Welcome to the InfoSci Platform
Reference Hub1
New Approach Based on Termite's Hill Building for Prediction of Successful Simulations in Climate Models

New Approach Based on Termite's Hill Building for Prediction of Successful Simulations in Climate Models

Mohamed Elhadi Rahmani, Abdelmalek Amine, Reda Mohamed Hamou
Copyright: © 2017 |Volume: 8 |Issue: 3 |Pages: 18
ISSN: 1947-9263|EISSN: 1947-9271|EISBN13: 9781522513551|DOI: 10.4018/IJSIR.2017070103
Cite Article Cite Article

MLA

Rahmani, Mohamed Elhadi, et al. "New Approach Based on Termite's Hill Building for Prediction of Successful Simulations in Climate Models." IJSIR vol.8, no.3 2017: pp.43-60. http://doi.org/10.4018/IJSIR.2017070103

APA

Rahmani, M. E., Amine, A., & Hamou, R. M. (2017). New Approach Based on Termite's Hill Building for Prediction of Successful Simulations in Climate Models. International Journal of Swarm Intelligence Research (IJSIR), 8(3), 43-60. http://doi.org/10.4018/IJSIR.2017070103

Chicago

Rahmani, Mohamed Elhadi, Abdelmalek Amine, and Reda Mohamed Hamou. "New Approach Based on Termite's Hill Building for Prediction of Successful Simulations in Climate Models," International Journal of Swarm Intelligence Research (IJSIR) 8, no.3: 43-60. http://doi.org/10.4018/IJSIR.2017070103

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

Quantitative analysis of the failures and crashes in simulation of climate models can yield useful insights to better understanding and improvement of the models results using Intergovernmental Panel on Climate Change (IPCC) class. In this paper, the authors propose a new technique inspired by termite's hill building behavior to analyze a series of simulation in climate models and predict which one was succeeded within the Parallel Ocean Program (POP2) component of the community Climate System Model (CCSM4). The authors' approach is a distance based approach used to predict the success of the values of 18 POP2 parameters. And in order to predict better results, they used for each experiment one of the studies as a training set and two as a test set, then they used two of the studies as a training set and one as a test set. Results of classification were very satisfactory (Accuracy > 0.87). This paper gives a very useful method to quantify, predict, and understand simulation success in climate models.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.