SL.125 Dataset
=============

Description
-----------
* Target Soil Properties: SOM, pH, Clay
* Groups of Features: ERa, vis-NIR 
* Sample size: 125
* Number of Features: 2,082
* Coordinates: Without coordinates because of privacy concerns instead with dummy coordinates (EPSG: 4326)
* Location: Skåne Län, Sweden
* Sampling Design: Three sampling designs over multiple adjacent fields: (1) regular grid sampling, targeted sampling through surface tortoise sampling (Persson et al. 2023) based on (2) ERa and (3) reflectance from remote sensing
* Study Area Size: 78 ha
* Geological Setting: High spatial variability of sandy till, clay till with chalk and glacial clay
* Previous Data Publication: None
* Contact Information:
    * Johanna Wetterlind (Johanna.Wetterlind@slu.se), Swedish University of Agricultural Sciences
    * Bo Stenberg (Bo.Stenberg@slu.se), Swedish University of Agricultural Sciences
* 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 SL.125 dataset contains soil measurements and features organized in a dataframe with the following properties:

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

SOM - Soil Organic Matter
    * Code: ``SOM_target``
    * Unit: %
    * Protocol: Measured through the weight difference before and after ignition of the soil with additional correction for structural water from clay by using the formula: SOM = LI − 0.46 − 0.047 × clay content (%) (Ekström 1927)
    * Sampling Date: September 2006
    * Sampling Depth: 0 - 20 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
    * Sampling Date: September 2006
    * Sampling Depth: 0 - 20 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 (Gee and Bauder 1986)
    * Sampling Date: September 2006
    * Sampling Depth: 0 - 20 cm

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

ERa – Apparent Electrical Resistivity
    * Number of Features: 1
    * Code(s): ``ERa``
    * Unit: Ω m
    * Sensing: EM38 sensor (Geonics Ltd., Mississauga, Canada) with exploration depth of 0 - 150 cm, in-situ
    * Processing: Ordinary Kriging to align sensing- with soil sampling locations
    * Sampling Date: April 2006

vis-NIR – Visible and Near Infrared Spectroscopy
    * Number of Features: 2,081
    * Code(s): ``wl_420``, ``wl_421``, ``wl_422`` ... ``wl_2500``
    * Unit: % (Reflectance)
    * Sensing: FieldSpec Pro FR scanning instrument (Analytical Spectral Devices Inc., Boulder, USA), on dried and sieved samples (<2 mm) in the laboratory, spectral range was 350 – 2,500 nm at 1.4 – 2.0 nm intervals
    * Processing: Discarding noisy edges of the spectrum (350 - 420 nm), resampling to 1 nm intervals
    * Sampling Date: September 2006
    * Spectral Information (After Data Processing):
        * Data Representation: Wavelength (in nm)
        * Spectral Resolution: 1 nm
        * Spectral Range: 420 - 2,500 nm

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("SL.125")
    dataset = data["Dataset"]
    folds = data["Folds"]
    coords = data["Coordinates"]  # Note: Contains dummy coordinates

    # Split into train/test using fold 1
    X_train, X_test, y_train, y_test = split_dataset(
        data=data,
        fold=1,
        targets=["pH_target", "SOM_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
----------

Ekström, G. (1927). Klassifikation av Svenska Åkerjordar (Classification of Swedish arable soils). Sveriges Geologiska Undersökning, Ser C. 345, 161 pp.

Gee, G.W. & Bauder, J.W. (1986) Particle-Size Analysis. In: Klute, A., Ed., Methods of Soil Analysis, Part 1. Physical and Mineralogical Methods, Agronomy Monograph No. 9, 2nd Edition, American Society of Agronomy/Soil Science Society of America, Madison, WI, 383-411.

Persson, K., Söderström, M. & Mutua, J. (2023). SurfaceTortoise: Find Optimal Sampling Locations Based on Spatial Covariate(s). R package version 2.0.1.