To celebrate Independence Day, I thought it would be fun to dress up the ocean in a little red, white and blue.
If you’re curious, the image above represents the gradient of sea surface temperature (SST) at each point, and is based off of today’s 7-day composite of SST collected by the AVHRR instrument on NOAA’s polar orbiting satellites.
For every point, if the temperature change from its left (or below) neighbor to its right (or above) neighbor increases, then the gradient is considered positive and is colored red. Similarly, if the temperature decreases as you go from left to right or bottom to top, then the gradient is negative and is colored in blue. If the temperature doesn’t change much, white is used. The darkest colors (red or blue) represent a temperature change of around 2-3 degrees Fahrenheit over 2km of distance.
The map above is actually the sum of the horizontal and vertical gradients (i.e. dT/dx + dT/dy), so areas that are blue, indicate areas where colder temperatures can be found towards the Northeast.
In the image, a few patterns stand out, particularly the north wall of the Gulf Stream which shows up as blue streaks, indicating colder temperatures to the North. The waters off New England show up as a dark mess of blue and red, due to the large number of clouds in the area over the past few days, not to mention all the hot weather warming up the cold waters, both of which resulted in uneven measured temperatures, and therefore chaotic calculated gradient values.
Scientists often use a gradient calculations to identify large-scale features in the ocean, like fronts and eddies, which can be seen in the image. However, to do so they generally would calculate the gradient over larger areas than the 1km pixels used above or use a combination of filtering or averaging to smooth out the features and make them stand out more.
Using raw SST data to calculate gradients results in an image that is very noisy. But then, fireworks are noisy too… and they are also quite beautiful to behold.
Comments are closed.