SOIL PHYSICS · ENVIRONMENTAL SYSTEMS · PHD

Samuel Araya

Soil, environmental & data science

My background is in soil physics and environmental science. I work with agricultural and environmental data using field measurements, geospatial analysis, statistical models, and machine learning.

View projects View résumé

About me
My background and education.

A fixed-wing drone flying over the landscape during soil moisture research. — Drone imagery for soil moisture research

WORK & EDUCATIONVylor / CortevaUSDAStanfordUniversity of California, Merced

01 / EXPERIENCE

Professional
experience

View résumé ↗

2023 to present

Data Scientist

Vylor / Corteva Agriscience

Crop-rotation prediction, crop-model validation, and Python/SQL data pipelines. Collaboration with engineering, product, agronomy, and modeling teams.

2021 to 2023

Soil Scientist

USDA Natural Resources Conservation Service

Soil and environmental analysis, geospatial databases, and terrain and satellite workflows supporting soil mapping and land-use planning.

2019 to 2021

Postdoctoral Researcher

Stanford University

Remote sensing, environmental modeling, and spatial and temporal analysis connecting surface observations with subsurface biogeochemistry.

PhD & MS, Environmental SystemsUniversity of California, MercedSoil physics and soil biogeochemistry

BSc, Land Resources and EnvironmentUniversity of Asmara

02 / RESEARCH

Selected
research

Selected projects from my doctoral research in soil physics and environmental science.

Testing-set measured and predicted volumetric soil moisture, with a one-to-one line and model evaluation metrics. — Measured and predicted soil moisture on the testing sets. Araya et al. (2021), Fig. 7. Click the figure to view full size.

REMOTE SENSING + MACHINE LEARNING

Soil moisture
from drone imagery

Estimating surface soil moisture using drone imagery, terrain attributes, and weather variables.

Doctoral research · Published 2021University of California, MercedRead project ↗

Model-predicted saturated hydraulic conductivity across organic-carbon values, grouped by soil texture, with logarithmic axes. — Project plot: model responses to organic carbon, by soil texture. Related paper, Araya & Ghezzehei (2019). Click the figure to view full size.

PREDICTIVE MODELING + OPEN SOURCE

Predicting saturated
hydraulic conductivity

Machine-learning models for predicting saturated hydraulic conductivity from soil properties. The code and a graphical interface are publicly available.

Doctoral research · Published 2019University of California, MercedRead project ↗

DOCTORAL RESEARCH · PUBLISHED 2022Conservation agriculture and soil water

03 / SELECTED PUBLICATIONS