Inspiration
Plant growth is fun :grin: We want to know does having more species under a family means the family fits the area better? Do geographical locations affect plant diversity? We want to investigate if there is any relationship between what’s growing in an area and what “fits” the area. We also want to find if temperature and precipitation will affect the species of plants that are growing or would be suitable in an area.
What it does
In this project, we will be looking at some existing plant data in several counties of Alabama, and see if they are similar with earlier expectations of what should grow in that region. We will also look at a period of time to see the extent of weather data like temperature and precipitation could change over years, potentially meaning that what grows/ should grow there can be affected.
How we built it
The existing plant data were available as excel files and we counted the top 10 families in each county and made pie charts. The suggested plant data were manually made by finding the most recommended plants in a county and then recording their families. We compare pie charts to see if any significant difference exists between two datasets. We also take 2001 to 2009’s temperature and precipitation data out to see if the weather condition changes significantly over years.
Challenges we ran into
The original goal was to predict what plant can grow in the given region by soil and weather. In the coding portion of data manipulation, specifically merging geopandas and the shapefile as there were no common values, and the cleaning / manipulation of data. At the beginning of data cleaning the steps were done manually using excel functions after the realization that there will be a lot of data downloaded for counties and with a bit of extra time left functions were created to pass shapefile and csv files as parameters, to remove and add rows. Importing libraries challenge goals: The environment worked on was SpyderIDE and VScode, there were many issues when importing libraries such as matplotlib was not found even if it was downloaded using anaconda, geopandas was installed but errors of geopandas not installed kept appearing, to fix these issues these steps were followed for troubleshooting, making sure the correct environment had the libraries installed within Anaconda, installing the libraries using Anaconda, deleting and importing libraries again and restarting VSCode after installation seemed to fix the errors.
Accomplishments that we're proud of
This is a good chance for us to practice data analyzing skills, like creating charts and cleaning the data frames. We are proud that we were able to see some pattern in datasets that were extremely messy. The visualizations we made are new to us as well, like making gif files, creating function to draw plots, etc.
What we learned
We found some differences in top 10 plants across counties, while Poaceae, Fabaceae, and Asteraceae are the three most popular plants in all counties. We also found some contradiction between the actual data and suggested plants: Magnolia is not mentioned in top 10 families in every county, but it appeared in the plant suggestion a lot. Cyperaceae is a one of the major families that take about 5% of the total species in each county, yet it is not recommended that strongly according to the suggested data. We took a few years to look at temperature and precipitation change for each county. We did not see any significant change in temperature, while the change in precipitation was up to higher than 200% of the lower bound, we think the change in precipitation is significant enough to be concluded as, the precipitation does change a lot across years, which could potentially affect plant diversity in the year.
What's next for Plant Growth in Counties of Alabama, US
Our original plan was to create a model that could predict where vegetation is supposed to grow based on previous vegetation growth, temperature and precipitation but because of the time constraints we created visualizations to analyze the difference between years. In terms of visualizations a gif for the whole Alabama state that shows precipitation and temperature per each county will portray a visual representation of Alabama State. This can provide better insight within zones.
Log in or sign up for Devpost to join the conversation.