Modeling temporal patterns of greenhouse gas emissions from excavated fish ponds

Aquaculture is the world’s dominant source of fish production, having recently surpassed capture fisheries. Most aquaculture operations occur in freshwater environments, utilizing either excavated ponds or cage culture where mesh enclosures are deployed in existing water bodies such as reservoirs to confine culture animals. Recent studies have demonstrated that freshwater aquaculture systems are non-negligible sources of greenhouse gases (GHG) such as carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O). These emissions are temporally variable, but measurements are costly and time-consumingand there is a notable scarcity of predictive models tailored to the aquaculture sector. Drawing on temporally and spatially resolved data from aquaculture systems in Brazil, we use machine learning to model the temporal variability in CO2, CH4 and N2O fluxes from excavated ponds and cage culture systems. Through our research, we aim to elucidate drivers and patterns of temporal variability in GHG emissions from these distinct yet widespread aquaculture production systems. This could help correct datasets lacking temporally resolved measurements.

Primary Presenter: Maria Camila Mejia Garcia, University of Texas Rio Grande Valley (mariacmejiag80@gmail.com)

Authors:

Maria Camila Mejia Garcia, University of Texas Rio Grande Valley  (mariacmejiag80@gmail.com)

Rafael M. Almeida, University of Texas Rio Grande Valley  (rafael.almeida@utrgv.edu)

Hansapani Rodrigo, University of Texas Rio Grande Valley  (hansapani.rodrigo@utrgv.edu)

Marcelo Gomes da Silva, University of Texas Rio Grande Valley  (marcelo.gomesdasilva@utrgv.edu)

Nathan Oliveira Barros, University Federal de Juiz de Fora  (nathan.barros@ufjf.edu.br)