Digital Innovations for Agriculture Group (DIAG)

Using data to transform agriculture

Mission: To Enable data intensive science in CALS and beyond.

Vision: Faster and more collaborative research and management through open science.

The University of Arizona Digital Agriculture Group (DIAG) enables open science and data intensive research in the College of Agriculture and Life Sciences at CALS. Our mission is to enable scientists at UA and beyond with their analysis, software, data, and computing needs. Our goal is to enable more efficient discovery and more accurate actionable information through data synthesis and analysis. Our vision is that digital collaboration will enable more effective and sustainable management of systems that produce food, energy, and ecosystem services.

High resolution thermal image of Sorghum bicolor

What We Do

Data Curation

We build software and processes to support data curation and harmonization. We have extensive experience with combining existing databases and extracting results and metadata from publications. We help scientists organize and publish data: from writing a data management plan to publishing data to developing custom databases. We work with CyVerse, the UA libraries, and other data repositories and can help you find the best way to share and get credit for your data. 

Custom Pipelines

We have experience developing custom data processing and analysis pipelines. Our projects include aggregation and processing of weather data, using crop and ecosystem simulation models for inference and prediction, analysis of remote sensing data and more.

Crop & Ecosystem Simulation Models

We have a variety of models appropriate for different applications, from predicting the yield, carbon, and water balance of crop monocultures to understanding plant communities and biogeochemical cycling. We can help parameterize, calibrate, improve and run models including the Ecosystem Demography Model, BioCro, Sipnet, and more. We are also familiar with common analyses used in modeling - including sensitivity analysis, uncertainty propagation, forecasting, data assimilation, and assessing model skill.

Training

Much of what we do, we teach. Most of our lessons are centered on agricultural systems and they range from data science basics to remote sensing and crop and ecosystem modeling.