Kelsey Ennis

PhD Candidate | Wegener Center für Klima & Globalen Wandel

Universität Graz | Department of Geography and Regional Science

Email:

kelsey.ennis@uni-graz.at

About me/ My Story

Welcome to my site! I am currently a PhD Candidate at Universität Graz in Austria. In June 2025, I completed my Master's degree in Atmospheric Science at Colorado State University where I was co-advised by Dr. Libby Barnes and Dr. Eric Maloney.

In 2019, I moved to Daytona Beach, Florida for undergraduate. I attended Embry Riddle Aeronautical University where I pursued a Bachelor's of Science in Meteorology and minored in applied mathematics. During my time there I was involved in undergraduate research, worked as a tutor and as teaching assistant for the meteorology department, President of Tri Sigma Sorority, Vice President in Senior Class Council, and I worked for the university's Campus Safety Administration. I graduated in May 2023 and was awarded: Meteorology Student of the Year by the Meteorology dept., President of the Year by Fraternity & Sorority life, and ERAU's most prestigious undergraduate award, the John C. Adams, Jr. Community Service Award. The community service award is presented to one graduating senior per year in recognition of all the time and efforts the student has given to the university and the local community during their time at ERAU.

In July 2023, I moved to Fort Collins, Colorado to pursue my Master's degree working in the Barnes group. My research focused on how skillful two artificial intelligence weather prediction models forecasted heat waves in the continental United States. While at Colorado State I was able to be involved with many organizations. I served as secretary & treasurer on the CSU FORTCAST executive board. Within this organization, we placed a strong emphasis on community outreach and professional development. In addition to this, I served as a student mentor and AMS BRAID volunteer during my time here. I defended my thesis on June 12th, 2025.

In August 2025, I accepted an amazing opportunity abroad in Graz Austria to complete my PhD at Karl-Franzens-Universität Graz (University of Graz). I currently work at the Wegener Center for Climate and Global Change in the regional climate working group. My research develops physics-informed, probabilistic Artificial intelligence for high-resolution climate modelling in mountainous regions in Austria. Coarse global datasets miss the fine-scale detail that matters most in complex terrain like the Austrian Alps, where climate impacts unfold at the scale of individual valleys. I build & design Deep Learning emulators that downscale coarse data to the kilometre scale while respecting the underlying physics, and that quantify their own uncertainty. The goal of this project is reliable, high-resolution climate information that can strengthen impact assessments and decision-making, a step toward a regional digital twin of the Austrian Alpines.

I fell in love with the chaotic nature of the atmosphere when I was 3 years old. On what was supposed to be a relaxing vacation, my family was instead greeted with the alarming news that a hurricane was meandering in the Atlantic. Hurricane Jeanne would make landfall on the East coast of Florida on September 26th, 2004. With little time to prep and two young kids, my parents got the supplies they could and we hunkered down and rode out the storm in Cocoa Beach, located on the 'Space Coast'. This is my earliest memory from my childhood. When I think back to it I remember the look of fear on both my parents' faces when it was announced that a storm was coming. I can still hear the screams of the wind and the sound of the rain striking the windows with incredible force. After the storm passed, the aftermath was unbelievable. And just like that, on September 28th, 2004 a meteorologist was born.

After many years of hard work, my passion has turned into my career.

EXPERIENCE

October 2025-Present

PhD Candidate

Universität Graz

Working in collaboration with both the Wegener Center and the Know Center in Graz, Austria.
Our project focuses on improving climate change impact assessment in densely populated mountainous regions within the Austrian Alps. Novel downscaling models with reliable uncertainty quantification will be developed using physics-informed and probabilistic machine learning (ML) methods.
This effort aims to generate climate data for Austria’s alpine regions with unprecedented spatial/ temporal resolution and quality, resulting in a significant step towards a
digital twin of the region.

June 2023-June 2025

Colorado State University

Graduate Research Assistant

Completed a Masters in Atmospheric Science in the Barnes research group.
My research focuses on understanding temperature prediction skill during extremes (i.e., heat waves) for NCEP's dynamical S2S model the NOAA Unified Forecast System (UFS GEFS) and various AI weather prediction models (i.e., Pangu) over North America.

April 2022-May 2023

Undergraduate Research Assistant

Embry Riddle Aeronautical University

Worked on a funded NOAA Climate Program Office project on Southeast U.S. heat stress trends.
Created heat stress climatologies and trend analyses using R and Python coding programs.
Investigated the physical mechanisms responsible for Southeast U.S. heat stress trends and their impacts on human health.
Presented my results at the 2023 American Meteorological Annual Meeting.
Prepared manuscript for Journal of Applied Meteorology and Climatology (Summer 2023).

June 2022-August 2022

Internship

Munich Reinsurance

Statistical modeling of stochastic events to estimate losses due to catastrophes (perils).
Responsibilities included using computer applications such as R Studio and SQL Server Management to produce accurate loss budgets for MRSI specialty insurance branches.
Mapped & tracked real time perils (i.e., SCS, wildfire, etc.) as they occurred using ArcGIS to estimate losses & amount of claims.

EDUCATION

2025-2028

Doctorate Degree

Universität Graz

Doctorate degree in Earth System & Natural Sciences

Specialization: Climate change and Environmental technology

Skills

Artificial intelligence / Machine learning techniques
Data science
Weather & climate extremes
Climate change impacts risk analysis

2023-2025

COLORADO STATE UNIVERSITY

Master's Degree

Masters of Science in Atmospheric Science

Graduate Research Assistant in the Barnes Research Group

Colorado State FORTCAST Board Secretary

Start: June 2023
Finish: June 2025

2019-2023

Bachelor's Degree

EMBRY RIDDLE AERONAUTICAL UNIVERSITY

Bachelor of Science in Meteorology Minor: Applied Mathematics

GPA: 3.71; Major GPA: 3.82

Awarded the Presidential Scholarship and Women of Excellence Award
Graduated Magna Cum laude
Awarded sorority and fraternity life President of the year for outstanding leadership and initiative.
Awarded Outstanding Meteorology Student of the Year awarded to one graduating meteorology senior a year.
Awarded the prestigious John C. Adams Community Service at my Spring 2023 graduation, it is awarded to one graduating senior a year at Spring graduation.

2015-2019

High School Diploma

HIGHTSTOWN HIGH SCHOOL

Yearbook Committee Lead
English Tutor
Peer Advisor/Mentor
Drama Club member
Lifeguard Instructor assistant

PUBLICATIONS

Cruel Summer (and Autumn): Humid Heat Trends, Extremes, and Mechanisms in the Southeast United States

Publication date: 18 March 2025

Doi: https://doi.org/10.1002/joc.8837

Authors: Kelsey. E. Ennis, Shawn M. Milrad, Desiree A. Knight, Kathy D. Dello, Lily Raye, Kyle R. Wodzicki

This study employs an established wet bulb globe temperature (WBGT) estimation formula and high-resolution reanalysis data to develop a custom gridded WBGT dataset for the Southeast. Summer and autumn WBGT (humid heat) trends and extremes are elucidated for 1950–2023. Overall, trends in average and extreme (90th percentile) WBGT are larger and more widely statistically significant at night and during summer. Autumn WBGT increases are largest and most significant in the Florida Peninsula and near the Gulf Coast, where the heat content of the adjacent water peaks in September. Furthermore, there are large increases in the frequency and duration of summer night-time extremes; regions near the Gulf and in Florida are experiencing nearly three additional extreme summer nights per decade, and extreme events are approximately one night longer per decade. A quantification of WBGT components shows that moisture (dewpoint) increases exceed temperature increases and are the largest contributors to WBGT across most of the Southeast, except in parts of the Appalachian and Piedmont regions where temperature changes dominate.

Man, it's a hot one: Trends and Extremes in Florida Autumn Heat Stress

Publication date: 04 March 2024

Doi: https://doi.org/10.1002/joc.8415

Authors: Kelsey. E. Ennis & Shawn M. Milrad

This study uses WBGT to examine heat stress trends and extremes during climatological autumn from the 1950s to 2022 at eight Florida cities. Results show that average autumn heat stress has increased significantly throughout much of Florida. Daytime heat stress increases were larger than their night-time counterparts in northern Florida, while the opposite was true in central and southern Florida. Extreme autumn heat stress events also increased in intensity and frequency. Results show that heat stress in Florida is no longer just a summer hazard and poses a particular threat to coastal areas in early autumn when ocean heat content is still near its climatological peak.

Heat Stress Metrics, Trends, and Extremes in the Southeastern United States

Publication date: 29 July 2024

Doi: https://doi.org/10.1175/JAMC-D-24-0009.1

Authors: Kyle R. Wodzicki, Kelsey E. Ennis, Desiree A. Knight, Shawn M. Milrad, Kathie D.
Dello, Corey Davis,a Sean Heuser, Blaine Thomas, and Lily Raye

This study evaluates several WBGT estimation formulas for the southeastern United States using North Carolina ECONet and U.S. Military measurement campaign data as verification. The estimation algorithm with the smallest mean absolute error was subsequently chosen to evaluate summer WBGT trends and extremes at 39 ASOS stations with long continuous (1950–2023) data records. Trend results showed that summer WBGT has increased throughout much of the southeastern U.S., with larger increases at night than during the day. Although there were some surprisingly large WBGT trends at higher elevation locations far from coastlines. Increases in the intensity and frequency of extreme WBGT's were particularly stark in large coastal urban centers. Some locations have experienced more than two additional extreme heat stress days and nights per decade since 1950, with an exponential escalation in the number of extreme summer nights during the most recent decade.

Assessing Summer Humid Heat in Europe: Trends, Extremes, and Drivers

Publication date: 22 May 2025

Doi: https://doi.org/10.1002/joc.8913

Authors: Shawn M. Milrad and Kelsey E. Ennis

Using wet bulb globe temperature (WBGT), this study uses high-resolution reanalysis data to investigate trends in the intensity, frequency, and duration of humid heat throughout Europe during climatological summers, 1950–2023. Results show that average and extreme (90th percentile) WBGT intensity exhibit statistically significant increases throughout Europe. Extreme humid heat intensity is increasing faster than average humid heat intensity across much of northern Europe and the Mediterranean coast, especially at night. Daytime humid heat trends are largest in western and northern Europe, while nighttime trends exceed daytime trends in eastern and southern Europe. Furthermore, there are statistically significant and large frequency and duration trends in extreme humid heat, especially at night near the Mediterranean, where extreme events are occurring approximately three extra nights per decade and extreme event duration is more than 1 day longer each decade. Unlike the Southeast United States, where dewpoint (humidity) increases are most responsible for WBGT increases, temperature increases are the dominant driver of positive WBGT trends across most of Europe. Results demonstrate that while humid heat is worsening in many locations across the world, it can amplify for different reasons in different regions.

Turning up the Heat: Assessing 2-m Temperature Forecast Errors in Artificial Intelligence Weather Prediction Models During Heat Waves

Publication date: 30 September 2026

Doi:https://doi.org/10.1002/met.70249

Authors: Kelsey E. Ennis, Marybeth C. Arcodia, Martin A. Fernandez, Eric D. Maloney, Elizabeth A. Barnes

Extreme heat is the deadliest weather-related hazard in the United States. Furthermore, heat extremes are increasing in intensity, frequency, and duration, making skillful forecasts vital to protecting life and property. Traditional numerical weather prediction (NWP) models struggle with extreme heat forecasts on longer (i.e., medium-range- and subseasonal) timescales. Meanwhile, artificial intelligence-based weather prediction (AIWP) models are progressing rapidly. However, it is largely unknown how well AIWP models predict extreme heat, especially for longer-range forecasts. This study investigates 2-m temperature forecasts for 60 heat waves across four contiguous United States regions (Northwest, Midwest, Northeast, Southeast) and the four boreal seasons at medium-range to subseasonal (i.e., 10–20-day) lead times, using two deterministic AIWP models (GraphCast and Pangu-Weather) and one probabilistic AIWP model (GenCast). All three models exhibit persistent mean cold biases throughout the forecast evolution, especially near the start of the heat wave. However, the GenCast ensemble mean consistently exhibits smaller biases and errors than the two deterministic models. In all three AIWP models, the smallest regional average errors during the heat wave period are in the Northwest, with the largest in the Midwest. In all boreal seasons, mean cold biases occur before the heat waves in the three AIWP models except Pangu-Weather in winter, which exhibits a mean warm bias before heat wave onset. A brief comparison of 2-m temperature forecast errors is then illustrated among the AIWP models and a traditional NWP model (NOAA United Forecast System Global Ensemble Forecast System (UFS GEFS)), for which GenCast particularly performs strongly. Overall, results show that the probabilistic AIWP model GenCast offers the most promising accuracy toward medium-range and subseasonal forecasts of extreme heat, albeit with a consistent slight mean cold bias.

SKILLS

High Performance Computing (HPC)

Data Analysis

Climate & Weather Extremes

Python Programming Language

Risk Communication Methods/Techniques

AI & ML Computing / Design