sampling Date: September 2020

NIR – Near Infrared Spectroscopy
    * Number of Features: 1,401
    * Code(s): ``wl_1000``, ``wl_1001``, ``wl_1002`` ... ``wl_2400``
    * Unit: % (Reflectance)
    * Sensing: NIR spectrometer (C11118GA, Hamamatsu Photonics K.K., Shizuoka Prefecture, Japan) on RapidMapper platform, in-situ, spectral range was 900 - 2550 nm at 15 nm intervals
    * Processing: Kriging to align sensing- with soil sampling locations, discarding noisy edges of the spectrum (900 - 1,000 nm & 2,400 - 2,550 nm), resampling to 1 nm intervals
    * Sampling Date: September 2020
    * Spectral Information (After Data Processing):
        * Data Representation: Wavelength (in nm)
        * Spectral Resolution: 1 nm
        * Spectral Range: 1,000 - 2,400 nm

pH-ISE – Ion Selective Electrodes for pH Determination
    * Number of Features: 1
    * Code(s): ``pH-ISE``
    * Unit: Unitless
    * Sensing: Soil pH Manager (VERIS Technologies, Salinas, USA) on RapidMapper platform, in-situ
    * Processing: Ordinary Kriging to align sensing- with soil sampling locations
    * Sampling Date: September 2020

VI - Vegetation Indices
    * Number of Features: 2
    * Code(s): ``NDVI``, ``GNDVI``
    * Unit: Unitless
    * Sensing: Sentinel-2 image during vegetative period (Level-2A) from "Copernicus Open Access Hub"
    * Processing: Calculating ``NDVI`` as (B08 - B04) / (B08 + B04) and ``GNDVI`` as (B08 - B03) / (B08 + B03), extracting VI values from raster at soil sampling locations
    * Sampling Date: March 2017

Examples
--------

.. code-block:: python

    from LimeSoDa import load_dataset, split_dataset
    from sklearn.linear_model import LinearRegression
    from sklearn.metrics import r2_score, mean_squared_error
    import numpy as np

    # Load and explore the dataset
    data = load_dataset("RP.62")
    dataset = data["Dataset"]
    folds = data["Folds"]
    coords = data["Coordinates"]  # Will be NA for RP.62

    # Split into train/test using fold 1
    X_train, X_test, y_train, y_test = split_dataset(
        data=data,
        fold=1,
        targets=["pH_target", "SOC_target", "Clay_target"]
    )

    # Fit model and get predictions
    model = LinearRegression()
    model.fit(X_train, y_train)
    predictions = model.predict(X_test)

    # Calculate performance metrics
    r2 = r2_score(y_test, predictions)
    rmse = np.sqrt(mean_squared_error(y_test, predictions))
    print(f"R-squared: {r2:.7f}")
    print(f"RMSE: {rmse:.7f}")

References
----------

Walkley, A. & Black, I. A. (1934). An examination of the Degtjareff method for determining soil organic matter, and a proposed modification of the chromic acid titration method. Soil science, 37(1), 29-38.