Jaden Dicopoulos, Data Scientist,
Engineer,
Cloud Practitioner
Collection of figures from my papers, presentations, and related works
A hexbin plot used to show how the wind shear coefficient (alpha) correlates to the air sea temperature difference. This was part of my conclusion in my paper (Weather Research and Forecasting Model Validation with NREL Specifications over the New York/New Jersey Bight for Offshore Wind Development) which talked about the practice of validating models such as WRF and using them to observe and evaluate the offshore wind space.
A hexbin plot to show where most of the model data matches with the observation data. Also included are simple statistics to show how strong the correlation is. I created this figure in tandem with other figures for our inhouse reporting but also in meetings with our funders to showcase our model's powerful correlation in comparison to other models at this observation al point. This figure was also generated automatically every month, season and year automatically.
A figure demonstrating my center of mass locator algorithm for Sargassum Seaweed patches off the coast of Puerto Rico.
On the left an image from a satellite of the Alternate Floating Algae Index, which is then scanned by my code for patches of Sargassum Seaweed with a set of parameters removing false positives.
Histogram with difference plot to to visualize where discrepancies in wind speeds are found. Weibull curves and values are included to help tell the story of the shapes of the distributions.
A wind profile plot to show how winds vary at different altitudes. The colored lines are made from using a variable derived from the observational data and fed into a dynamic best fit line.
A hexbin plot to show where wind speeds match with wind shear alphas.
Simple map making and marking in python for publication figures and easy reproducibility.
A Taylor diagram for a better way to visualize what statistical data means for large data sets in the grand scheme of things. Often when looking at data we narrow in on smaller differences which may not be significant on larger scales.
Power averages with 25th and 75th quartiles during months where sea breezes were not present.
Figures like this one were made weekly for years to give us insight on our model's performance against observational points. Statistics were also run along side this for better numerical analysis. These were put into reports which were shared with management and interested stakeholders.
Taking the wind shear formula and plotting by its height and wind components gives us this figure where we are looking to create a fitted profile with a best wind shear alpha.
A look into how power averages vary between sea breeze days and non seabreeze days. The quartiles help show how the sea breeze days are far less variable.
Simple time series showing a months worth of data. The models are aggregated together to form a highest highs and lowest lows area on the graph so the figure is not cluttered by a multitude of lines.
This is a still from an animation of the Sargassum particles moving after being through a drifter simulation fed by coastal wave measurements made by HF radar. When they reach land or dissolve they turned blue, this is how we would know where they would land.
This analysis lead us to discover that a static variable assigned to the log wind profile was actually due to two peaks around the number rather than being an actual representation. We saw this not only in our model but in the observations as well.
Feel free to reach me at jadendicopoulos@gmail.com or 908-216-6136
Other helpful links;
Previous job page https://rucool.marine.rutgers.edu/people/jaden-dicopoulos/
Github for previous job https://github.com/JadenD