Nowadays, with emerging research interest in hurricanes, a hurricane should be presented as a data structure so that hurricane information can be efficiently stored, organized and manipulated . Hurricanes research involves in hurricane circulation modeling, predicting hurricanes path, evaluate hurricane track forecast… Those researches have to deal with lot of hurricane information such as name, time, location, size, wind speed, ocean water temperature, humidity, moisture in atmosphere… [3,4] The information is not only numbers but also image, GIS data… [3,4] It is not a good idea to store each of this information as a separated data structure. By implementing a hurricane as a data structure, we no longer care about individual factors of hurricane but look at a hurricane as a whole thing. Different kinds of data structures are suited to different kinds of applications, therefore standardized hurricane data structure can be used especially for hurricane research. It will help researchers discuss with each other easier or they can cooperate easier by sharing their data with less effort and time. They just simply process hurricane data with standardized operations accompanied with the data structure . Hurricane data is large because they contain not only many factors but also real time information. Representing a hurricane as data structure will provide a means to manage this amount of data efficiently .
However, represent hurricane as only one data structure is not enough because “different kinds of data structures are suited to different kinds of applications, and some specialize to specific takes” . Therefore, it is more suitable if a hurricane can be represented internally as a “collections framework”. A framework will provide generic, basic interfaces, basic necessary methods, operations for accessing, retrieving, and updating data. Moreover, add-on operations can be implemented later for a framework that will suitable for more complex data processing. Another reason for considering representing a hurricane as a “collections framework” is memory usage. If there is a hurricane data structure, there will be variables defined as type of that hurricane data structure. Because hurricane data is large, large amount of memory is needed to load those variables in runtime. A hurricane collections framework will ensure the most generic interface, operations and will not take lot of memory compared to full hurricane data structure.
In conclusion, a hurricane should be presented as a data structure or a “collections framework” which is a better choice. However, there must be a lot of efforts to create an efficient hurricane data structure that effectively contains enough information and provides useful operations to manipulate hurricane data.
 Wikipedia’s data structure definition, http://en.wikipedia.org/wiki/Data_structure
 Note on Amortization, D. Sleator, http://www.cs.cmu.edu/afs/cs/academic/class/15451- s04/www/Lectures/amortize.pdf
 National Hurricane Center website, http://www.nhc.noaa.gov/  http://en.wikipedia.org/wiki/Tropical_cyclone