This project sought to implement the analysis of the different data that can affect the behavior of birds, as well as to predict the possible changes that could arise during their migration. Our main ideal is to know how the factor data collected by earthdata can reveal the causes of fluctuating migration of birds.
We developed a prediction model based on machine learning, cross validation and LTSM capable of graphically generating the behavior of bird migration, using the data collected, showing how they progress according to changes in the ecosystem.
Knowing how animals behave due to changes in their environment is essential to be able to act and prevent new species from entering danger of extinction and also to be able to generate predictions that help study ways to improve the quality of life of migratory or stationary animals.
Basically our proposal tries to predict the behavior of some species of migratory birds, using data that could influence and affect flight routes, as well as nesting points.
It works in a similar way to other prediction models, however we tried to use more data that, we presume, could affect the common behavior of these birds, such as rainfall, winds, as well as old flight paths. We hope to inform you about the importance of the changes that species make, based on environmental variations and how these can affect their behavior and life, and how they subsequently affect other species, including people.
Our inspiration for the realization of this project is the study of the planet's species, mainly in how the behavior of different life forms are affected due to the changes that can occur on the planet itself (climate change, temperature change , precipitation, vegetation, etc.). We also believe that these kinds of studies could help to discover or facilitate the discovery of factors that give rise to or affect the behavior of life forms on other planets.
The impact that we would have with this project is to be able to understand and observe the different classifications of birds and the environmental impact that this has. With this, we try to predict their migration and possible new species of birds that exist and have not yet been discovered.
Our approach was based on climatological data and data from different species of birds.
For development We use Python, Visual Studio Code, Anaconda and Excel (for csv data)
The implementation of the code and the mixing of the different information sources that could have a relationship or direct or indirect effects on the species or species chosen for the study.
Another factor that we must take into account when carrying out similar studies is the knowledge that in biology, and the behavior of the species in particular, since the basic notions would not allow an in-depth study of the changes that directly affect them.
As main data tool we use an external dataset, of the behavior of 2 flocks of migratory birds based on their life traffic in the year 2013-2014. We use it to know the behavior of various groups of birds and characteristics of life form.
We also use the earthdata species distribution tool, in which we can mention snow cover, temperature and precipitation. NASA data is very important as it helps us find those factors that affect bird migration. It could even help accurately map changes in real time as this data is associated with that of birds and matches are generated.
· Birds Data: https://www.datarepository.movebank.org/handle/10255/move.478
· behavioral factors: https://earthdata.nasa.gov/learn/pathfinders/biodiversity/species-distribution
· For the future, it could be used to map the migration: https://earthdata.nasa.gov/eosdis/science-system-description/eosdis-components/gibs