08 March 1999
Washington, D.C.

What Color is the Ocean?

and why do you need satellites to tell you?

One of the things that I had hoped to do while on board the Kaharoa during this expedition in addition to sitting in front of a computer screen working on these webpages, was to make measurements everyday around local noon with a little instrument called a Sunphotometer. This thing is about the size of a gameboy and more than once someone has yelled across the deck of a ship at me and said "Hey Gene, quit playing games and get back to work!". However, since it looked like I was not going to be able to spend any time at sea on board the Kaharoa, I decided to take at least one measurement as close to the ocean as I possibly could. Finding a nice, unobstructed part of the beach, I waited until just the right moment and took what I hoped would be at least one worthwhile set of observations.

Well, let me explain what I am doing and more importantly, why I am doing it. If someone were to ask you what color the ocean is, chances are that your answer would be "blue". And for most of the world's oceans, you would be right. However, wherever large numbers of microscopic, floating algae called phytoplankton are found, the water turns shades of green, red and brown. The more phytoplankton in the water, the greener it is....the less phytoplankton, the bluer it is. Pretty simple.

But what we really want to know is "how much phytoplankton is there"? and also "how does the distribution and abundance of phytoplankton change in time and space?". There are many reasons why we should care about the answers to these questions, and these along with some of the basic principles of remote sensing (I'll be getting to that in a minute) can be found in the Teacher's and Student's Guide at NASA's ocean color monitoring program called SeaWiFS.

Remote Sensing is simply observing something from far away. In many ways, much of what Clyde and Bernard will doing here on this expedition is remote sensing. When Clyde uses the sonar on the Deep Rover, or when Bernard listens to the sound of the sperm whales with his hydrophones, that is a form of remote sensing. What I do is to observe the earth from space. In particular, I work on a project whose mission is to monitor ocean color from space using a satellite called SeaStar that has just one instrument on it, called SeaWiFS. Every day, SeaWiFS scans the earth sending back nearly 2 gigabits of data which are processed to produce spectacular images of the earth's vegetation, both in the oceans and on land.

A little closer to home, the image on the right was taken by SeaWiFS on February 21, 1999 and shows the land and ocean features of New Zealand in much the same way as you would see them if you were flying high above the earth. Without getting into too much detail, in order to measure the effect of phytoplankton on the color of the ocean, we need to measure the amount of sunlight reflecting back from the earth after it has passed into the ocean and been reflected back out to space. Remember what I said earlier? The more phytoplankton in the water, the greener the light is that comes back out of the ocean. However, before that light reaches SeaWiFS which is flying overhead at 705 kilometers above the earth, it must pass through the atmosphere. The atmosphere has a very great effect on the light before it reaches the satellite and we must be able to remove that effect before we can calculate the part of the light that came out of the ocean. That is where my little sunphotometer come in. I bet you thought I had forgotten all about it, didn't you?

What I do is to point the sunphotometer at the sun at about the same time that the satellite passes overhead (usually around local noon) and measure the total amount of sunlight at very specific wavelengths. These measurements are then used to help verify our correction for the effect of the atmosphere on the SeaWiFS image take at the same time. This process is called Ground Truth. Without it, it would be very difficult to have very much confidence in our satellite observations. Other researchers make measurements in the water, both of the amount of light entering and leaving the ocean, and of the concentrations of phytoplankton in the water.

The measurements I make are very easy to do. All I do is:

  1. find a nice, sunny spot on the ship I happen to be on (the bow is my favorite spot)
  2. find something to brace myself against to avoid the effects of the ship's rolling (which is one of the reasons that I didn't try taking any measurements the other day when I went out with Bernard and Keith - the boat was just a little too small)
  3. enter some basic information from the little keypad such as time and location
  4. point the top of the sunphotometer at the sun and keep it centered in the cross-hair target
  5. press the SCAN button and wait a few seconds for it to run through all its measurements

I usually take ten measurements each day (scientists call this taking replicate samples) to minimize the effect of a single, incorrect sample. These measurements are then compared with the SeaWiFS observations made at the same time and same location. This comparison allows us to determine the accuracy of the algorithms (the computer programs that we use to process the SeaWiFS data) and our understanding of how the sensor might be changing over time.

SeaWiFS 8 year Biosphere animation

The result of all this can be seen in the image above. This remarkable view of the Global Biosphere, derived from SeaWiFS data shows the distribution of vegetation (plants and trees on land and microscopic phytoplankton in the ocean) around the world. What we are doing now is making maps like these every day to see how the earth's vegetation changes from day to day, from season to season and soon, from year to year. By monitoring these changes along with the processes that might be producing the changes we observe, we will be in a much better position to be able to predict how future changes in the earth's environment might influence life on this planet.


<-- Previous Journal | Next Journal -->