Irrigation Science
Modeling how water moves through soil and crops to decide when and how much to irrigate.
- Irrigation scheduling
- FAO-56 / PyFAO56
- Soil water balance
- Evapotranspiration
- Rosetta soil properties
- Decision support systems
- Irrigation
- Agriculture
- Technology
Abinava Yeshwanth KJ
Bridging irrigation, agriculture & technology.
I use machine learning and software development to close the gap between field data and farm decisions — building decision support tools that help growers irrigate at the right time, with the right amount of water.
PhD Researcher · Biosystems & Agricultural Engineering, Oklahoma State University
RISE Lab · Dr. Sumon Datta
BridgeML +
Software
Software
- Irrigation scheduling
- Soil moisture
- Machine learning
- Evapotranspiration
- Weather forecasts
- Remote sensing
- Decision support
- Sensor networks
- Water balance
- Sustainable water use
01Skills & expertise
Where irrigation, agriculture and technology meet.
Machine learning and software development are the bridge — connecting field measurements and water science to tools growers can use.
Machine Learning & AI
Predictive models for soil moisture and crop-water use.
- Random Forest
- XGBoost
- LightGBM
- LSTM
- Transformers
- scikit-learn
- TensorFlow
- PyTorch
- OpenCV
- Generative AI
Field Instrumentation
Installing and maintaining the sensors that feed the models.
- Soil moisture sensors
- Rain gauges
- IRT canopy temperature
- Eddy covariance
- ZL6 data loggers
Software Development
Full-stack tools that put models in growers' hands.
- React
- FastAPI
- Flask
- REST / SOAP APIs
- Docker
- AWS
- Git
- Linux
- JavaScript
- C++
- Java
Data Engineering
Pipelines that turn raw weather and sensor data into model-ready inputs.
- Python
- Pandas
- NumPy
- SQL / SQLite
- Plotly
- Batch & parallel processing
- HPC
Weather & Geospatial
Forecasts, station networks and satellite data across the landscape.
- NOAA NDFD / GRIB2
- Oklahoma Mesonet
- Google Earth Engine
- ArcGIS
- OpenET
02Decision support for irrigation
From field data to a clear irrigation decision.
Good irrigation depends on what is happening in the soil, the crop and the sky. A decision support system brings those signals together and turns them into practical guidance for growers.
- 01
Sense
Soil moisture probes, rain gauges, canopy temperature sensors and weather stations measure what is happening in the field.
Field research - 02
Forecast
Weather forecasts and satellite data add what is coming next and what is happening across the wider landscape.
Forecast pipeline - 03
Model
Machine learning and FAO-56 water-balance models estimate crop water use and how much water is left in the root zone.
Soil moisture ML - 04
Decide
A web-based decision support tool turns it all into a clear recommendation: when to irrigate, and how much.
OASIS platform
Feedback from growers and new field data keep improving the models.
03About
A computer scientist working on agricultural water.
My background is in computer science — software, data pipelines and machine learning. Today I apply those skills to a practical question: how can growers use soil, weather and sensor data to decide when and how much to irrigate?
As a PhD researcher in Biosystems & Agricultural Engineering at Oklahoma State University, I build irrigation decision support tools, develop machine-learning models for soil moisture, process weather forecasts for crop-water models, and work in the field installing the sensors that make it all possible.
I'm part of the RISE Lab (Research in Irrigation Scheduling and Engineering), led by Dr. Sumon Datta, where I work on irrigation decision support with him and the team.
Research interests
- Artificial Intelligence
- Machine Learning
- Soil Moisture Modeling
- Irrigation Scheduling
- Decision Support Systems
- Evapotranspiration
- Weather Data Processing
- Remote Sensing
- Geospatial Data
- Sustainable Water Management
The journey
Computer Science
2019 – 2023B.Tech in Computer Science & Engineering — algorithms, systems and the first ML projects.
Software Engineering
2022 – 2023Building web tools and data pipelines as a student assistant and software developer intern.
Machine Learning
2022 – 2025Applying ML to real data — disaster prediction, sentiment analysis, soil fertility — then an MS in Computer Science.
Agricultural Engineering
2024 – nowGraduate research in Biosystems & Agricultural Engineering; PhD from January 2026.
Irrigation & Water
nowSoil moisture, weather and decision-support systems that help producers use water well.
04Research & projects
Building decision support for irrigation.
Projects that connect software, machine learning and field data to everyday irrigation decisions.
01Active development
Decision Support • Irrigation • Full-Stack Engineering
OASIS
Operational Agriculture and Soil Information System
Formerly the Oklahoma Agricultural Scientific Irrigation Scheduler
An irrigation decision-support platform that brings soil, weather and crop information together to give producers irrigation recommendations.
- Automated soil and weather data extraction, using the Rosetta model to estimate soil hydraulic properties.
- React front end and FastAPI back end with interactive Plotly charts for producers.
- REST and SOAP integrations pull weather, soil and other external data sources into one workflow.
- Containerized with Docker, hosted on a DASNR IT virtual machine, with AWS SES email notifications.
- Usability feedback gathered from growers through the Master Irrigator Program.
- React
- FastAPI
- Python
- Plotly
- Rosetta
- PyFAO56
- Mesonet
- REST / SOAP
- Docker
- AWS SES
Schematic
02Active development
Machine Learning • Soil Moisture • Irrigation
Multi-Depth Soil Moisture Prediction
Machine learning for irrigated fields
Predicting soil moisture at several depths so irrigation decisions can be supported where sensors are sparse or absent.
- Built and compared five models — Random Forest, LightGBM, XGBoost, LSTM and Transformer.
- Lag-based temporal features (up to three lags) capture soil-water memory.
- Fully autoregressive prediction without relying on in-situ sensor inputs.
- Scaling toward a model that generalizes across diverse agricultural regions of Oklahoma.
- Python
- scikit-learn
- XGBoost
- LightGBM
- PyTorch
- TensorFlow
- Pandas
Schematic
03Built
Data Engineering • Weather • Agricultural Modeling
NOAA NDFD Forecast Pipeline
From GRIB2 forecasts to agronomic inputs
A Python pipeline that turns high-frequency National Digital Forecast Database files into daily, locally-aligned forecasts for irrigation models.
- Processes GRIB2 forecast files from NOAA's NDFD S3 bucket into daily CSVs.
- 7-day forecasts of TMAX, TMIN, RHMAX, RHMIN, QPF, dew point, wind speed and sky cover across Mesonet stations.
- UTC → Central Time alignment so forecasts match daily agronomic models.
- Computes solar radiation from sky cover with FAO radiation equations using station elevation and latitude.
- Python
- GRIB2
- AWS S3
- Pandas
- FAO-56
- Mesonet
Forecast → Field pipeline
- 01NOAA NDFD
- 02GRIB2
- 03Python
- 04UTC → CST
- 05Daily aggregation
- 06Ag variables
- 07Irrigation models
TMAX · TMIN · RHMAX · RHMIN · QPF · DEW POINT · WIND · SKY COVER → SRAD
Current research directions
Work in progress — findings will be shared once complete.
OpenET Root-Zone Depletion Evaluation
Comparing OpenET-based root-zone depletion estimates against soil-moisture sensor observations in irrigated agricultural fields.
- Remote Sensing
- Evapotranspiration
Historical Irrigation Water Requirements
Lugert-Altus Irrigation District
Using long-term weather, soil, field and crop information with FAO-56 / PyFAO56 modeling to study historical irrigation water requirements.
- FAO-56 Modeling
- Water Resources
OASIS Kansas–Texas Expansion
Extending the OASIS scheduler beyond Oklahoma — automating weather, soil and crop inputs for irrigation decision support across a wider region.
- Decision Support
- Regional Scaling
05Publications & presentations
Papers, talks and posters.
3 papers and 8 conference presentations across IEEE, ASABE and Oklahoma water-research venues.
2026
- PaperConference paper
Data-Driven Modeling of Soil Moisture in Irrigated Fields
Kalayambakam Janardhan, A. Y., Datta, S., Taghvaeian, S., Crick, C., & Palash Kumar, K.
19th IEEE Dallas Circuits and Systems Conference (DCAS) · Dallas, TX · April 10–12, 2026
DOI 10.1109/DCAS69364.2026.11544580 - PaperConference paper
Canopy Temperature Sensor Data Optimization for Irrigation Scheduling
Meadi, S., Kalayambakam Janardhan, A. Y., Datta, S., & Das, H.
2026 IEEE Dallas Circuits and Systems Conference (DCAS) · Dallas, TX · April 10–12, 2026
DOI 10.1109/DCAS69364.2026.11545108 - Oral
Data-Driven Modeling of Multi-Depth Soil Moisture in Irrigated Fields
Kalayambakam Janardhan, A. Y., Datta, S., Taghvaeian, S., Crick, C., & Palash Kumar, K.
ASABE 119th Annual International Meeting · Indianapolis, IN · July 12–15, 2026
- Oral
Can We Accurately Estimate Canopy Temperature Without Adjusting Sensor Heights to Reduce Labor Requirements?
Meadi, S., Kalayambakam Janardhan, A. Y., & Datta, S.
ASABE 119th Annual International Meeting · Indianapolis, IN · July 12–15, 2026
- Oral
Paving the Path Forward to a National Scale Irrigation Scheduler by Leveraging PyFAO56
Datta, S., Meadi, S., & Kalayambakam Janardhan, A. Y.
ASABE 119th Annual International Meeting · Indianapolis, IN · July 12–15, 2026
- Oral
Canopy Temperature Sensor Data Optimization for Irrigation Scheduling
Meadi, S., Kalayambakam Janardhan, A. Y., Datta, S., & Das, H.
19th IEEE Dallas Circuits and Systems Conference · Dallas, TX · April 10–12, 2026
- Poster
Data-Driven Modeling of Soil Moisture in Irrigated Fields
Kalayambakam Janardhan, A. Y., Datta, S., Taghvaeian, S., Crick, C., & Palash Kumar, K.
19th IEEE Dallas Circuits and Systems Conference · Dallas, TX · April 10–12, 2026
2025
- Oral
Data-Driven Modeling of Soil Moisture for Irrigation Scheduling
Kalayambakam Janardhan, A. Y., Datta, S., Taghvaeian, S., Crick, C., & Palash Kumar, K.
2025 Oklahoma Governor's Water Conference & Research Symposium · Midwest City, OK · December 3–4, 2025
- Oral3-Minute Presentation
Oklahoma Irrigation Scheduler: Paving the Path for Sustainable Water Management
Kalayambakam Janardhan, A. Y., & Datta, S.
2025 ASABE Oklahoma Sectional Meeting · Stillwater, OK · March 7, 2025
2024
- Poster
Oklahoma Agricultural Scientific Irrigation Scheduler
Kalayambakam Janardhan, A. Y., & Datta, S.
2024 Oklahoma Governor's Water Conference & Research Symposium · Midwest City, OK · November 19–20, 2024
2022
- PaperJournal article
Extreme Learning Mechanism for Classification & Prediction of Soil Fertility Index
Ganesh, D., Abinava Yeshwanth, K. J., Meadi, S., Guru Sai Vignesh Reddy, M., Chirudeep, T., & Naveen Kumar Polisetty, S.
Journal of Pharmaceutical Negative Results · 13 (Special Issue 06), 37–43
DOI 10.47750/pnr.2022.13.S06.006
06Experience & education
Where I've worked and studied.
Experience
May 2024 — Present● Current
Graduate Research Assistant
Oklahoma State University · Biosystems & Agricultural Engineering · RISE Lab — Research in Irrigation Scheduling and Engineering (Dr. Sumon Datta)
Stillwater, OK
Building the software, data pipelines and models behind irrigation decision support — and helping instrument the fields that feed them.
- Develop OASIS (Operational Agriculture and Soil Information System), a full-stack irrigation decision support platform (React, FastAPI, Plotly).
- Built and compared five ML models for multi-depth soil moisture prediction.
- Engineered NOAA NDFD GRIB2 pipelines producing daily 7-day forecasts across Mesonet stations.
- Field instrumentation and bi-weekly data collection across Oklahoma research sites.
- Irrigation
- ML
- Full-stack
- Field work
Nov 2022 — Jul 2023
Software Developer Intern
Federal Soft Systems
Tirupati, AP, India
Media sentiment analysis for marketing insights.
- Designed a Python, R and SQL pipeline to collect, clean and analyze Twitter and Facebook data.
- Processed over 100,000 data points with Pandas and scikit-learn for sentiment classification.
- Built Matplotlib visualizations to surface sentiment trends and popular topics.
- Data pipelines
- NLP
- Python
Jan 2022 — Nov 2022
Student Assistant
Sree Vidyanikethan Engineering College · Computer Science & Engineering
Tirupati, AP, India
Web-based natural disaster prediction tool.
- Developed the tool with HTML5, JavaScript and Flask APIs, backed by a Random Forest model.
- Responsive interface with dynamic data visualization.
- Web
- ML
- Flask
Education
- Jan 2026
Agricultural engineering
PhD in Biosystems & Agricultural Engineering
Oklahoma State University
Stillwater, OK
– May 2029 (expected) - Aug 2023
Computer science
MS in Computer Science
Oklahoma State University
Stillwater, OK
– Dec 2025 - Aug 2019
Computer science
B.Tech in Computer Science & Engineering
Sree Vidyanikethan Engineering College
Tirupati, AP, India
– May 2023
Two computer science degrees, then a PhD that applies them to agricultural water.
07Field research
Grounded in real field data.
Good models need good measurements. I help install and maintain sensor networks on research stations and growers' fields across Oklahoma.
- producer fields instrumented
- 15
- producer fields instrumented
- regions of Oklahoma
- 3
- regions of Oklahoma
- logger data collection
- Bi-weekly
- logger data collection
Oklahoma Panhandle
Producer-field monitoring network
Assisted in installing soil moisture sensors and rain gauges in 15 producers' fields in 2025 and 2026 as part of a USDA-funded project.
- Soil moisture sensors
- Rain gauges
Oklahoma Panhandle Research & Extension Center
Eddy covariance flux tower
Assisted the LI-COR team with the installation of an eddy covariance flux tower.
- Eddy covariance
- LI-COR
Southwest Oklahoma Research & Extension Station
Canopy temperature & soil moisture
Assisted with installation of IRT canopy temperature sensors and Acclima soil moisture sensors.
- IRT sensors
- Acclima
Central Oklahoma
Profile soil moisture
Soil moisture sensor and rain gauge installation.
- Sentek Drill & Drop
- Rain gauge
Across Oklahoma
Data logger collection
Bi-weekly data collection from ZL6 data loggers across multiple regions of Oklahoma.
- ZL6 loggers
08Earlier projects
Foundations in computer science.
Software and machine-learning projects from before agricultural research — the foundation for today's work.
Undergraduate · published 2022
Soil Fertility Prediction
Extreme Learning Machine approach for classification and prediction of a soil fertility index — an early link between computing and agriculture.
ELM · Classification
DOIStudent Assistant · 2022
Natural Disaster Prediction
Web application predicting natural disasters with a Random Forest model served through Flask APIs.
Random Forest · Flask · JavaScript
Internship · 2022–23
Media Sentiment Analysis
Data pipeline and sentiment classification over social-media data to inform marketing strategy.
Python · R · SQL · scikit-learn
Undergraduate project
Vali-Doc
Document validation using digital signatures and SHA-512 hashing for authentication and integrity.
Digital signatures · SHA-512
09Beyond research
Leadership & community.
At Oklahoma State
Current
CurrentVice Treasurer
BAE Graduate Student Association
Oklahoma State University
Aug 2025 – May 2026
Sports Secretary
Indian Student Association (ISA)
Oklahoma State University
Undergraduate
CSE Executive
ISTE Student Chapter
Jun 2021 – Jan 2023
Organized and conducted technical and student events. Helped run a techno-cultural fest in a leadership capacity.
Campus Ambassador
IUCEE
Jun 2021 – Jan 2023
Promoted campus activities. Contributed to club development and student engagement.
Social Media Handler
CETA
Jun 2021 – Jan 2023
Supported social-media communication. Conducted outreach activities within the college.
Professional activities
ASABE
Member, American Society of Agricultural & Biological Engineers
IEEE
Member, Institute of Electrical and Electronics Engineers
FE Exam
Passed the Fundamentals of Engineering exam on the first attempt
Master Irrigator Program
Participated and gathered grower feedback to improve the usability of OASIS
PhD Interview Panel
Panel member for BAE PhD applicant interviews (RISE Lab); set tasks to evaluate conceptual and written skills
Off the clock
CricketBadmintonChessCarromReading
10Contact
Let's talk about water, data and agriculture.
Open to research collaborations, conversations from conferences, and opportunities at the intersection of AI and agricultural engineering.
Download CV
Stillwater, Oklahoma