--- START OF FILE README.txt ---

README: Regional Groundwater Database for Arequipa and Surroundings

Dataset Title: Regional Groundwater Database from Unstructured Sources Based on an Integrated Optical Character Recognition and Large Language Model Workflow for Arequipa and Surroundings in Southern Peru
Authors: Héctor L. Venegas-Quiñones, Madeleine Guillen, et al.
Repository Platform: HydroShare
Date of Last Update: November 2025

1. OVERVIEW
-----------------
This repository contains the first comprehensive, open-access groundwater database for the Arequipa region and surrounding areas in Southern Peru. The dataset consolidates 4,775 multi-parameter data points encompassing depth to water, aquifer thickness, transmissivity, and storage coefficient, and is augmented with 931 depth-to-water records from Peru's National Water Resources Information System (SNIRH) spanning the period 1966–2024.

The data was derived from two primary streams:
1. "Gray Literature" Mining: 3,675 hydrogeological records (hydraulic conductivity, transmissivity, storage coefficient, lithology, etc.) extracted from 606 technical reports and academic theses using a semi-automated AI workflow (OCR + LLM) with 100% manual verification.
2. National Databases: 931 depth-to-water records sourced from Peru's National Water Resources Information System (SNIRH).

2. FILE STRUCTURE AND ACCESS
-----------------
IMPORTANT: Due to the comprehensive nature of the dataset, the database is compressed into a multi-part 7-Zip archive. These files contain the full raw data extraction in Excel format.

You must download ALL parts to access the data.

Files in this Repository:
- database.7z.001: Part 1 of the compressed archive.
- database.7z.002: Part 2 of the compressed archive.
- README.txt: This technical documentation file.

Instructions to Open:
1. Download both "database.7z.001" and "database.7z.002" to the same folder on your computer.
2. You need software capable of handling split archives (e.g., 7-Zip for Windows, Keka for macOS).
3. Right-click on the first file ("database.7z.001") and select "Extract Here".
4. The software will automatically combine Part 1 and Part 2.
5. This will extract the main Master File: 
   "arequipa_final_gw_database.csv"

3. DATA DICTIONARY (VARIABLE TRANSLATION)
-----------------
The column headers in the Excel file are in Spanish. The tables below provide the English translation and technical description.

A. SOURCE IDENTIFICATION (TRACEABILITY)
Spanish Header      | English Translation    | Description
------------------- | ---------------------- | --------------------------------------------------------
ID_Registro         | Record ID              | Unique identifier.
Codigo_Archivo      | File Code              | Internal code for the source PDF (e.g., ANA0000209).
Fuente_Completa     | Full Source            | Name of the institution or thesis repository.
Titulo_Documento    | Document Title         | Exact title of the original report/thesis.
Autor_es            | Author(s)              | Authors of the study.
Pagina_Referencia   | Page Reference         | Specific page(s) in the PDF where data was found.

B. SPATIAL & TEMPORAL
Spanish Header      | English Translation    | Description
------------------- | ---------------------- | --------------------------------------------------------
Fecha_Medicion      | Measurement Date       | YYYY-MM-DD.
Latitud_WGS84       | Latitude (WGS84)       | Standardized decimal latitude.
Longitud_WGS84      | Longitude (WGS84)      | Standardized decimal longitude.
Cota_msnm           | Elevation              | Meters above sea level.
CUENCA              | Basin                  | Hydrographic unit name.

C. HYDROGEOLOGICAL PARAMETERS
Spanish Header               | English Translation          | Unit
---------------------------- | ---------------------------- | --------
Nivel_Freatico               | Water Table Depth            | Meters (m)
Espesor_Saturado             | Saturated Thickness          | Meters (m)
Conductividad_Hidraulica_K   | Hydraulic Conductivity (K)   | See Unidad_K
Transmisividad_T             | Transmissivity (T)           | See Unidad_T
Coeficiente_Almacenamiento_S | Storage Coefficient (S)      | Dimensionless
Litologia                    | Lithology                    | Text

4. EXAMPLE: HOW TO TRACE DATA TO THE SOURCE
-----------------
To verify the origin of any data point in the "database.7z" files, use the 'Titulo_Documento' and 'Pagina_Referencia' columns.

Example (Row 2 in Excel):
- ID_Registro: 1
- Parameter: Transmisividad_T (Not listed in example row, but T would be here if measured) or Nivel_Freatico (176.05 masl).
- Source File: ANA0000209_1_compressed.pdf
- Document Title: "INVENTARIO Y EVALUACIÓN DE LAS FUENTES DE AGUA SUBTERRÁNEA EN EL VALLE DE ACARI"
- Author: MINISTERIO DE AGRICULTURA
- Page Reference: "54, 115"

Interpretation: The user can verify this record by finding the report "INVENTARIO... VALLE DE ACARI" (1981) and looking at pages 54 and 115, where this specific well data (ID 04/03/12-1) appears.

5. METHODOLOGY
-----------------
Data was extracted using a "Human-in-the-Loop" AI workflow:
1. OCR (Tesseract) converted scanned PDFs to text.
2. LLM (Gemini 2.5 Pro) parsed the text to extract hydrogeological parameters.
3. Validation: 100% of records were manually cross-checked by the authors against the source PDFs to ensure accuracy.

6. CITATION
-----------------
When using this dataset, please cite:
Venegas-Quiñones, H. L., et al. (2025). Regional Groundwater Database from Unstructured Sources Based on an Integrated Optical Character Recognition and Large Language Model Workflow for Arequipa and Surroundings in Southern Peru. HydroShare.
--- END OF FILE README.txt ---