Observing Flooding Extent in South Florida from ‘Super Camera’ 700km above the Earth
Author: Boya Zhang
Looking
at the title, you must be thinking I am crazy. And you will confirm that I am
crazy, if I tell you this will be part of my PhD dissertation. You may also be wondering
what is ‘super camera’? Why it is flying or floating in the 700 km in the outer
space? Even if it exists, why and how to use this strange thing to observe South
Florida flooding? Let me solve the puzzle.
The
‘super camera’ is not like a Canon digital camera. Indeed, it is a radar, which
is short for ‘radio detection and ranging’. Simply speaking, one of the basic
tasks of radar is to detect the signal and measure the distance. But, detect what signal, and measure the
distance of what? My research uses a special type of radar, and it has a very
unfriendly name- ‘Synthetic Aperture Radar’ or SAR. (Let’s skip the meaning of
the first two words of the name, because it may take another article to
explain!) SAR is able to actively transmit microwave signal towards the Earth
surface. Indeed, the signal wavelength is very similar to the microwave oven at
your home, but don’t worry, it will not cook the Earth like eggs, or heat
anybody walking on the streets. The
transmitted microwave is able to reach the Earth surface, interacts with
objects on the ground and bounces back towards SAR (It sounds crazy, but
believes me, it is true…). Therefore, the answers to the first question is that
SAR detects the signal it transmits by itself. And the answer to the second
question is: it measures the distance between itself and Earth surface.
You may
be thinking that SAR is similar with other optical sensors like the popular
Landsat. Indeed, SAR has both advantages and disadvantages when compared with
the optical sensors. One of the advantages is that SAR also works during the
night! It is simply because it transmits signal by itself, unlike the optical
sensors, which rely on the sun light. Another advantage is that microwave is
generally able to penetrate atmosphere, especially clouds. When the cloud is
dense, the sight of optical sensors can be completely blocked. Especially, when
there is heavy rainfall, the optical sensors can hardly take ground information.
The disadvantage of SAR is that, it requires more complicated data processing.
Now, we know SAR is able to remotely sense the ground, and it is able to work regardless of illumination or clouds, but how does it observe the flood on the surface? Remember that previously, it is said SAR signal interacts with objects on the ground. The flood observation is associated with the interaction. But before we dig into the concepts, let’s take a look at some SAR images of South Florida, taken from a satellite called Sentinel-1, launched by European Space Agency.
Fig 1.1: Aug 29th |
Fig 1.2: Sept 10th |
Figure
1. Sentinel-1 SAR intensity image acquired in South Florida on Aug 29th
(left) and Sep 10th (right), 2017 respectively
Just by
looking at both figures, you can immediately tell that the left one is much
brighter than the right one. Wait a minute, what does this brightness and
darkness mean? The brightness means there is more energy bounced back to SAR,
whereas the darkness means a lot of energy bounces away from SAR. Why the two
images, only 12 days apart, have such a large difference? Do you remember where
you were on Sep 10th, 2017? I remember it so clearly, because it was
only 3 weeks after I moved to Miami and had just started my PhD program (with
full excitement!). Hurricane Irma landed in Florida on that very day, Sep 10th,
2017! My wife and I had to escape to Orlando with ‘fear’ that Irma will catch
up us on the highway. Now, it is so
clear that the difference between the two images is attributed to flooding! I
guess tons of rain fell in the Everglades and changed everything. But you may
want to ask a deeper question: why did flooding make the image darker? Let’s
take a look at Figure 2 below:
Fig 2.1 |
Fig 2.2 |
Fig 3.3 |
Figure
2 The interaction between the SAR signal and ground surface.
Fig 2.1 shows when SAR signal interacts with rough surface, it scatters and SAR is
able to receive the portion of scattering back towards it. Fig 2.2 one shows
when the surface is flooded, the water surface is like a mirror and the signal single-bounces
away from the SAR. It actually explains the darkness of the Figure 1 Sep
acquisition. Fig 3.3 shows, at the presence of a tree, the signal
experiences double-bounce, at the water surface and then on the tree trunk,
which makes high energy signal to bounce back to the satellite. It explains the
bright part of the same figure, which is located at west and east parts. Indeed,
this part can have higher intensity than the left plot of Figure 1. It does not
look that bright, because of the stretch of pixel brightness for visualization
effect. At last, I will show you two more figures.
Fig 3.1 |
Figure 3.1 Land Cover distribution in South Florida. Fig 3.2: Flooding detection
result. Red patches mean higher SAR signal intensity during flooding comparing
with the unflooded situation, whereas the blue patches mean lower intensity. Both
red and blue patches are detected with flooding.
Fig 3.2 |
These
two figures are from one of my publications (Zhang et al. 2018). It is
interesting to see that the marshes are corresponding to the blue patches on
the right figure, while mangroves in southwest, other woody vegetation in
northwest, and urban areas in the east match with red patches. It is perfectly
consistent with our model, that submerged marshes have single-bounce while
flooded woody vegetation and urban areas have double-bounce effect (building
plays the same role as tree trunk). In this way, we finally explain how we
observe flooding extent in South Florida from ‘Super Camera’ 700km above! If
you look at the Figure 3, there is a legend, which provides a range of an
index, which is used to tell us whether there is flooding or not. The detail of
this index is written in my paper ‘Mapping the Extent and Magnitude of Severe Flooding
Induced by Hurricane Irma with Multi-Temporal Sentinel-1 SAR and InSAR
Observations.’ This also includes zoom-in views of the three yellow frames. Feel
free to contact me if you are interested and have more questions! My next blog
will show what else SAR can bring to us and benefit our society!
Contact:
Boya Zhang (Paul), PhD student in Florida International University, Earth and
Environment Department. Email Address: bzhan018@fiu.edu
Comments
Post a Comment