My name is Olya (Ola) Skulovich, I am
near completion of my Ph.D. in Earth
and Environmental Engineering at
Columbia University. During my Ph.D.
program, I worked with remote sensing
data and utilized machine learning to
create long-term consistent soil
moisture and vegetation optical depth
datasets. In particular, my work
included analyzing, regridding, and
deseasonalizing remote sensing data
from SMAP, SMOS, AMSR-E, and AMSR-2
satellite missions to prepare the data
for machine learning. On the
methodological side, I developed,
tested, and fine-tuned Deep and
Convolutional neural networks and
built a unique transfer learning
training scheme to merge the patched
remote sensing data into a consistent
dataset. The soil moisture dataset and
the corresponding paper (Scientific
Data – Nature family journal,
https://doi.org/10.1038/s41597-023-02053-x
) have been published. The soil
moisture dataset is the only
consistent quality dataset available
globally, covering 18 years,
explicitly targeting soil moisture
extremes and anomalies. The vegetation
optical depth dataset is the only
L-band dataset that spawns back to
2002. After developing the datasets,
my research was focused on analyzing
trends and variability of soil
moisture, including spatiotemporal
statistical analysis and identifying
regions of different dynamics. A
part of my research was dedicated to
modifying and analyzing the
process-based carbon cycle model
(CARDAMOM). It is a model that
simulates carbon fluxes and pools by
assimilating data using the
Metropolis-Hastings Markov chain Monte
Carlo method. My part of the project
included developing two new model
modules for assimilating solar-induced
fluorescence and vegetation optical
depth data, including developing model
formulation, incorporating the modules
in the main model (C and Python),
adjusting model uncertainties,
likelihood functions, and ecological
dynamical constraints, as well as
analyzing the updated model’s
performance, information content
effect, effect on constraining
respiration flux and carbon
pools. In addition to that, I
participated as a collaborator in
several research projects
investigating the effects of soil
moisture and land-atmosphere feedback
on European and Siberian droughts and
the spatiotemporal relationship
between soil moisture dynamics and
vegetation productivity. I
presented the results of my research
at the American Geophysical Union
(AGU) Fall Meetings in 2020, 2022, and
2023, the USMILE Kickoff Meeting 2020,
USMILE Meeting 2022, and LEMONTREE
Science Meeting: Soil Moisture Stress
2023. I take pride in building a
compelling story from scientific
findings and enjoy communicating my
research to various stakeholders,
creating captivating presentations,
and engaging public speaking.