On Monday 7/20 my mentor invited me to visit her workplace, the USGS, where I got to meet with a team of scientists. They work at the Eastern Geographic Science Center, which is a division of the USGS dedicated to evaluating the impacts of land use and land cover changes on natural resources and environmental health. There were two Information Technology specialists and a member of the team who helped put together the Hurricane Sandy data set I have been visualizing with Tableau. I showed them the visuals that I had created, and they asked me questions about my use of color and trends in the data.
A great suggestion that they gave me is to layer some of the maps that I have created to look for correlations. This is similar to what I did for my total organic carbon dashboard, where I used a US Population map to show that areas with higher amounts of organic carbon tend to have a larger population. Another interesting thing to try would be to see how the data set about flooded businesses compares to the chemical composition of the sediment in nearby areas. This could help identify which areas are more likely to be a greater environmental concern in case of future natural disasters when businesses are destroyed and they release certain substances and chemicals into the outside environment.
I would also like to find a way to standardize the data that I have been working with. Certain regions have greater numbers of sample sites, so when I put this on a graph it may look like an area was affected more than others when in fact it only appears that way because more data was taken from there.
I really enjoyed this opportunity to share my work with professionals in the field. I got a lot of great feedback and ideas for future projects. They also showed some interesting tools they've been trying out, such as R statistical software, which also creates cool visualizations. I'm excited to continue exploring different USGS data as well as different software.
A great suggestion that they gave me is to layer some of the maps that I have created to look for correlations. This is similar to what I did for my total organic carbon dashboard, where I used a US Population map to show that areas with higher amounts of organic carbon tend to have a larger population. Another interesting thing to try would be to see how the data set about flooded businesses compares to the chemical composition of the sediment in nearby areas. This could help identify which areas are more likely to be a greater environmental concern in case of future natural disasters when businesses are destroyed and they release certain substances and chemicals into the outside environment.
I would also like to find a way to standardize the data that I have been working with. Certain regions have greater numbers of sample sites, so when I put this on a graph it may look like an area was affected more than others when in fact it only appears that way because more data was taken from there.
I really enjoyed this opportunity to share my work with professionals in the field. I got a lot of great feedback and ideas for future projects. They also showed some interesting tools they've been trying out, such as R statistical software, which also creates cool visualizations. I'm excited to continue exploring different USGS data as well as different software.
No comments:
Post a Comment