Colorimetric assays are an important tool in point-of-care testing that offers several advantages to traditional testing methods such as rapid response times and inexpensive costs. A factor that currently limits the portability and accessibility of these assays are methods that can objectively determine the results of these assays. Current solutions consist of creating a test reader that standardizes the conditions the strip is under before being measured in some way. However, this increases the cost and decreases the portability of these assays. The focus of this study is to create a machine learning algorithm that can objectively determine results of colorimetric assays under varying conditions. To ensure the flexibility of a model to several types of colorimetric assays, three models were trained on the same convolutional neural network with different datasets. The images these models are trained on consist of positive and negative images of ETG, fentanyl, and HPV Antibodies test strips taken under different lighting and background conditions. A fourth model is trained on an image set composed of all three strip types. The results from these models show it is able to predict positive and negative results to a high level of accuracy.

Hydrology and biogeochemistry are coupled in all systems. However, human decision-making regarding hydrology and biogeochemistry are often separate, even though decisions about hydrologic systems may have substantial impacts on biogeochemical patterns and processes. The overarching question of this dissertation was: How does hydrologic engineering interact with the effects of nutrient loading and climate to drive watershed nutrient yields? I conducted research in two study systems with contrasting spatial and temporal scales. Using a combination of data-mining and modeling approaches, I reconstructed nitrogen and phosphorus budgets for the northeastern US over the 20th century, including anthropogenic nutrient inputs and riverine fluxes, for ~200 watersheds at 5 year time intervals. Infrastructure systems, such as sewers, wastewater treatment plants, and reservoirs, strongly affected the spatial and temporal patterns of nutrient fluxes from northeastern watersheds. At a smaller scale, I investigated the effects of urban stormwater drainage infrastructure on water and nutrient delivery from urban watersheds in Phoenix, AZ. Using a combination of field monitoring and statistical modeling, I tested hypotheses about the importance of hydrologic and biogeochemical control of nutrient delivery. My research suggests that hydrology is the major driver of differences in nutrient fluxes from urban watersheds at the event scale, and that consideration of altered hydrologic networks is critical for understanding anthropogenic impacts on biogeochemical cycles. Overall, I found that human activities affect nutrient transport via multiple pathways. Anthropogenic nutrient additions increase the supply of nutrients available for transport, whereas hydrologic infrastructure controls the delivery of nutrients from watersheds. Incorporating the effects of hydrologic infrastructure is critical for understanding anthropogenic effects on biogeochemical fluxes across spatial and temporal scales.