# A Circular Sea Ice Puzzle (WEEK 6)

As I mentioned two blog posts ago, I suspect that less surrounding sea ice concentration correlates with higher rates of rift propagation. To determine sea ice concentration (SIC), I downloaded a bunch of data sets that give SIC values (between 0 and 1) in areas of 50,000-by-50,000 meters, and I extracted the values from right around the Amery Ice Shelf. However, clipping the SICs proved to be much harder than clipping the ocean temperatures.

The available SIC data

Similarly to the ocean temperature data, the sea ice data are available in dimensions of longitude, latitude, and time. However, these data are instead projected as if the south pole were flattened into a two-dimensional shape, with the center of the shape being the southernmost point on earth, latitude radiating outwards from that central point, and longitude appearing in a spiral pattern (see figures 1 and 2). In these files, the x-axis represents meters east or west from the center, while the y-axis represents meters north or south from the center. In turn, each of these data points has a corresponding latitude, longitude, and sea ice concentration value.

Figure 1: Degrees latitude around the South Pole

Figure 2: Degrees longitude around the South Pole

In order to determine the area of SIC around the Amery Ice Shelf, I altered the colors of the above pictures to isolate the targeted latitude (-68o) and longitude (72o) values. I then placed a dot there to see roughly how many meters (“Unknown Units” in figure 3) the latitude and longitude are away from the center.

Figure 3: Intersection of latitude and longitude around the Amery Ice Shelf in ArcMap

I then used the following x- and y- coordinate (or grid) pattern, as it is represented on the files.

 Starting value & meters Ending value & meters Interval (in meters) between each datum x grid 1         -3,937,500 316          3,937,500 25,000 y grid 1          4,337,500 332         -3,937,500 25,000

Finally, the table below shows an example of some of the data that I extracted. Around the point of 2,287,500 meters east of the south pole (250th x-grid value) and 712,500 meters north of the south pole (146th y-grid value), the water at (-68.1416o, 72.6995o) was 93% covered by sea ice and 7% uncovered (with a standard error of 0.0337). Around the point of 2,262,500 meters east of the south pole (249th x-grid value) and 662,500 meters north of the south pole (148th y-grid value), the water at (-68.5455o, 73.8481o) was 95% covered by sea ice and 5% uncovered (with a standard error of 0.3323). Also note that the 0’s represent areas with no sea ice coverage, and the NA’s and -0.03’s represent areas that would never have any sea ice coverage because they are generally over land.

How much data?

Unlike the ocean temperature data – which gives temperature estimates for 3-month ranges – the sea ice concentration data provides daily values. While the time resolution of these data is therefore much better, it also brings forth the unrealistic possibility of my extracting over 7,000 days’ worth of SIC. To keep myself from repeating this process 7,000+ times, I instead selected four days from each month, spaced roughly evenly apart.

I then began analyzing the data to find trends and patterns within it. However, I will end this blog post here, and for week 7 introduce my other data set (atmospheric / surface temperatures).