Table of Contents
- 1. Research Setting
- 2. Primary Research Question
- 3. Investigation Plan
- 4. Data Access and Visualization Methods
- 5. Preliminary Analysis
- 6. Refinement of Analysis
- 7. Statement of Results
- 8. Discussion of Results
- 9. Statement of Conclusions
- 10. Questions for Further Investigation
1. Research Setting
The research settings for this tutorial are locations in the world ocean. The world ocean constitutes several major oceans and oceanic basins. The major oceans of the world ocean are the Atlantic, the Pacific, the Indian, the Arctic, and the Southern Circumpolar. The oceans are further subdivided into gyres that are generally defined by a quasi-circular circulation pattern: these are notably found in the north and south Atlantic and the north and south Pacific. Obvious there are other, smaller divisions with various names, such as the Arabian Sea and Bay of Bengal in the Indian Ocean, and the Bering and Chukchi Seas in the Arctic Ocean. The world ocean also includes a few bodies of water that are mostly separated from the primary oceanic basins. Noteworthy bodies of water in this category would be the Mediterranean Sea, Red Sea, South China Sea, Sea of Japan, Sea of Okhotsk, Caribbean Sea, Gulf of Mexico, and several other semi-enclosed smaller basins.
2. Primary Research Question
The primary research question in this tutorial concerns the further use of sea surface temperature (SST) versus chlorophyll scatter plots for diagnostic purposes. In this tutorial, the question to be examined is:
Do SST versus chlorophyll scatter plots produce patterns that characterize specific locations in the world ocean?
3. Investigation Plan
To answer this question, this tutorial will proceed in the form of a "mystery" investigation. Three different scatter plots will be presented, the characteristics of the water body will be determined solely from the scatter plot, and then the location which yielded the scatter plot will be revealed.
4. Data Access and Visualization Methods
The system we will be using for this tutorial is Giovanni’s “Ocean Color Multi-Parameter Intercomparison System”. This system allows comparison of several data parameters in a variety of different ways. The system utilizes SeaWiFS data and MODIS-Aqua data, so at this time only a couple of years of MODIS-Aqua data are available for intercomparison. (Future plans include the addition of MODIS-Terra SST data, extending the SST data in the system back to the year 2000.)
If you have entered this tutorial without seeing Tutorial 5, "Sea Surface Temperature versus Chlorophyll Scatter Plots, Part I", now would be a good time to read Section 4 in that tutorial. (Just click the linked title of the tutorial.)
5. Preliminary Analysis
To commence the Preliminary Analysis, the three "mystery" scatter plots are shown below. Click on each one to see the full-size version.
It should be somewhat obvious that the appearance of each of these scatter plots is quite different from the others. Note that each of the scatter plots was generated for the same month of data, April 2005.
6. Refinement of Analysis
Now each scatter plot will be examined in turn, with the important features highlighted.
The first one:
The three highlighted features are: one, most of the chlorophyll concentration values are low, which means they are less than 1.0 mg m-3; two, there is an inverse relationship between SST and chlorophyll concentration, meaning that higher chlorophyll concentrations are found with lower SST values; and three, there is a small cluster of high SST, high chlorophyll data points.
The second one:
The three highlighted features of the second scatter plot are: one, the majority of the chlorophyll concentrations are very low, none higher than 0.6 mg m-3 and a lot of concentrations less than 0.2 mg m-3, while the SST values are warm, between 18 and 29°C; two, there is an inverse relationship between SST and chlorophyll concentration; and three, there is an intriguing "three-fingered" pattern with high chlorophyll values and SST at approximately 28, 24, and 19° C.
The third one:
Three highlighted features of this scatter plot are: one, SST values get down to cold temperatures, as low as 2° C; two, in this plot there is a positive correlation between SST and chlorophyll concentration, meaning that higher SST is associated with higher chlorophyll concentration; and three, there aren't very many high chlorophyll values in this plot, with the majority between 0.06 and 0.6 mg m-3.
7. Statement of Results
The three scatter plots show decidedly different patterns for the relationship between SST and chorophyll concentration in the month of April 2005. Because these patterns are distinct, these scatter plots do provide a means of distinguishing surface water body characteristics based on the distribution of SST and chlorophyll concentration within the water body. However, the pressing question is: what locations generated the data for these three scatter plots?
To determine the locations that provided the data, the general characteristics of each scatter plot should be considered. For the first plot, the generally warm SST values indicate a temperate or tropical ocean, not a polar ocean. The inverse correlation between SST and chlorophyll concentration which dominates the data distribution is likely indicative of the presence of upwelling, where colder subsurface water comes to the surface with nutrients that foster phytoplankton growth. The cluster of high SST, high chlorophyll values is generally a sign of warm shallow water (similar to the coastal Yucatan waters discussed in Tutorial 5).
There are a few locations that have these characteristics, but the region that generated the data for the first scatter plot is the eastern equatorial Pacific Ocean, shown below. (Click any of the images below to see the full-size plot.)
The Peru Current upwelling zone on the coast of South America provides the majority of the high chlorophyll, lower SST data points, and the shallow waters near the Galapagos Islands and on the coast of Colombia likely provide the high SST, high chlorophyll data points.
For the second plot, SST values are still warm to very warm, indicating the likelihood of a tropical location. The chlorophyll concentrations are very low, which is characteristic of the centers of the large oceanic gyres (see section 10 for more about this). The three-fingered pattern indicates that the area touches on waters with higher and lower SST values, so this also is indicative of a gyre, where the latitudinal extent of the gyre includes mid-latitude and low-latitude (tropical) waters. There is also an inverse correlation between SST and chlorophyll concentration, which might indicate the presence of upwelling — but not always.
The location which provided the data for the second scatter plot is the Sargasso Sea, in the central north Atlantic Ocean. In the area plot below, the islands in the lower right corner are the Cape Verde Islands off the coast of Africa. (Click the image for the full-size version.) Note that the customized color palette option in Giovanni was used to generate this image, due to the very low chlorophyll concentrations. The color scale didn't quite fit in the plotting box; the minimum value is 0.01 mg m-3 and the maximum value is 0.5 mg
The reason that there is an inverse correlation between SST and chlorophyll concentrations here is due to the boundary of the Gulf Stream, which marks the northern extent of the Sargasso Sea. Even though the Gulf Stream is a warm current, the water in the Gulf Stream is still a bit cooler than the Sargasso Sea gyre, and the Gulf Stream has slightly higher chlorophyll concentrations. This is the bottom "finger" of the distribution. The center "finger" is due to the Cape Verde Islands, where water temperatures are higher but not as high as the coast of South America, which is just off of the bottom left corner of the plot above. The outflow of the Amazon River provides nutrients that increase the phytoplankton growth, and this creates the high chlorophyll, high SST top "finger" in the scatter plot. The MODIS-Aqua SST plot for April 2005 for the Sargasso Sea illustrates this well. 18-19° C water in the northern Sargasso Sea is purple; 24-25° C water near the Cape Verde Islands is light blue; and 28-29° C water off of South America is light green.
In consideration of the third scatter plot, one of the dominant characteristics is the chilliness of the water, which gets down to near 2° C. This is certainly not a tropical region! And the SST values are low, which indicates that the region did not include a coastal area. So this data distribution is probably from a polar ocean, the Arctic or Southern Circumpolar. One hint — April 2005 is early spring in the Northern Hemisphere and early autumn in the Southern Hemisphere, and the SST in the plot gets as warm as 16° C, so this is probably not early spring SST in the Arctic.
The location that provided the data for the third scatter plot is the Southern Circumpolar Ocean south of Australia. The top area plot below shows just the area that provided the data; the bottom plot moves the northern boundary of the area box 10 degrees north, to show where Australia is located. The 0.01 to 0.5 mg m-3 dynamic color scale was also used for these two plots.
SST in the southern range of this region does indeed approaching the freezing mark, but the Southern Circumpolar Ocean warms up considerably toward its northern boundary. The somewhat higher chlorophyll concentration band south of Australia marks the position of one of the three fronts that are found in the circulation of the Southern Circumpolar Ocean (see Section 10 for more on this topic).
9. Statement of Conclusions
In this tutorial, further examination of "mystery" SST versus chlorophyll concentration scatter plots has indicated that this type of scatter plot does provide a way of determining water mass characteristics for locations in the world ocean. Although the relationship between SST and chlorophyll will vary seasonally, particularly in locations where there is a strong seasonal cycle, in many locations the general relationship will persist though its shape will change somewhat over a year. The three locations considered here indicate the following for SST versus chlorophyll concentration scatter plots:
- Central oceanic basin gyres such as the Sargasso Sea are characterized by very low chlorophyll concentrations, elevated SST, and may display a multi-"finger" pattern if the boundaries of the gyre are included in the area generating the scatter plot data.
- Locations which include a permanent upwelling zone, such as the Equatorial Pacific including the Peru Current upwelling zone, will display a dominant inverse relationship between SST and chlorophyll concentration.
- Polar oceans such as the Southern Circumpolar Ocean have cold SST values and low chlorophyll concentrations.
Note that higher chlorophyll concentrations will generally be found in any oceanic location where oceanic fronts or convergence zones are included in the area which provides the data.
10. Questions for Further Investigation
This tutorial has touched on a few topics that can be considered for further investigation. One topic is the extremely low chlorophyll concentrations that are found near the center of the major oceanic gyres, a condition that could be referred to as "hyper-oligotrophy". Read the Science Focus! article "The Blue, the Bluer, and the Bluest Ocean" for more information.
Another topic of interest is the concentric circulation of the Southern Circumpolar Ocean, which is discussed in two Science Focus! articles: "The Low Zone" and "South Georgia: A View Through the Clouds". Also, the LOCUS Education Module "The Southern Ocean" describes and illustrates this region. Look for the three major frontal zones that are primary features of this important oceanic current.
Finally, much of the Southern Circumpolar Ocean is called an "HNLC" region — where HNLC stands for "High Nutrient Low Chlorophyll". The reason for the low chlorophyll concentrations in the presence of sufficient nutrient concentrations is the low concentration of dissolved iron. Several Science Focus! articles address this topic: