Data for Bio-Digital Convergence and Sustainable Artificial Intelligence: Evaluating the Tools of the Jach'a Qh’anax Model in the Andean-Amazonian and Mesoamerican Bioeconomy
Description
This dataset and its complementary software tools form the empirical validation framework of the Jach'a Qh'anax Model Ecosystem. This project explores bio-digital convergence and the application of sustainable Artificial Intelligence (AI) within the bioeconomy of the Andean-Amazonian and Mesoamerican regions, specifically covering Bolivia, Mexico, Guatemala, Honduras, and El Salvador. The primary objective of the ecosystem is to mitigate technical information asymmetries and climate risks faced by traditional rural agricultural producers through decentralized climate sensors and mobile tools, while actively integrating indigenous and ancestral knowledge. The collected data evaluates technological adoption, algorithmic predictive accuracy, the substitution of traditional chemical inputs for bioeconomic practices, and the direct impact on household income stability and food security across 200 multi-country participants during the 2026 agricultural cycle.
Files
Steps to reproduce
Step 1: Environment Setup & DependenciesR Environment: Ensure you have R (version > o = 4.4.3) and RStudio installed.Python Environment: Ensure Python 3.10+ is installed along with a functional environment manager (such as conda or venv).Required Libraries:In R: Install the required analytical packages by running: install.packages(c("readxl", "tidyverse", "ggplot2", "scales", "patchwork")) In Python: Install the Google Earth Engine API and geospatial dependencies: pip install earthengine-api pandas openpyxl Step 2: Google Earth Engine (GEE) Authentication & Climate Ingestion Open SI_Computational_Scripts.txt.docx and extract the Python script. Initialize and authenticate your GEE account in your terminal: earthengine authenticate Run the script to extract the historical climate parameters (ERA5-Land for precipitation/temperature and NASA-USDA SMAP for soil moisture). The script uses the geographic coordinates embedded in Data.xlsx to fetch localized pixel values from the cloud. Export the generated remote sensing metrics as a .csv or directly map them back into your local workspace. Step 3: Data Preparation and Codebook Mapping Locate Data.xlsx and ensure it sits in the same working directory as your execution scripts. Review S3_CodeBook_Data_Structure.xlsx to verify variable types, explicit scaling (such as the 1–5 Likert structures for ancestral integration), and column constraints. Use the S2_Tool_Validation_Questionnaire.docx to cross-reference categorical codes (e.g., crop types, pest IDs, and regional weed classifications) if manual data filtering is required. Step 4: Empirical Analysis and Visualization in R Open S4_R_Script_for_Empirical_Results_and_Bio-Digital_Validation_Metrics.docx and copy the code into an .R or .Rmd file inside RStudio. Execute the script. The routine automatically: Handles non-ASCII encoding fixes for regional place names. Cleans missing parameters and isolates the 200 validated producers (40 per country). Executes descriptive statistical modeling and computes model predictive accuracy indices. Outputs high-resolution figures saved directly to your workspace as .tiff files configured at 300 DPI. Step 5: Launching the Interactive Interface Extract the source code from S5_Script_Tool_Interactive_HTML.docx and save it locally as index.html. Open the index.html file in any modern web browser (Chrome, Firefox, or Edge). The interface will render the Calculadora Climática ConciencIA locally, utilizing Tailwind CSS via CDN. You can interactively input the field indicators from Data.xlsx to simulate real-time climate risk alerts, digital diagnostic reports, and view localized indigenous language translations.
Institutions
- Universidad Mayor de San AndrésLa Paz Department, La Paz
- Universidad Nacional Autónoma de Nicaragua-LeónLeón Department, León
- National Autonomous University of HondurasFrancisco Morazán Department, Tegucigalpa