In the realm of global agriculture

This dataset was made by augmenting optimum soil and environmental characteristics for crop growth

Analysis of ‘Crop Recommendation Dataset’ provided by Analyst-2 (analyst-2.ai), based on source dataset retrieved from https://www.kaggle.com/siddharthss/crop-recommendation-dataset on 30 September 2021.

--- Dataset description provided by original source is as follows ---

THE INFORMATION IN THE DATASET IS PROVIDED TO THE BEST OF KNOWLEDGE OF ICAR. THE BELOW DATA CAN BE USED PUBLICALLY UNDER ALL PUBLIC AND PRIVATE UNDERTAKINGS

Context Precision agriculture is in trend nowadays. It helps the farmers to get informed decision about the farming strategy. Here, we present you a dataset which would allow the users to build a predictive model to recommend the most suitable crops to grow in a particular farm based on various parameters.**

Source This dataset was build by augmenting datasets of rainfall, climate and fertilizer data available for India. Gathered over the period by ICFA, India.

Data fields N - ratio of Nitrogen content in soil P - ratio of Phosphorous content in soil K - ratio of Potassium content in soil temperature - temperature in degree Celsius humidity - relative humidity in % ph - ph value of the soil rainfall - rainfall in mm

COPYRIGHT: Indian Chamber of Food and Agriculture https://www.icfa.org.in/ https://www.google.com/url?sa=i&url=https%3A%2F%2F10times.com%2Fcompany%2Findian-council-of-food-and-agriculture&psig=AOvVaw0S9UpuXsVmmje0SgSjybK5&ust=1622035121203000&source=images&cd=vfe&ved=0CAIQjRxqFwoTCNie_uv15PACFQAAAAAdAAAAABAn" alt="">

--- Original source retains full ownership of the source dataset ---

Soil Classification Datasets

Please ensure to cite the paper when utilizing the dataset in a research study. Refer to the paper link or BibTeX provided below.

This repository contains comprehensive datasets for soil classification and recognition research. The Original Dataset comprises soil images sourced from various online repositories, which have been meticulously cleaned and preprocessed to ensure data quality and consistency. To enhance the dataset's size and diversity, we employed Generative Adversarial Networks (GANs), specifically the CycleGAN architecture, to generate synthetic soil images. This augmented collection is referred to as the CyAUG Dataset. Both datasets are specifically designed to advance research in soil classification and recognition using state-of-the-art deep learning methodologies.

This dataset was curated as part of the research study titled "An advanced artificial intelligence framework integrating ensembled convolutional neural networks and Vision Transformers for precise soil classification with adaptive fuzzy logic-based crop recommendations" by Farhan Sheth, Priya Mathur, Amit Kumar Gupta, and Sandeep Chaurasia, published in Engineering Applications of Artificial Intelligence.

Links

• Paper: An advanced artificial intelligence framework integrating ensembled convolutional neural networks and Vision Transformers for precise soil classification with adaptive fuzzy logic-based crop recommendations
• Code: Development Code (model and recommendation system)

Application produced by this research is available at:

• Agro Companion: Agro Companion
• Server Code: Agro Companion Server

Note: If you are using any part of this project; dataset, code, application, then please cite the work as mentioned in the Citation section below.

Dataset

Both dataset consists of images of 7 different soil types.

The Soil Classification Dataset is structured to facilitate the classification of various soil types based on images. The dataset includes images of the following soil types:

• Alluvial Soil
• Black Soil
• Laterite Soil
• Red Soil
• Yellow Soil
• Arid Soil
• Mountain Soil

The dataset is organized into folders, each named after a specific soil type, containing images of that soil type. The images vary in resolution and quality, providing a diverse set of examples for training and testing classification models.

Original Dataset Details

• Total Images: 1189 images
• Image Format: JPG/JPEG
• Image Size: Varies
• Source: Collected from various online repositories and cleaned for consistency.

CyAUG Dataset Details

• Total Images: 5097 images
• Image Format: JPG/JPEG
• Image Size: Varies
• Source: Generated using CycleGAN to augment the original dataset, enhancing its size and diversity.

Input and Output Parameters

• Input Parameters:
  • Image: The images of the soils (JPG/JPEG format).
  • Label: The labels are in the format 'soil types' (folder names).
• Output Parameter:
  • Classification: The predicted class (soil type) based on the input image.

Citation

If you are using any of the derived dataset, please cite the following paper:

@article{SHETH2025111425,
  title = {An advanced artificial intelligence framework integrating ensembled convolutional neural networks and Vision Transformers for precise soil classification with adaptive fuzzy logic-based crop recommendations},
  journal = {Engineering Applications of Artificial Intelligence},
  volume = {158},
  pages = {111425},
  year = {2025},
  issn = {0952-1976},
  doi = {https://doi.org/10.1016/j.engappai.2025.111425},
  url = {https://www.sciencedirect.com/science/article/pii/S0952197625014277},
  author = {Farhan Sheth and Priya Mathur and Amit Kumar Gupta and Sandeep Chaurasia},
  keywords = {Soil classification, Crop recommendation, Vision transformers, Convolutional neural network, Transfer learning, Fuzzy logic}
}

This dataset contains agricultural data for 1,000,000 samples aimed at predicting crop yield (in tons per hectare) based on various factors. The dataset can be used for regression tasks in machine learning, especially for predicting crop productivity.

• Region: The geographical region where the crop is grown (North, East, South, West).
• Soil_Type: The type of soil in which the crop is planted (Clay, Sandy, Loam, Silt, Peaty, Chalky).
• Crop: The type of crop grown (Wheat, Rice, Maize, Barley, Soybean, Cotton).
• Rainfall_mm: The amount of rainfall received in millimeters during the crop growth period.
• Temperature_Celsius: The average temperature during the crop growth period, measured in degrees Celsius.
• Fertilizer_Used: Indicates whether fertilizer was applied (True = Yes, False = No).
• Irrigation_Used: Indicates whether irrigation was used during the crop growth period (True = Yes, False = No).
• Weather_Condition: The predominant weather condition during the growing season (Sunny, Rainy, Cloudy).
• Days_to_Harvest: The number of days taken for the crop to be harvested after planting.
• Yield_tons_per_hectare: The total crop yield produced, measured in tons per hectare.

This dataset was transcribed and analyzed to support the article:Obrycki, J. F., and D. L. Karlen. 2018. “Optimizing Iowa Land Use: Past Perspectives for Current Questions.” Journal of Soil and Water Conservation 73 (6): 693–704. https://doi.org/10.2489/jswc.73.6.693The dataset contains transcribed tables from the county pamphlets “How To Use Your Soil And Keep It Too” distributed by county agricultural planning committees in 1937/1938. There are 90 pamphlets available. One table was included in each pamphlet and was titled “Land Use and Cropping Recommendations For Various Soil Types.” The table included recommended crop rotations by soil type in the county. These tables were created by county agricultural planning committees. The values in the tables were committee recommendations for the types of crop rotations that would protect soils and provide suitable crop production. The dataset contains additional columns that evaluated the county agricultural planning committee recommendations. Two supporting datasets are provided to explain the column assignments using in the primary dataset.

Cropland Index

The Cropland Index evaluates lands used to produce crops based on the following input datasets: Revised Storie Index, California Important Farmland data, Electrical Conductivity (EC), and Sodium Adsorption Ratio (SAR). Together, these input layers were used in a suitability model to generate this raster. High values are associated with better Croplands

This dataset includes soil health, crop biomass, and crop yield data for a 13-year corn stover harvest trial in central Iowa. Following the release in 2005 of the Billion Ton Study assessment of biofuel sources, several soil health assessments associated with harvesting corn stover were initiated across ARS locations to help provide industry guidelines for sustainable stover harvest. This dataset is from a trial conducted by the National Laboratory for Agriculture and Environment from 2007-2021 at the Iowa State University Ag Engineering and Agronomy farm. Management factors evaluated in the trial included the following. Stover harvest rate at three levels: No, moderate (3.5 ± 1.1 Mg ha-1 yr-1), or high (5.0 ± 1.7 Mg ha-1 yr-1) stover harvest rates. No-till versus chisel-plow tillage. Originally, the 3 stover harvest rates were evaluated in a complete factorial design with tillage system. However, the no-till, no-harvest system performed poorly in continuous corn and was discontinued in 2012 due to lack of producer interest. Cropping sequence. In addition to evaluating continuous corn for all stover harvest rates and tillage systems, a corn-alfalfa rotation, and a corn-soybean-wheat rotation with winter cover crops were evaluated in a subset of the tillage and stover harvest rate treatments. One-time additions of biochar in 2013 at rates of either 9 Mg/ha or 30 Mg/ha were evaluated in a continuous corn cropping system. The dataset includes: 1) Crop biomass and yields for all crop phases in every year. 2) Soil organic carbon, total carbon, total nitrogen, and pH to 120 cm depth in 2012, 2016, and 2017. Soil cores from 2005 (pre-study) were also sampled to 90 cm depth. 3) Soil chemistry sampled to 15 cm depth every 1-2 years from 2007 to 2017. 4) Soil strength and compaction was assessed to 60 cm depth in April 2021. These data have been presented in several manuscripts, including Phillips et al. (in review), O'Brien et al. (2020), and Obrycki et al. (2018). Resources in this dataset:Resource Title: R Script for Phillips et al. 2022. File Name: Field 70-71 Analysis Script_AgDataCommons.RResource Description: This R script includes analysis and figures for Phillips et al. "Thirteen-year Stover Harvest and Tillage Effects on Soil Compaction in Iowa". It focuses primarily on the soil compaction and strength data found in "Field 70-71 ConeIndex_BulkDensityDepths_2021". It also includes analysis of corn yields from "Field 70-71 CornYield_2008-2021" and weather conditions from "PRISM_MayTemps" and "Rainfall_AEA".Resource Software Recommended: R version 4.1.3 or higher,url: https://cran.r-project.org/bin/windows/base/ Resource Title: Field 70-71 ConeIndex_BulkDensityDepths_2021. File Name: Field 70-71 ConeIndex_BulkDensityDepths_2021.csvResource Description: This dataset provides an assessment of soil strength (penetration resistance) and soil compaction (bulk density) to 60 cm depth, in continuous corn plots. Penetration resistance was measured in most-trafficked and least-trafficked areas of the plots to assess compaction from increased traffic associated with stover harvest. This spreadsheet also has associated data, including soil water, carbon, and organic matter content. Data were collected in April 2021 and are described in Phillips et al. (in review, 2022).Resource Title: Field 70-71 CornYield_2008-2021. File Name: Field 70-71 CornYield_2008-2021_ForR.csvResource Description: This dataset provides corn stover biomass and grain yields from 2008-2021. Note that this dataset is just for corn, which were presented in Phillips et al., 2022. Yields for all crop phases, including soybeans, wheat, alfalfa, and winter cover crops, are in the file "Field 70-71 Crop Yield File 2008-2020".Resource Title: PRISM_MayTemps. File Name: PRISM_MayTemps.csvResource Description: Average May temperatures during the study period, obtained from interpolation of regional weather stations using the PRISM climate model (https://prism.oregonstate.edu/). These data were used to evaluate how spring temperatures may have impacted corn establishment.Resource Title: Rainfall_AEA. File Name: Rainfall_AEA.csvResource Description: Daily rainfall for the study location, 2008-2021. Data were obtained from the Iowa Environmental Mesonet (https://mesonet.agron.iastate.edu/rainfall/). Title: Field 70-71 Plot Status 2007-2021. File Name: Field 70-71 Plot Status 2007-2021.xlsxResource Description: This file contains descriptions of experimental treatments and diagrams of plot layouts as they were modified through several phases of the trial. Also includes an image of plot locations relative to NRCS soil survey map units.Resource Title: Field 70-71 Deep Soil Cores 2012-2017. File Name: Field 70-71 Deep Soil Cores 2012-2017.xlsxResource Description: Soil carbon, nitrogen, organic matter, and pH to 120 cm depth in 2012, 2016, and 2017.Resource Title: Field 70-71 Baseline Deep Soil Cores 2005. File Name: Field 70-71 Baseline Deep Soil Cores 2005.csvResource Description: Baseline soil carbon, nitrogen, and pH data from an earlier trial in 2005, prior to stover trial establishment.Resource Title: Field 70-71 Crop Yield File 2008-2020. File Name: Field 70-71 Crop Yield File 2008-2020.xlsxResource Description: Yields for all crops in all cropping sequences, 2008-2020. Some of the crop sequences have not been summarized in publications.Resource Title: Field 70-71 Surface Soil Test Data 2007-2021. File Name: Field 70-71 Surface Soil Test Data 2007-2021.xlsxResource Description: Soil chemistry data, 0-15 cm, collect near-annually from 2007 to 2021. Most analyses were performed by Harris Laboratories (now AgSource) in Lincoln, Nebraska, USA. Resource Title: Iowa Stover Harvest Trial Data Dictionary. File Name: Field 70-71 Data Dictionary.xlsxResource Description: Data dictionary for all data files.

This data comprises processed weather, soil, yield, and cultivation area for corn yield prediction in Sub-Sahara Africa, with emphasis on Nigeria. The data was collected to design a corn yield prediction model to help smallholder farmers make smart farming decisions. However, the data can serve several other purposes through analysis and interpretation.

The reference study region in Africa is Nigeria. The focuses on corn crop because there are over 211.4 million people, of which a large percentage of the population are smallholder farmers. Nigeria [9.0820° N, 8.6753° E] is within an arable land area of 34 million hectares located on the west coast of Africa. The region comprises of 36 states with the most and least number of districts being 214 and 10, respectively. For each state, the environment data are collected as follows.

Grid map climate data – This data spans spatial resolutions between ~1 km2 to ~340 km2 from the high spatial resolution WorldClim global climate database22. Each grid point on the map is monthly data from January to December between 1970 and 2000 years and records 8 climate variables. The variables are average temperature C0, minimum temperature C0, maximum temperature C0, precipitation (mm), solar radiation (kJ m^(-2) day(-1), wind speed (m s(-1)), and water vapor (kPa) taken at 30 seconds (s), 2.5 minutes m, 5 m, and 10 m.

Grid map soil data – This data is obtained from 250 minutes of spatial resolution AfSIS soil data23 from year 1960 to 2012. The variables are wet soil bulk density, dry bulk density (kg dm-3), clay percentage of plant available water content, hydraulic conductivity, the upper limit of plant available water content, the lower limit of, organic matter percentage, pH, sand percentage (g 100 g-1), silt percentage (g 100 g-1) and, clay percentage (g 100 g-1), and saturated volumetric water content variables measured at depths 0–5, 5–10, 10–15, 15–30, 30–45, 45–60, 60–80, 80–100, and 100–120 measured in centimeters (cm).

Corn yield data – This data is available on Kneoma Corporation website24. It ranged from years 1995 to 2006 and consisted of a corn yield of 1000 metric tonnes and a cultivation area of 1000 hectares.

Geolocation coordinates (latitude and longitude) – The geolocation of each of the 36 states with their districts is sampled from Google Maps. The output feds into the Esri-ArcGIS 2.5, a professional geographical software, for extracting the point-cloud values of each environmental variable (weather and soil) at specific geolocation of the 36 states of Nigeria.

Other Descriptions: Data type - Continous and Categorical Dataset Characteristics - Tabular Associated Tasks - Regression Feature Type - Real Number of Instances - 1828 Number of Features: 12

A dataset of 6 paddocks at six sites in Queensland. Data includes paddock boundaries, point data for soil chemistry, EM38, elevation and yield (sorghum, wheat and barley). The dataset collection is includes measurements from 2005 - 2020. The collection includes raw versions of this data and versions which have been pre-processed for Machine Learning analytics.

Crop Variety Recommendation AI Market Outlook

As per our latest research, the global Crop Variety Recommendation AI market size reached USD 1.82 billion in 2024, with a robust growth trajectory supported by a CAGR of 18.5% from 2025 to 2033. By the end of 2033, the market is forecasted to attain a value of USD 9.87 billion. The rapid adoption of artificial intelligence in agriculture, driven by the need for increased productivity, sustainability, and precision in crop selection, is fueling this remarkable expansion. The integration of AI-based recommendations is transforming traditional farming methodologies, ensuring optimal crop choices tailored to specific climatic, soil, and market conditions.

The primary growth driver for the Crop Variety Recommendation AI market is the rising global demand for food security amidst changing climatic conditions and shrinking arable land. With the worldÂ’s population projected to surpass 9 billion by 2050, agricultural stakeholders are under increasing pressure to maximize yields and minimize resource wastage. AI-powered crop recommendation systems leverage vast datasets, including historical yield data, weather patterns, soil health, and pest prevalence, to provide actionable insights. This significantly enhances farmersÂ’ decision-making processes, enabling them to select crop varieties that are both high-yielding and resilient to biotic and abiotic stresses. The resulting improvements in productivity and profitability are compelling both smallholder and large-scale farmers to invest in AI-driven solutions.

Another critical factor propelling market growth is the increasing government and institutional support for digital transformation in agriculture. Many countries are launching initiatives and funding programs aimed at modernizing the agri-food sector, with AI and data analytics at the core of these strategies. Governments, in collaboration with research institutes and agri-tech companies, are investing in the development and deployment of intelligent crop recommendation platforms to support farmers in optimizing their crop selection, reducing input costs, and promoting sustainable practices. These public-private partnerships are accelerating the penetration of Crop Variety Recommendation AI, particularly in regions with significant agricultural output and export potential.

The proliferation of affordable smart devices and IoT sensors in rural and semi-urban areas is also contributing to the rapid adoption of AI-based crop recommendation systems. Farmers now have access to real-time data on soil moisture, nutrient levels, weather forecasts, and pest threats, all of which can be seamlessly integrated into AI platforms for more accurate recommendations. The convergence of AI with precision agriculture tools, such as drones and satellite imagery, is further enhancing the effectiveness of these solutions. As digital literacy improves and connectivity expands, even smallholder farmers in developing regions are increasingly able to harness the benefits of AI-driven crop variety recommendations, democratizing access to advanced agronomic expertise.

AI-Powered Crop Yield Forecasting is becoming an integral part of modern agriculture, offering unprecedented accuracy in predicting crop outputs. By leveraging machine learning algorithms and historical data, these systems can analyze variables such as weather patterns, soil conditions, and crop health to forecast yields with remarkable precision. This capability is crucial for optimizing resource allocation, planning harvests, and managing supply chains effectively. As farmers and agribusinesses increasingly rely on AI-driven insights, the ability to anticipate yield fluctuations and adapt strategies accordingly is enhancing resilience against climate variability and market volatility. The integration of AI-powered forecasting tools is not only improving productivity but also contributing to more sustainable agricultural practices by reducing waste and optimizing input use.

Regionally, Asia Pacific stands out as the fastest-growing market, driven by the regionÂ’s vast agricultural base, rapid digitalization, and proactive government policies supporting smart farming. North America and Europe follow closely, with established agri-tech ecosystems and high rates of technology adoption among comm

CY-Bench: A comprehensive benchmark dataset for sub-national crop yield forecasting

Overview

CY-Bench is a dataset and benchmark for subnational crop yield forecasting, with coverage of major crop growing countries of the world for maize and wheat. By subnational, we mean the administrative level where yield statistics are published. When statistics are available for multiple levels, we pick the highest resolution. The dataset combines sub-national yield statistics with relevant predictors, such as growing-season weather indicators, remote sensing indicators, evapotranspiration, soil moisture indicators, and static soil properties. CY-Bench has been designed and curated by agricultural experts, climate scientists, and machine learning researchers from the AgML Community, with the aim of facilitating model intercomparison across the diverse agricultural systems around the globe in conditions as close as possible to real-world operationalization. Ultimately, by lowering the barrier to entry for ML researchers in this crucial application area, CY-Bench will facilitate the development of improved crop forecasting tools that can be used to support decision-makers in food security planning worldwide.

* Crops : Wheat & Maize
* Spatial Coverage : Wheat (29 countries), Maize (38).
See CY-Bench paper appendix for the list of countries.
* Temporal Coverage : Varies. See country-specific data

Data

Data format

The benchmark data is organized as a collection of CSV files, with each file representing a specific category of variable for a particular country. Each CSV file is named according to the category and the country it pertains to, facilitating easy identification and retrieval. The data within each CSV file is structured in tabular format, where rows represent observations and columns represent different predictors related to a category of variable.

Data content

All data files are provided as .csv.

Folder structure

The CY-Bench dataset has been structure at first level by crop type and subsequently by country. For each country, the folder name follows the ISO 3166-1 alpha-2 two-character code. A separate .csv is available for each predictor data and crop calendar as shown below. The csv files are named to reflect the corresponding country and crop type e.g. **variable_croptype_country.csv**.
```
CY-Bench
│
└─── maize
│ │
│ └─── AO
│ │ -- crop_calendar_maize_AO.csv
│ │ -- fpar_maize_AO.csv
│ │ -- meteo_maize_AO.csv
│ │ -- ndvi_maize_AO.csv
│ │ -- soil_maize_AO.csv
│ │ -- soil_moisture_maize_AO.csv
│ │ -- yield_maize_AO.csv
│ │
│ └─── AR
│ -- crop_calendar_maize_AR.csv
│ -- fpar_maize_AR.csv
│ -- ...
│
└─── wheat
│ │
│ └─── AR
│ │ -- crop_calendar_wheat_AR.csv
│ │ -- fpar_wheat_AR.csv
│ │ ...
```

Example : CSV data content for maize in country X

```
X
└─── crop_calendar_maize_X.csv
│ -- crop_name (name of the crop)
│ -- adm_id (unique identifier for a subnational unit)
│ -- sos (start of crop season)
│ -- eos (end of crop season)
│
└─── fpar_maize_X.csv
│ -- crop_name
│ -- adm_id
│ -- date (in the format YYYYMMdd)
│ -- fpar
│
└─── meteo_maize_X.csv
│ -- crop_name
│ -- adm_id
│ -- date (in the format YYYYMMdd)

│ -- tmin (minimum temperature)
│ -- tmax (maximum temperature)
│ -- prec (precipitation)
│ -- rad (radiation)
│ -- tavg (average temperature)
│ -- et0 (evapotranspiration)
│ -- cwb (crop water balance)
│
└─── ndvi_maize_X.csv
│ -- crop_name
│ -- adm_id
│ -- date (in the format YYYYMMdd)
│ -- ndvi
│
└─── soil_maize_X.csv
│ -- crop_name
│ -- adm_id
│ -- awc (available water capacity)
│ -- bulk_density
│ -- drainage_class
│
└─── soil_moisture_maize_X.csv
│ -- crop_name
│ -- adm_id
│ -- date (in the format YYYYMMdd)
│ -- ssm (surface soil moisture)
│ -- rsm ()
│
└─── yield_maize_X.csv
│ -- crop_name
│ -- country_code
│ -- adm_id
│ -- harvest_year
│ -- yield
│ -- harvest_area
│ -- production

Data access

The full dataset can be downloaded directly from Zenodo or using the ```zenodo_get``` library

License and citation

We kindly ask all users of CY-Bench to properly respect licensing and citation conditions of the datasets included.

Although soil and agronomy data collection in Ethiopia has begun over 60 years ago, the data are hardly accessible as they are scattered across different organizations, mostly held in the hands of individuals (Ashenafi et al.,2020; Tamene et al.,2022), which makes them vulnerable to permanent loss. Cognizant of the problem, the Coalition of the Willing (CoW) for data sharing and access was created in 2018 with joint support and coordination of the Alliance Bioversity-CIAT and GIZ (https://www.ethioagridata.com/index.html). Mobilizing its members, the CoW has embarked on data rescue operations including data ecosystem mapping, collation, and curation of the legacy data, which was put into the central data repository for its members and the wider data user’s community according to the guideline developed based on the FAIR data principles and approved by the CoW. So far, CoW managed to collate and rescue about 20,000 legacy soil profile data and over 38,000 crop responses to fertilizer data (Tamene et al.,2022). The crop response dataset (N=1,550 observations) is extracted, transformed, and uploaded into a harmonized template, consisting of 76 variables. Recent efforts by the Federal and Regional research centres in collaboration with the MoA, RBoA’s and ATA have shown that there was a significant potassium deficiency in significant agricultural lands of the country. Potassium deficiency was observed through soil fertility assessment surveys and crop response studies.Hence, the promotion of potassium fertilizer use in the agricultural system would be of great importance to increase the balanced fertilizer use system in the country.

In the year 2016/ 2017, a project known as “Large Scale Popularization of Potassium Fertilizer Use in Ethiopia” was implemented from October 2015 to March 2017 by Sasakawa Africa Association/Sasakawa Global 2000 in collaboration with the Ministry of Agriculture, ATA, AGRA and other stakeholders. To achieve the set goals and objectives of the project, in the 2016/2017 cropping season, 18,203 KCL demonstrations were implemented in the four project regions, Amhara, Oromia, SNNPRs and Tigray on five crops, Teff, wheat, Maize, Barley and Sesame. Accordingly, voluminous crop response to the fertilizer dataset was generated by this project.

Reference: Ashenafi, A., Tamene, L., and Erkossa, T. 2020. Identifying, Cataloguing, and Mapping Soil and Agronomic Data in Ethiopia. CIAT Publication No. 506. International Center for Tropical Agriculture (CIAT). Addis Ababa, Ethiopia. 42 p. 10.13140/RG.2.2.31759.41123. Tamene L; Erkossa T; Tafesse T; Abera W; Schultz S. 2021. A coalition of the Willing - Powering data-driven solutions for Ethiopian Agriculture. CIAT Publication No. 518. International Center for Tropical Agriculture (CIAT). Addis Ababa, Ethiopia. 34 p. https://www.ethioagridata.com/Resources/Powering%20Data-Driven%20Solutions%20for%20Ethiopian%20Agriculture.pdf. The Coalition of the Willing (CoW) website: https://www.ethioagridata.com/index.html. TERMS: Access to the data is limited to the CoW members until the national soil and agronomy data-sharing directive of MoA is registered by the Ministry of Justice and released for implementation. DISCLAIMER: The dataset populated in the harmonized template consisting of 76 variables is extracted, transformed, and uploaded from the source document by the CoW. Hence, if any irregularities are observed, the data users have referred to the source document uploaded along with the dataset. Use of the dataset and any consequences arising from using it is the user’s sole responsibility.

Although soil and agronomy data collection in Ethiopia has begun over 60 years ago, the data are hardly accessible as they are scattered across different organizations, mostly held in the hands of individuals (Ashenafi et al.,2020; Tamene et al.,2022), which makes them vulnerable to permanent loss. Cognizant of the problem, the Coalition of the Willing (CoW) for data sharing and access was created in 2018 with joint support and coordination of the Alliance Bioversity-CIAT and GIZ (https://www.ethioagridata.com/index.html). Mobilizing its members, the CoW has embarked on data rescue operations including data ecosystem mapping, collation, and curation of the legacy data, which was put into the central data repository for its members and the wider data user’s community according to the guideline developed based on the FAIR data principles and approved by the CoW. So far, CoW managed to collate and rescue about 20,000 legacy soil profile data and over 38,000 crop responses to fertilizer data (Tamene et al.,2022). The crop response dataset (N=1,686 observations) is extracted, transformed, and uploaded into a harmonized template, consisting of 76 variables. Supporting Soil Health Intervention (SSHI) crop response dataset. Reference: Ashenafi, A., Tamene, L., and Erkossa, T. 2020. Identifying, Cataloguing, and Mapping Soil and Agronomic Data in Ethiopia. CIAT Publication No. 506. International Center for Tropical Agriculture (CIAT). Addis Ababa, Ethiopia. 42 p. 10.13140/RG.2.2.31759.41123. Tamene L; Erkossa T; Tafesse T; Abera W; Schultz S. 2021. A coalition of the Willing - Powering data-driven solutions for Ethiopian Agriculture. CIAT Publication No. 518. International Center for Tropical Agriculture (CIAT). Addis Ababa, Ethiopia. 34 p. https://www.ethioagridata.com/Resources/Powering%20Data-Driven%20Solutions%20for%20Ethiopian%20Agriculture.pdf. The Coalition of the Willing (CoW) website: https://www.ethioagridata.com/index.html. TERMS: Access to the data is limited to the CoW members until the national soil and agronomy data-sharing directive of MoA is registered by the Ministry of Justice and released for implementation. DISCLAIMER: The dataset populated in the harmonized template consisting of 76 variables is extracted, transformed, and uploaded from the source document by the CoW. Hence, if any irregularities are observed, the data users have referred to the source document uploaded along with the dataset. Use of the dataset and any consequences arising from using it is the user’s sole responsibility.

SNAP (Soil Nutrient Assessment Program), a component of the USDA/ARS Soil and Water Hub, is a web-based tool that provides an estimate of plant-available nutrients that the soil naturally provides. Soil test fertilizer recommendations have long been predicated upon response curves generated from fertility trials across the country. These response curves have been compared to relative yield which provide probability ranges for a response to varying fertilizer inputs. Category responses include very low, low, adequate, high or very high inversely related to probability of a response to various inputs of nitrogen, phosphate, and potassium (N, P, and K). New soil test methods, increases in computing power and access to the internet have enabled development of an interactive tool that is based on plant available NPK from both the inorganic fraction and organic pool of the soil. The new methods provide an estimate of plant available nutrients that the soil naturally provides, which has largely been ignored for decades. Since we have access to large datasets we can calculate the amounts of NPK required growing crops in lbs NPK per bu of the desired crop. For example, it requires 100 lbs of N, 50 lbs P2O5, 50 lbs K2O to grow 100 bu corn. These are the base numbers from which we subtract the soil test data after converting from the analytical ppm to Lbs P2O5 or lbs K2O. This is a straight subtraction. It also eliminates the need for "calibration data" since the soil tests reflect the soils inherent fertility. Using the example above, of 100, 50, 50 of N, P, and K required and soil test results of 25, 35, 45 then the fertilizer needed would be 75 N, 15 P2O5 and 5 K2O. This is a simple approach that doesn't get lost in relative yield-crop response curves that have been used for decades from differing geographical areas. This tool will include current fertilizer prices, soil test inputs, and crop based county averages for the last 15 years that will predict the chances of making the yield goal the user inputs compared to historical yield data for their county and calculate the fertilizer cost with and without soil testing compared to user input yield goal and county average. This tool will allow the user via the internet to produce a more straightforward approach to realistically planning next year's fertilizer inputs and associated cost. It will also show the benefits of soil testing for increased fertilizer efficiency and reduced environmental impact. Resources in this dataset:Resource Title: Website Pointer to SNAP - Soil Nutrient Assessment Program. File Name: Web Page, url: https://snap.brc.tamus.edu/Home/Index The web dashboard interface for estimating local yield based on field location (state/county), crop (, area, and yield goal; and soil NPK test results (lb/acre), Results returned illustrate local yield, fertilizer cost/acre, fertilizer needed (lb/acre), and overall chance of success (%).

The Soil and Terrain Database for Northeastern Africa contains land resource information on soils, physiography, geology and vegetation for the following ten countries: Burundi, Djibouti, Egypt, Eritrea, Ethiopia, Kenya, Rwanda, Somalia, Sudan and Uganda. The information is accessible with an easy-to-use viewer program and is also stored in vector Arc/Info export format. Information on individual soil properties with class values is also given. A land suitability assessment for irrigated and upland crops for each unit is included. The scale ofthe source material is variable and ranges between 1:1 million and 1:2 million. A user manual for the viewer program and background information on the collected and correlated land resource materials are contained in filed documents.

Soils are classified in the Revised Legend; physiographic and lithology information was collected using an earlier draft version of the SOTER manual.

The Inter-Governmental Authority on Development (IGAD) -- Sudan, Kenya, Djibouti, Somalia, Uganda, Eritrea, Ethiopia -- Crop Production System Zones (CPSZ) software is a detailed database that provides background information about actual farming in the region. It comes with a program (CVIEW, a CPSZ viewer) that displays maps, zooms in and out, and provides export facilities for the maps in image format and for the actual data in text format. The elementary mapping unit is a compromise between administrative units and agro-ecological zones: whenever steep agro-ecological gradients exist, administrative units are subdivided, thus resulting in 1200 mapping units that are homogeneous from an agro-ecological point of view, while retaining the compatibility with the administrative units used for most socio-economic variables in agricultural planning.

The just over 500 mappable variables are subdivided into several categories covering the spectrum from agronomy and livestock to the physical environment. For each mapping unit, detailed information is also presented on the crop calendar, typical yields and main pests and diseases.

This CD-ROM contains a collection of land and natural resource information for Northeastern Africa, in particular for the IGAD countries bordering the Nile basin. It includes data on administrative boundaries, rivers and lakes, soil and terrain, climatology, land use, physiography, geology and natural vegetation in easily accessible format.

Soil and Terrain Database for Northeasterm Africa (1:1 Million Scale) and Crop Production System Zones of the IGAD Subregion is provided on CD-ROM by the FAO, Land and Water Digital Media Series (Number 2). The CD-ROM can be purchased (Price: US$40) from FAO, Sales and Marketing Group, Viale delle Terme di Caracalla 0100 Rome, Italy (Fax: +39-06-5705-3360 E-mail: publications-sales@fao.org).

Under the AI4SoilHealth project we have created a dataset (“RapidCrops”) to support the automatic mapping of crop types across Europe. Crop type information is essential for monitoring soil health as it provides systematic insights into crop rotations over time and supports efforts to detect other cropping practices that affect soil health (e.g. tillage & cover crops). The RapidCrops dataset provides approximately 99M agricultural parcel boundaries with harmonised crop type information across a wide spatio-temporal extent; with coverage across seven EU countries for 5-7 years. Based on parcel boundaries and crop type information reported under the EU IACS programme, our methodology seeks to improve the usability of the parcel boundaries without diluting their integrity. Additional attributes are provided to support the use of the data in ML workflows; especially for those leveraging EO data. The dataset builds on top of the EuroCrops [1,2] crop type harmonisation initiative and the fiboa [3] open data standard for parcel boundaries. For enhanced access to the data, the dataset is also made freely available on Source Cooperative [4]. The dataset was also utilised under the Horizon Europe project Open-Earth-Monitor to perform pan-European crop identification across 51M parcels from the year 2022; classifying each parcel into one of 29 crop types [5]. Please see the license terms for underlying datasets below: Data source Data licensing terms Austria INSPIRE public access license & CC-BY-AT 4.0 Denmark INSPIRE public access license & INSPIRE no conditions & CC0 1.0 Universal France Custom open license Germany Custom open licenses: NRW, Brandenberg, LS Netherlands INSPIRE public access license & INSPIRE no conditions & Dutch creative commons license Portugal CC BY 4.0 Spain Custom open license

The Crops Suitability Tool combines soil types, aspect (slope orientation), and percentage of slope to determine the best and least suitable sites in which to grow crops in Loudoun County. It includes different types of Agricultural Soils (Prime Farmland, Secondary Cropland, Grassland Agriculture, Orchard Land, Woodland Use and Wildlife) and its grade of suitability for grapes, tree fruits, hops, vegetables, flowers, herbs, small fruits, field crops, pasture and hay.A spatial model uses existing geographic data to predict an outcome. In this application, we combined soil types, aspect (slope orientation), and percentage of slope to determine the best and least suitable site in which to grow crops in Loudoun County, Virginia. It includes different types of Agricultural Soils (Prime Farmland, Secondary Cropland, Grassland Agriculture, Orchard Land, Woodland Use and Wildlife) and its grade of suitability for grapes, tree fruits, hops, vegetables (ethnic crops), flowers, herbs, and small fruits, field crops, pasture, and hay.This tool does not account for the incidence and prevalence of any type of pests (weed, insects, and diseases -nematodes, fungi, bacteria, or viruses) or weather conditions that can affect crops. The accuracy of the predicted outcomes is not 100% (for example: 17B soils in a concave position are not suitable for growing perennial crops or high cash valued crops); therefore, it is highly recommended to contact VCE Loudoun Commercial Horticulturist Beth Sastre to get a soil map report of the property and/or to have a site evaluation for further recommendation.We encourage farmers, beginner farmers, people interested in farming, and realtors to use this tool to make guided decisions before starting a crop for the first time or buying land. If you see major discrepancies while using this tool, please report them.

Although soil and agronomy data collection in Ethiopia has begun over 60 years ago, the data are hardly accessible as they are scattered across different organizations, mostly held in the hands of individuals (Ashenafi et al.,2020; Tamene et al.,2022), which makes them vulnerable to permanent loss. Cognizant of the problem, the Coalition of the Willing (CoW) for data sharing and access was created in 2018 with joint support and coordination of the Alliance Bioversity-CIAT and GIZ (https://www.ethioagridata.com/index.html). Mobilizing its members, the CoW has embarked on data rescue operations including data ecosystem mapping, collation, and curation of the legacy data, which was put into the central data repository for its members and the wider data user’s community according to the guideline developed based on the FAIR data principles and approved by the CoW. So far, CoW managed to collate and rescue about 20,000 legacy soil profile data and over 38,000 crop responses to fertilizer data (Tamene et al.,2022). The crop response dataset (N=946 observations) is extracted, transformed, and uploaded into a harmonized template, consisting of 76 variables.

The CASCAPE project was designed to support the Agricultural Growth Programme (AGP) by further strengthening the capacity of AGP stakeholders in identifying, documenting and disseminating best practices in agricultural production. CASCAPE was jointly executed by Ethiopian researchers from Jimma University, Bahir Dar University, Hawassa University, Mekelle University, Addis Ababa University and Dutch researchers from Wageningen University and Research Centre. In each site researchers from the universities and from the RARIs from different disciplines work on the CASCAPE project. The CASCAPE project is financed by the Dutch Ministry of Foreign Affairs through the Embassy of the Kingdom of The Netherlands (https://www.wur.nl/en/show/cascape-1.htm). To solve the soil fertility and plant nutrition problems in Ethiopia, CASCAPE Project conducted various soil surveys and fertilizer experiments, during the period of 2011 to 2019, and generated crop responses to the fertilizer application dataset covering Ethiopia.

Reference: Ashenafi, A., Tamene, L., and Erkossa, T. 2020. Identifying, Cataloguing, and Mapping Soil and Agronomic Data in Ethiopia. CIAT Publication No. 506. International Center for Tropical Agriculture (CIAT). Addis Ababa, Ethiopia. 42 p. 10.13140/RG.2.2.31759.41123. Tamene L; Erkossa T; Tafesse T; Abera W; Schultz S. 2021. A coalition of the Willing - Powering data-driven solutions for Ethiopian Agriculture. CIAT Publication No. 518. International Center for Tropical Agriculture (CIAT). Addis Ababa, Ethiopia. 34 p. https://www.ethioagridata.com/Resources/Powering%20Data-Driven%20Solutions%20for%20Ethiopian%20Agriculture.pdf. The Coalition of the Willing (CoW) website: https://www.ethioagridata.com/index.html. TERMS: Access to the data is limited to the CoW members until the national soil and agronomy data-sharing directive of MoA is registered by the Ministry of Justice and released for implementation. DISCLAIMER: The dataset populated in the harmonized template consisting of 76 variables is extracted, transformed, and uploaded from the source document by the CoW. Hence, if any irregularities are observed, the data users have referred to the source document uploaded along with the dataset. Use of the dataset and any consequences arising from using it is the user’s sole responsibility.

Although soil and agronomy data collection in Ethiopia has begun over 60 years ago, the data are hardly accessible as they are scattered across different organizations, mostly held in the hands of individuals (Ashenafi et al.,2020; Tamene et al.,2022), which makes them vulnerable to permanent loss. Cognizant of the problem, the Coalition of the Willing (CoW) for data sharing and access was created in 2018 with joint support and coordination of the Alliance Bioversity-CIAT and GIZ (https://www.ethioagridata.com/index.html). Mobilizing its members, the CoW has embarked on data rescue operations including data ecosystem mapping, collation, and curation of the legacy data, which was put into the central data repository for its members and the wider data user’s community according to the guideline developed based on the FAIR data principles and approved by the CoW. So far, CoW managed to collate and rescue about 20,000 legacy soil profile data and over 38,000 crop responses to fertilizer data (Tamene et al.,2022). The legacy soil profile dataset (consisting of Profiles Site = 1,776 observations with 37 variables; Profiles Layer Field = 1,493 observations with 64 variables; Profiles Layer Lab= 1,386 observations with 76 variables) is extracted, transformed, and uploaded into a harmonized template (adapted from Batjes 2022; Leenaars et al, 2014) from the below source: Bilateral Ethiopian-Netherlands Effort for Food, Income and Trade (BENEFIT) Partnership which is a portfolio of five programs (ISSD, Cascape, ENTAG, SBN, and REALISE) and is funded by the government of the Kingdom of Netherlands through its embassy in Addis Ababa. The BENEFIT-REALISE program implements its interventions in 60 PSNP weredas in four regions (Tigray, Amhara, Oromia, and SNNPR).Accordingly, in 2019, BENEFIT-REALISE along with the MoA initiated a wereda-wide soil resource characterization and mapping task at1:50,000 scale in 15 BENEFIT-REALISE intervention weredas: 3 of Tigray, 6 of Amhara, 3 of Oromia, and 3 of SNNPR. Reference: Ashenafi, A., Tamene, L., and Erkossa, T. 2020. Identifying, Cataloguing, and Mapping Soil and Agronomic Data in Ethiopia. CIAT Publication No. 506. International Center for Tropical Agriculture (CIAT). Addis Ababa, Ethiopia. 42 p. 10.13140/RG.2.2.31759.41123. Ashenafi, A., Erkossa, T., Gudeta, K., Abera, W., Mesfin, E., Mekete, T., Haile, M., Haile, W., Abegaz, A., Tafesse, D. and Belay, G., 2022. Reference Soil Groups Map of Ethiopia Based on Legacy Data and Machine Learning Technique: EthioSoilGrids 1.0. EGUsphere, pp.1-40. https://doi.org/10.5194/egusphere-2022-301 Tamene L; Erkossa T; Tafesse T; Abera W; Schultz S. 2021. A coalition of the Willing - Powering data-driven solutions for Ethiopian Agriculture. CIAT Publication No. 518. International Center for Tropical Agriculture (CIAT). Addis Ababa, Ethiopia. 34 p. https://www.ethioagridata.com/Resources/Powering%20Data-Driven%20Solutions%20for%20Ethiopian%20Agriculture.pdf. The Coalition of the Willing (CoW) website: https://www.ethioagridata.com/index.html. Batjes, N.H., 2022. Basic principles for compiling a profile dataset for consideration in WoSIS. CoP report, ISRIC–World Soil Information, Wageningen. Contents Summary, 4(1), p.3. Carvalho Ribeiro, E.D. and Batjes, N.H., 2020. World Soil Information Service (WoSIS)-Towards the standardization and harmonization of world soil data: Procedures Manual 2020. Elias, E.: Soils of the Ethiopian Highlands: Geomorphology and Properties, CASCAPE Project, 648 ALTERRA, Wageningen UR, the Netherlands, library.wur.nl/WebQuery/isric/2259099, 649 2016. Leenaars, J. G. B., van Oostrum, A.J.M., and Ruiperez ,G.M.: Africa Soil Profiles Database, Version 1.2. A compilation of georeferenced and standardised legacy soil profile data for Sub Saharan Africa (with dataset), ISRIC Report 2014/01, Africa Soil Information Service (AfSIS) project and ISRIC – World Soil Information, Wageningen, library.wur.nl/WebQuery/isric/2259472, 2014. Leenaars, J. G. B., Eyasu, E., Wösten, H., Ruiperez González, M., Kempen, B.,Ashenafi, A., and Brouwer, F.: Major soil-landscape resources of the cascape intervention woredas, Ethiopia: Soil information in support to scaling up of evidence-based best practices in agricultural production (with dataset), CASCAPE working paper series No. OT_CP_2016_1, Cascape. https://edepot.wur.nl/428596, 2016. Leenaars, J. G. B., Elias, E., Wösten, J. H. M., Ruiperez-González, M., and Kempen, B.: Mapping the major soil-landscape resources of the Ethiopian Highlands using random forest, Geoderma, 361, https://doi.org/10.1016/j.geoderma.2019.114067, 2020a. 740 . Leenaars, J. G. B., Ruiperez, M., González, M., Kempen, B., and Mantel, S.: Semi-detailed soil resource survey and mapping of REALISE woredas in Ethiopia, Project report to the BENEFIT-REALISE programme, December, ISRIC-World Soil Information, Wageningen, 2020b.

TERMS: Access to the data is limited to the CoW members until the national soil and agronomy data-sharing directive of MoA is registered by the Ministry of Justice and released for implementation. DISCLAIMER: The dataset populated in the harmonized template consisting of 76 variables is extracted, transformed, and uploaded from the source document by the CoW. Hence, if any irregularities are observed, the data users have referred to the source document uploaded along with the dataset. Use of the dataset and any consequences arising from using it is the user’s sole responsibility.