Presenter Profile
Erika Barcelos
Postdoctoral Scholar
Case Western Reserve University, Department of Materials Science and Engineering
Erika is currently a postdoctoral scholar in the Department of Materials Science and Engineering at Case Western Reserve University(CWRU) . Her work at SDLE/CWRU focuses on geospatial analytics, fairification of data and statistical learning as a part of CASFER NSF Engineering Research Center and Material Data Science for Stockpile Stewardship Center of Excellence (MDS3-COE). She obtained a duo PhD in Polymer Engineering from the Macromolecular Science and Engineering department at CWRU and Mechanical Engineering from PUC, Rio in Brazil. Previous to joining CWRU, she obtained a Master in Chemical Engineering from Ecole de Mines de Saint Etienne in France and a bachelor in Chemical Engineering. Her main experiences relies on molecular simulation and data science, where she has been working for several years. Erika is passionate about solving real world problems using data science and advanced computational tools.
TALK TITLE
An overview of geospatiotemporal analytics, applications of satellite imagery data to monitoring nutrient flows through the environment
KEYWORDS
Geospatiotemporal predictive models, nutrient flows, satellite imagery
ABSTRACT
Watersheds are land areas where water streams coming from rainfall, snowmelt and rivers flow to larger water body areas such as reservoirs, lakes, bays and oceans. Nitrogen and phosphorus are some of the main components of fertilizers in agriculture and in combination with animal waste from concentrated feeding animal operations(CAFO) it causes serious environmental problems. The amount of nitrogen and phosphorus that ends in the water bodies may vary depending on a series of variables such as climate conditions, land elevation, soil type, presence of CAFOS etc. This work focuses on monitoring and tracking the contribution of nitrogen from various sources in watersheds using spatiotemporal analytical tools. To fulfill these objectives, different datasets having nationwide coverage have been collected, fairified and ingested into our CARDLE/HPC environment. The datasets having different formats, data types, spatial and temporal resolution were integrated and important insights about nitrogen distributions were acquired. We could map the nutrient flow and draw correlations between properties in soil, water, elevation and crops data in the states of Ohio, Texas and Florida. Our efforts in collecting and integrating new data sources to the existing data is ongoing. In addition to new data being collected and ingested, we are also expanding our analysis from individual states to continental United States.