Description: Describe the project and its functionality?
The project we created is an AI model that predicts when Ice caps will melt. This model will be extremely helpful for arctic and popular animal rescue services as well as scientists. Our AI model is extremely user-friendly as it only requires simple basic data from the user, such as ocean level and location. Due to the complexity and intelligence level of our AI model, only a few of these factors are needed to predict when the ice cap will melt. The function of the project is to serve as a tool in rescuing polar and arctic animals before the ice cap they live on melts. Another function of our project is that it can be used as a research tool for the melting factors of an ice cap. We also hope this Model will help raise awareness about the melting ice caps that are caused by climate change, and how we day by day are putting these vulnerable species such as the Arctic Fox and Polar bear, and how we humans are pushing them closer and closer to the brink of extinction. Our AI model’s results, as our AI has predicted that there are some Ice caps at risk of melting completely in the nearest couple hundred years, may change an individual's perspective about climate change, and may even cause them to help slow down the process of these ice caps melting.
Why this project is helpful to the community or the environment
The ice caps in the arctic environment are melting day by day, faster and faster, because of climate change. Many of these activities that contribute to the melting of these ice caps are human activities. Melting ice caps, sheets, and glaciers, causes sea levels to drastically rise, destroying the habitat for arctic animals in the process as their homes become one with the sea. The water that comes from these pieces of ice ends up in the ocean, making ocean levels rise. These higher ocean levels turn low-lying coastlines into a part of the ocean and cause flooding and major destruction of habitats all around the world. With our model, we are able to predict when these ice caps will melt. With the results and information that we get from our AI Model, we are able to easily identify when and where we need to interfere with the habitat of these high-risk melting ice caps.
Implementation / Methods: How did you create the project? What tools did you use? What was your implementation workflow?
First, we researched to find many sources of information about ice cap melting. We then came across a study about Greenland and the Icecaps that had been melting there. After that, we compiled many data sets from the Greenland studies into one master copy. The variables that our Dataset included were; Numerics, Stress Balance, Resolution, Initialization, Initial Year, Melt in Partially Floating Cells, Ice Front, Open Melt Parameterization, Standard Melt Parameterization, Initial SMB, and Initial Basal Melt. The next step was to create an AI model that can predict when the ice caps will melt. I decided to use IBM Watson Data Studio for this part. So first I uploaded the data file to Watson Studio. Then I added an AutoExperiment and asked it to predict when the ice cap will melt based on the values of the other columns. AutoExperiments will use several different algorithms and show the error for each one of them. I chose the one with the smallest error, one of the Ridge algorithms.
Results - What are the key findings of your project? What has your project achieved?
Our project has created reliable and accurate results on when icebergs and glaciers will melt using artificial intelligence. Our AI Model has persistently and continuously given us accurate results of when ice caps are going to melt. The time periods the AI Model predicts are logical. We discovered that some of the icebergs are only a few hundred years from melting and that we need to take action now to prevent the loss of habitat for arctic animals and the animals that live along coasts and low-lying areas. We also learned that there was a lot of variation between the amount of time each iceberg took to melt due to location, age, etc..
Citation
Dataset Data citations: Data usage notice: If you use any of these results, please acknowledge the work of the people involved in the process producing this data set. Acknowledgements should have language similar to the below (if you only use CMIP5 forcing, remove CMIP6 and vice versa).
“We thank the Climate and Cryosphere (CliC) effort, which provided support for ISMIP6 through sponsoring of workshops, hosting the ISMIP6 website and wiki, and promoted ISMIP6. We acknowledge the World Climate Research Programme, which, through it's Working Group on Coupled Modelling, coordinated and promoted CMIP5 and CMIP6. We thank the climate modeling groups for producing and making available their model output, the Earth System Grid Federation (ESGF) for archiving the CMIP data and providing access, the University at Buffalo for ISMIP6 data distribution and upload, and the multiple funding agencies who support CMIP5 and CMIP6 and ESGF. We thank the ISMIP6 steering committee, the ISMIP6 model selection group and ISMIP6 dataset preparation group for their continuous engagement in defining ISMIP6."
You should also refer to and cite the following papers:
Heiko Goelzer, Sophie Nowicki, Anthony Payne, Eric Larour, Helene Seroussi, William H. Lipscomb, Jonathan Gregory, Ayako Abe-Ouchi, Andy Shepherd, Erika Simon, Cecile Agosta, Patrick Alexander, Andy Aschwanden, Alice Barthel, Reinhard Calov, Christopher Chambers, Youngmin Choi, Joshua Cuzzone, Christophe Dumas, Tamsin Edwards, Denis Felikson, Xavier Fettweis, Nicholas R. Golledge, Ralf Greve, Angelika Humbert, Philippe Huybrechts, Sebastien Le clec'h, Victoria Lee, Gunter Leguy, Chris Little, Daniel P. Lowry, Mathieu Morlighem, Isabel Nias, Aurelien Quiquet, Martin Rückamp, Nicole-Jeanne Schlegel, Donald Slater, Robin Smith, Fiamma Straneo, Lev Tarasov, Roderik van de Wal, and Michiel van den Broeke: The future sea-level contribution of the Greenland ice sheet: a multi-model ensemble study of ISMIP6 , The Cryosphere, 2020. doi:10.5194/tc-2019-319
Sophie Nowicki, Antony Payne, Heiko Goelzer, Helene Seroussi, William Lipscomb, Ayako Abe-Ouchi, Cecile Agosta, Patrick Alexander, Xylar Asay-Davis, Alice Barthel, Thomas Bracegirdle, Richard Cullather, Denis Felikson, Xavier Fettweis, Jonathan Gregory, Tore Hatterman, Nicolas Jourdain, Peter Kuipers Munneke, Eric Larour, Christopher Little, Mathieu Morlinghem, Isabel Nias, Andrew Shepherd, Erika Simon, Donald Slater, Robin Smith, Fiammetta Straneo, Luke Trusel, Michiel van den Broeke, and Roderik van de Wal: Experimental protocol for sea-level projections from ISMIP6 standalone ice sheet models, The Cryosphere, doi:10.5194/tc-2019-322, 2020.
Log in or sign up for Devpost to join the conversation.