Riverbed sediments are hotspots for biogeochemical transformations because they provide the physical, chemical, and biological conditions that reactions depend on. Recent research supports the hypothesis that reaction extent is primarily controlled by residence time within just the top few centimeters of sediment, the benthic biolayer, where riverine microbiota are concentrated and where mixing between channel water and deeper hyporheic zone solutions occurs. Our work focuses on relating residence time within the benthic biolayer, measured over a short distance, to total transformation observed at the full river scale.

Environmental nucleic acids are powerful monitoring tools, as they can be used to assess the presence and absence of endangered or invasive species, ecosystem functioning, and potential human health risks. Recent research from our group has demonstrated that antibiotic resistance genes added to genetically modified crops can mobilize with rainwater and runoff from soil into rivers, and that river sediment has the capacity to retain this material for extended periods of time. Our work focuses on developing frameworks for predicting when and where genetic signals are likely to be measured, as well as assessing the threats that exposure to such materials poses to microbial communities

Micro- and nano-particle (colloid) transport distances in sediment are of concern for both assessing contamination risk, such as from pathogens and microplastics, and for enabling the targeted delivery of contaminant remediation agents. The accurate prediction of colloidal transport has been long confounded by colloid-surface repulsion dynamics, which can cause observed transport patterns to strongly deviate from traditional model predictions. Our group is working to develop and improve colloid filtration models to better predict observed transport trends. 

Chemical reactions in natural and engineered water systems fundamentally depend on reactants coming into physical contact, a process governed by mixing. Computational models used to predict water quality typically assume that reactants are uniformly distributed and available to react at the rate their chemistry alone would predict. Experimental evidence shows this assumption is frequently violated in flowing systems, producing significant errors in predicted product concentrations. Our work focuses on improving the accuracy of reaction predictions by relating the rate at which reactants come into contact to their residence time in the environment.