1. (7 pts) The data set NASA contains 24 24 grid of central America atmosphere mea-surements from NASA. In the data folder, you can nd the readme.txt which contains the basic information about the data set. There are several measurements and for this problem, you are supposed to work on the High Cloud Coverage. Design a nice one-page gure to visualize the High Cloud Coverage average over years for each of the 12 months (e.g. average January HCC over years, average February HCC over years etc.) over the whole spatial area, located by the latitude and longitude over a map background. That means, for each of the 24 24 locations, you have 12 numbers to show. And you need to show the information for all the 596 locations according to geographical locations. The data structure might be seemingly complicated, so we will go through the basic data processing together. But once you fully understand the DataStructure.R, things will be easier. We will also illustrate some basic ideas about how to visualize it in the lecture. No words are allowed for this question except the necessary captions/label-s/legend etc. Due the large volume of data you need to show, it is acceptable if one has to zoom in to see all details clearly. But you need to ensure: 1) you do have all details visible after zooming in (hint: use pdf as your output gure); 2) without zooming in, the high-level trend should still be visible. (2.5 pts for all details available, 2.5 pts for visible high-level pattern, 2 pts for correct map layout)
1
