Master's Project Directory

Forests cover approximately 30% of Earth's land surface, absorb more carbon than all other terrestrial ecosystems, and provide trillions of dollars’ worth of ecosystem services (Food and Agriculture Organization of the United Nations, 2005). However, climate change-induced droughts pose a significant threat to these vital ecosystems. As climate change intensifies, it is critical for our planning and management that we understand how and where trees will be the most threatened. Previous research has examined the effects of these droughts on forests at a global scale, but these large-scale analyses are not particularly helpful for land managers that often focus on specific regions and only a limited number of species. Our project addresses this gap by assessing species-specific sensitivity to increasingly severe and frequent droughts, considering the variations within their ranges.

Consumer demand for sustainable products is increasingly driving companies to consider the life-cycle carbon emissions of products being sold. Darn Tough, a Vermont-based sock manufacturer, aims to continue their reputation of environmental stewardship through aligning its operations with the 2030 carbon mitigation target set by its largest retailer, Recreational Equipment, Inc. (REI). This project streamlines the calculation process with an improved computational method and builds an interactive visualization dashboard that allows the company to calculate yearly emissions and analyze the potential GHG impacts of changes in different fiber types for sock manufacturing.

Climate models are computer simulations that attempt to replicate the complex interactions between Earth’s systems. Improving the accuracy of climate models relies on evaluating uncertainty and minimizing error. The Climate and Global Dynamics Lab at the National Center for Atmospheric Research (NCAR) has recently carried out a Parameter Perturbation Experiment (PPE) to understand how the uncertainty of parameter values affected the output of their model, the Community Land Model (CLM); which simulates terrestrial processes. While the necessary data for the PPE has been collected, the data is stored in a collection of files that are difficult to interpret in their current form. The current website hosts visualizations for a portion of the PPE data, but contains no visualizations for data that more closely simulates Earth system interactions. These issues can be mitigated by developing an emulator with the internal complexity to isolate a one-to-one relationship between a parameter and climate variable, then display the predicted relationship. A publicly available emulator with these capabilities will allow scientists to easily interpret complex climate model outputs and offer insights on parameter-variable relationships that are not being predicted accurately by the model; which can lead to increased accuracy and precision of climate models. 

The Santa Barbara Botanic Garden is restoring monarch butterfly habitats in the Los Padres National Forest. Using habitat models and accessibility tools, the team identified key areas for milkweed planting to support monarch populations and streamline future surveys.

Community engagement in planning is essential for effective and just climate adaptation. However, historically underserved communities are often difficult to reach through traditional means of soliciting public input. The Climate Adaptation Solutions Accelerator (CASA) through School-Community Hubs project identifies public schools as promising sites for building both community engagement and community capacity for climate adaptation.

The Channel Islands Marine Sanctuary (CINMS) comprises 1,470 square miles surrounding the Northern Channel Islands: Anacapa, Santa Cruz, Santa Rosa, San Miguel, and Santa Barbara, protecting various species and habitats. However, these sensitive habitats are highly susceptible to climate-driven ‘shock’ events which are associated with extreme values of temperature, pH, or ocean nutrient levels. To better understand the drivers and statistics of climate-driven ecosystem shocks, a ‘large ensemble’ of simulations run with multiple climate models will be used. The objective of this project is to develop an educational toolkit and Python-based web application to visualize ecologically significant climate variables near the CINMS.

Dichlorodiphenyltrichloroethane (DDT) is an insecticide that is resistant to degradation and can cause increased risks of cancer, premature births, developmental abnormalities, and neurological diseases in humans and animals. Although banned in 1972, DDT was dumped into the ocean off the coast of southern California for decades. Recent documentation of the extent of this contamination has captured the attention of the public and raised concerns regarding the consumption of contaminated seafood. The State of California currently issues consumption advisories for coastal communities, but these advisories do not address DDT. To fill this gap and improve advisory accessibility, SaferSeafood has partnered with Scripps Institution of Oceanography and the California Cooperative Oceanic Fisheries Investigations, who collected data on the incidence of DDT in sediments and fish in nearshore southern California waters. Using this data, we updated and validated their spatiotemporal statistical model to predict DDT concentrations in fish based on species and location. We then integrated this model into an interactive web application that allows anglers to receive predicted DDT concentrations in the fish they have just caught, along with serving size recommendations and other relevant advisories. This project will allow individuals to make well-informed decisions about their seafood choices in the face of environmental challenges and health risks associated with DDT contamination.

This project examines how supply-side oil and gas regulations impact greenhouse gas emissions, employment, and the health of communities in California living near oil wells. The project is especially focused on the implications of Senate Bill 1137 (SB1137), which would prohibit the construction of new oil and gas wells within 3,200 feet of schools, hospitals, and other sensitive receptors.

This project estimates how changes to the rules governing private LUPs could affect the extent of forest cover and agricultural land in the Paraguayan Chaco through a four-step approach: (1) Assess property owners' compliance that deforestation only occurs in authorized areas and determine deforestation rates by using geospatial overlays, (2) Estimate protected forest area under different potential laws in the undeveloped Paraguayan Chaco region by creating mock properties with simulated land use plans, (3) Integrate the simulated LUP configurations into a machine learning model to examine how potential changes to land use laws might influence future deforestation patterns, and (4) To inform forest conservation regulations by providing INFONA and other stakeholders with an interactive platform for exploring the results.

We produced spatial layers of four different wildfire metrics, and conducted exploratory statistical analyses to assess how these wildfire metrics may be affected by GDEs in California. These layers and analyses are incorporated into an interactive dashboard that allows users to view general spatial relationships between groundwater variables, such as GDEs, and various wildfire metrics.

This project provides the first facility-level analysis of indirect carbon emissions from EAF steel plants in the US. As the US steel industry seeks to decarbonize and promote green steel, understanding the Scope 2 emissions of EAF steel production is essential. The calculated emissions and emissions intensity for each steel plant are available in a public dataset as well as an interactive Tableau dashboard.

Through this project, the capstone team created a reproducible workflow to streamline data entry for the MPS Tracker, as well as a data visualization tool to summarize and highlight trends. These deliverables allow WildAid Marine to increase their efficiency in monitoring and improving MPA enforcement, and successfully share progress with donors and stakeholders.

This project contributes to the larger Building Resilience to Wildfires initiative by presenting a process for creating these potential climate scenarios by stitching together pieces of existing climate model projections. To build each piece of a climate scenario, or each ‘segment’, climate models are searched for pieces that match the specified climate criteria of interest for that very segment. Once all desired segments are built, segments are then stitched together to make one continuous time-series: a ‘climate scenario’.

Adding solar PV to existing wind energy sites can boost profits and minimize the need for new transmission infrastructure. This study used a nonlinear optimization model to size solar PV additions at over 1,300 U.S. wind projects, ranking sites by environmental impact and cost efficiency.

The ecology of giant kelp is complex and the formation of kelp forests is impacted by many factors, including ocean nutrient concentrations, water depth, sea surface temperature, and seafloor habitat. The Santa Barbara Channel has been home to many long-term research projects that collect data on these factors. Unfortunately, this information is in many formats and data structures making it challenging for researchers to easily incorporate it into their work. This project creates a shortcut for acquiring and using these data by synthesizing a data set of nutrient concentrations, water depth, sea surface temperature, seafloor habitat, kelp area, and kelp biomass.

In this project, we conduct a case study of Zambia to demonstrate a recently developed machine learning pipeline. We use the “Multi-task Observations using Satellite Imagery & Kitchen Sinks” (MOSAIKS) machine learning approach. In this approach, we use processed numerical data from satellite images and agricultural survey data to develop machine learning models that predict various agricultural variables over time.