GIScience

---

Areas of expertise:

• Spatial Data Analytics
• GIS Programming (Python, R, React.js, C++)
• Spatial-temporal modeling and simulation

---

Ecosystem Structure Diversity Across the Continental US Using NEON Vegetation Structure Measurements

Characterizing the allometric relationships by plant functional type for parameterizing structural relationships in ecosystem dyanmics models.

---

Close

Remote Sensing

---

Areas of expertise:

• Small Footprint and Full Waveform LiDAR Remote Sensing
• Multi- and Hyper- Spectral Remote Sensing
• Time Series and Machine Learning Analyses
• Image Classification and Semantic Segmentation

---

Mapping Aboveground Biomass in the Continental U.S. using GEDI Waveforms

In this research, we provide estimates of above ground biomass across the continental U.S. using a full waveform based model. The model utilizes the full vertical information derived from waveforms, integrating from the bottom to the top of the canopy layer the product of the waveform and a power function of height. The model is parameterized using in-situ vegetation structure data provided by the NSF NEON program and compared with FIA AGB estimates. Theoretical basis paper.

---

Modeling Gross Primary Production from OCO-3 Solar Induced Flourescence (SIF) and CO2 Measurements

Utilized OCO-3 measurements to estimate diurnal and seasonal variability of CO2 concentration by land cover type and estimate GPP of forests surrounding three major metropolitan areas.

---

Quantifying Above-Ground Biomass in Tropical Forests Directly from Lidar Waveforms

Estimating above ground biomass using full waveform lidar measurements across tropical forests in Central Africa. Model outperforms leading small footprint and waveform based models. Theoretical basis paper.

---

U-NET CNN Semantic Segmentation of Land Cover Types using Airborne Imagery in Dubai

Utilizing airborne imagery from Dubai hosted on Kaggle, a large depth U-NET CNN model was trained for semantic segmentation on a simple 5 class land cover dataset.

---

Mapping Habitat Suitability for Pinus Rigada in Northeastern United States

Machine learning model to classify habitat suitability for P. rigada using WorldClim and SOILGRIDS data.

---

Close

Software Tools

---

Close