Sunday, November 6, 2011

A visit to Smart-R

I spent two hours in the Smart-R radar truck Friday noon. Smart-R is a typical mobile Doppler weather radar. Its antenna has a diameter of 2.4 meters, scanning in two directions: either vertically at constant azimuthal angle, or horizontally at constant elevation angle. A combination of different scan angles, called scan strategy, determines how radar "sees" cloud. It takes 10 minutes for Smart-R to finish a scan cycle. Here is a photo looking to southeast from half a mile away. Smart-R is tiny sitting in the red circle of this photo:

A closer look:

The bicycle to the left of the truck is the one Adam bought at upper west side of Manhattan and Daehyun brought all the way here. See the trees on the right of the truck; these are also in the first photo (all those tree to the right). These trees cause significant blockage of radar signals. Let's take a look at the workstation inside the truck:

Two radar images were on the top. There are lots of green area in the right top of the monitor from 5 clock to 10 clock. These are blind spots of Smart-R because of the tree blockage I mentioned above. Smart-R also leaves 15 degrees in north-northwest for people to walking to the truck. So Smart-R detects cloud systems only from 11 clock to 4 clock. Other than trees, buildings and ships and cranes can all possibly create blockages that we really want to avoid. There will be some construction work close to Smart-R during DYNAMO, and cranes will be a big concern.

At the moment, Smart-R is maintained by a small group of people: Courtney, Johathan and Emily. During my two weeks here, I have been working with them and Chidong, our big boss, to gather some statistical information from Smart-R data. Here is one of my outputs:

The colors are relative humidity collected from weather balloons; the blue curves shows counts of daily "Echotop", a quantity from Smart-R indicating the height of cloud top; the green curve in the bottom show the daily rain rate from Scott at S-Pol. This picture shows how the height of cloud top, relative humidity and rainfall varied over October. It's interesting to DYNAMO scientists because it may indicate some inherent relationship among these variables. In the end, we will use statistics like this to learn a great deal about climate in Indian ocean over the the 6 month period of DYNAMO.


  1. Nice plot! To first order it seems to show what we expect... deeper clouds with a deeper moist layer. Right?

    And, great post.

  2. Right. No surprise here.
    These are daily averaged data. It is a bit difficult to see which (deeper clouds or deeper layer) comes first in time.

  3. I agree that this is a very nice plot, Shuguang.