LOCUS Tutorial Research Project Two:
Seasonal Patterns and Mysteries in the Red Sea

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 setting for the second LOCUS tutorial research project is the Red Sea, the narrow body of water that lies between Africa and the Arabian Peninsula. For further information on the Red Sea, you may wish to read the Science Focus article, Ethiopia, the Red Sea, and the Nile River, which includes a SeaWiFS image of the entire Red Sea region.

The Red Sea is a body of water with a number of "limits" -- limited inflow of water from either end, through the Suez Canal in the north or the Bab el Mandeb to the south, which opens to the Gulf of Aden and the Arabian Sea. In many other bodies of water, restriction of water exchange is usually a main factor in the development of eutrophication, but the Red Sea is also very nutrient-limited, due to the desert areas that surround it. Therefore, the Red Sea has very clear water and many nearly-pristine coral reefs.

One of the main things that coral reefs require is very clear water, and the lack of nutrients in the Red Sea means that phytoplankton primary productivity is very low, too. This lack of phytoplankton productivity, combined with a lack of significant water or sediment input from rivers (about the only sediments that the Red Sea receives is dust from dust storms) means that the primary characteristic of the Red Sea's marine environment is very clear, very low productivity water.

2. Primary Research Question

This tutorial is an example of an open-ended question that allows investigation of the "unknown"; i.e., a question with an uncertain outcome. In the first tutorial, it was deemed likely that the El Niño/La Niña phenomenon would probably influence the productivity in the Gulf of Panama, and this supposition was confirmed by the investigation. In this case, the question was prompted by an intriguing question: since the Red Sea has such low primary productivity, does it have any seasonal productivity patterns that can be detected in the ocean color data set?

So, framing the primary research question more clearly:

"Can seasonal patterns of phytoplankton productivity be observed in the Red Sea?"

3. Investigation Plan

For this investigation, we again plan to utilize the SeaWiFS 9 kilometer resolution global data that are available for analysis using Giovanni. Use of these data provides the opportunity to examine patterns of chlorophyll concentration, which is an indicator of phytoplankton productivity, over several years. Our study area, as discussed above, will be the Red Sea.

4. Data Access and Visualization Methods

The SeaWiFS 9 km chlorophyll data are processed into monthly files containing the average chlorophyll concentration for each 9 x 9 km "square" area over the world's oceans. Giovanni accesses these files and provides the capability of selecting areas of interest for examination. Giovanni can be used to create a map of the chlorophyll concentrations for the area averaged over selected time intervals, concentration vs. time plots for chlorophyll concentration (averaged over the entire selected area), month-by-month animations of the data for the selected area, or Hovmoller plots for the area.

Hovmoller plots can be particularly useful data visualizations to detect variations over time and space. Hovmoller plots display data in a time vs. longitude or time vs. latitude format. Thus, for the same area, a comparison of the spatial variability over time is easy to comprehend.

5. Preliminary Analysis

The first step is to define the study area. The first figure shows the area chosen for this study, a box bounded by a northern latitude of 28.0 degrees N, a southern latitude of 12.5 degrees N, a western longitude of 33.0 degrees E, and an eastern longitude of 44.0 degrees E. This box includes the entire Red Sea and the Bab el Mandeb, but excludes the Gulf of Suez and Gulf of Aqaba to the north. The figure shows the monthly chlorophyll concentrations averaged over the year 2001, and indicates that there are apparently two distinctly different zones in the Red Sea, the northern zone with very low chlorophyll concentrations, and the southern zone with significantly higher concentrations.

The next step is to determine if there are seasonal patterns in the Red Sea chlorophyll concentrations. To determine this, the obvious choice again are the Hovmoller plots, which were also used in the first tutorial. So we first generate a Hovmoller longitude vs. time plot starting in January 1998 and ending in December 2003.

This Hovmoller plot indicates that there are definitely seasonal patterns in Red Sea chlorophyll concentrations! It also indicates that there may be a number of factors which likely affect the Red Sea's chlorophll concentrations. In particular, there is a regularly occurring pattern of missing data - the white spaces on the right side of the Hovmoller plot, which corresponds to the southern Red Sea. The left side also shows a recurring pattern, so there is also an apparent seasonal cycle here, in the northern Red Sea.

Note that the slight angle of the long axis of the narrow Red Sea (off the direct north- south line) makes this Hovmoller plot fairly easy to interpret, because the western latitudes of this body of water are in the north, and the eastern latitudes are in the south. This plot provides several different topics that can be investigated, but there is quite a bit to examine. It can be made simpler by looking at a Hovmoller longitude vs. time plot for one year. 2001 appears to be fairly typical, so now the data for just the year 2001 are plotted.

Now two distinct patterns can be easily distinguished. In the northern Red Sea (left side), there are higher concentrations in February, March, and April, most notably in March. In the southern Red Sea (right side), the missing data occur from May through September.

6. Refinement of Analysis

We'll examine the northern Red Sea first.

Northern Red Sea

Looking back at the area plot for 2001, it appears that southern boundary of the northern Red Sea can be set at about 21 degrees North. So now we use Giovanni to create a Hovmoller longitude vs. time plot for this area in the year 2001.

This looks familiar -- the reason for the white area to the right is that there is no Red Sea east of 39 degrees East at latitudes north of 21 degrees North! The next thing to do is to determine if there is anything that can be learned from a Hovmoller latitude vs. time plot of the same area.

While this plot looks familiar (even though north is on the left side of the plot rather than the right), there is an unusual feature, too: a narrow band of higher chlorophyll concentrations at about 23.3 degrees North. A 'spike' such as this might mean that the data are in error, so this is something requiring further investigation (this will be found in section 8.) But because it's there, we'll look at the seasonal pattern in the northern Red Sea north of the feature.

This is a nice 'clean' Hovmoller longitude vs. time plot for the northern Red Sea, from 25.0 degrees North to 27.5 degrees North. It indicates that the highest concentrations of chlorophyll are between 0.7 and 2.5 milligrams per cubic meter (mg m^-3) during the short productive period in early spring. However, Giovanni has a feature that allows the generation of a more accurate color scale. Up until now, all of the plots were generated using the "pre-defined" SeaWiFS color scale, which is a logarithmic color scale designed for the full range of chlorophyll concentrations found in the world's oceans. Giovanni allows the generation of a "dynamic" color scale, which uses the range of concentrations found in a particular time and space study area, such as the area being studied here. So now we generate the same Hovmoller longitude vs. time plot, using the dynamic color scale option.

Here we have a more precise display of the chlorophyll concentrations, showing that they only exceed 1.3 mg m^-3 during the spring. There is also a small area in the southern part of the study area with higher concentrations.

To get a better idea of what's happening, now we will generate an area plot for March 2001 using the dynamic color scale.

This plot looks strange, doesn't it? The reason it looks strange is that Giovanni chose a different dynamic color scale for March 2001 than it did for all of the data in the area in 2001 plotted in the Hovmoller plot. Giovanni makes the decision on the dynamic color scale based on all the data that are being analyzed. Because of this, there is a third option for the color scale: the "customized" option. This allows the creation of a specific color scale using minimum and maximum input values. So look back at the Hovmoller plot and choose that range, with a minimum of 0.2 mg m^-3 and a maximum of 1.3 mg m^-3.

This is much better. Higher chlorophyll concentrations can be seen in the northern part of the northern Red Sea, with some higher concentrations near the coast. And in the southern part of this region, we can see a small area of higher concentrations, near a small group of islands on the eastern coast. This is the same latitude as the higher concentrations seen in the Hovmoller plot, so it is likely that the islands create conditions favoring slightly higher chlorophyll concentrations.

So now let's see what the pattern in the northern Red sea looks like, when plotted as concentration vs. time from 1998 to 2003.

The seasonal cycle is well-defined, but note how low the average concentrations are in this plot. This is due to the averaging of areas with higher concentration and lower concentration. The higher concentration areas that occur in the spring are isolated to the most northerly region of the northern Red Sea.

This pattern is very characteristic of phytoplankton "spring blooms" which occur around the world in ocean basins, enclosed seas, and lakes. Because the Red Sea is near the Equator, spring occurs considerably earlier than in higher northern latitudes. So this analysis shows that there is a seasonal spring bloom pattern in the northern Red Sea. And it also raises a few questions -- see section 10 for more discussion.

Southern Red Sea

Now we turn our attention to the Southern Red Sea by generating a Hovmoller longitude vs. time plot for this region, the portion of the Red Sea south of 25 degrees North down to the Bab al Mandeb at 14 degrees North.

As noted earlier, there is a seasonal pattern of missing data. The color scale also indicates that concentrations in the southern Red Sea are considerably higher than in the northern Red Sea, and that's accurate. Yet there is still something more that can be learned by switching to the dynamic color scale:

This plot looks considerably different than the plot made with the pre-defined color scale. The concentrations here are still considerably higher than in the northern Red Sea: the minimum concentration chosen for this region is 0.5 mg m^-3, compared to 0.2 mg m^-3 in the northern Red Sea. What this plot demonstrates is that the concentrations are generally higher (which we already knew), but that there is a fairly dramatic "burst" of productivity occurring right around the time that missing data start to occur. Also note that the missing data are initially in the far south, and moves to the north as summer progresses.

There is no mystery here; the southern Red Sea seasonal pattern is dominated by the seasonal pattern that occurs in the Arabian Sea to which it is connected -- the monsoon. The summer monsoon winds stir up nutrients from deeper waters, which fosters productivity near the surface in the same fashion that occurs in the Gulf of Panama. During July and August, there is heavy cloud cover, which accounts for the missing data.

And there is one other factor in this desert region -- DUST. Dust is also stirred up and blown around by the monsoon winds. Giovanni also provides a way to look at dust, because Giovanni includes all of the SeaWiFS Standard Mapped Image products. Two of these products, the Angstrom coefficient at 510-865 nanometers (nm) and the aerosol optical thickness at 865 nm, are measurements of the amount of aerosol (very fine particles) suspended in the atmosphere. Dust is a form of aerosol.

The plot below is a Hovmoller longitude vs. time plot of the aerosol optical thickness at 865 nm over the entire Red Sea in 2001.

It is obvious from this plot that the highest aerosol optical thickness values occur in the same pattern as the missing data, which is due to the monsoon clouds. This plot shows that dust is being stirred up and transported over the southern Red Sea during the summer.

7. Statement of Results

• This investigation has determined that there are apparently two seasonal patterns in the chlorophyll concentrations in the Red Sea.
• The northern Red Sea has a short-lived spring bloom every year, occurring primarily in March.
• The southern Red Sea has generally higher chlorophyll concentrations than the northern Red Sea, and the highest chlorophyll concentrations occur during the months of the summer monsoon.
• Aerosol optical thickness over the southern Red Sea is also highest during the summer monsoon season, indicating dust transport by monsoon winds.

8. Discussion of Results

One of the remarkable aspects of the spring bloom in the northern Red Sea is that it is such a small bloom. The maximum chlorophyll concentrations in the northern Red Sea in 2001 were just over 1.3 mg m^-3, occurring in March. The bloom doesn't last long, and it doesn't cover a very large area. The obvious explanation for this is the lack of nutrients in the Red Sea. In Section 1, the fact that the Red Sea is very low in nutrients was noted, and it is a primary factor that favors the healthy coral reefs of the Red Sea. Even though nutrients will build up in the deeper waters (the reason for this is a lesson in marine chemistry), because the Red Sea is in a desert region, there is very little nutrient input from any other source. Therefore, the minimal productivity in the Red Sea relies on what are essentially "recycled" nutrients -- nutrients derived from the phytoplankton that thrive briefly each spring. See section 10 for questions that are related to this topic.

The southern Red Sea is different, but it is still in a desert region, so the nutrient supply is still limited. One source of nutrients might be the deeper waters of the Indian Ocean, if water flows from the Indian Ocean into the Red Sea. Also, the dust that is blown into the southern Red Sea is also a nutrient source, particularly for iron, but not for the essential nutrients nitrogen and phosphorus. See Section 10 for questions that are related to this topic.

The final topic that will be discussed here is the 'spike' seen in the Hovmoller latitude vs. time plot for the northern Red Sea. This spike occurs at 23.3 degrees North. Examination of the 2001 area plot shows that there is a small area of productivity at this latitude, right at the coast where the boundary between Egypt and Sudan meets the western coast, which is wavy line that resembles a river. The map is confusing, because there are two boundaries; the official (straight line) boundary at 22 degrees North, and the "provisional" boundary that goes above and below this line.

We'll use Giovanni one more time to focus in on this area in 2001.

In this plot, the small area of higher chlorophyll concentrations can be clearly seen. This plot is for the entire year: the Hovmoller plot indicated that this area of higher chlorophyll concentrations essentially disappeared between April and August -- which is summer in the desert, and the same time that the monsoon influences the southern Red Sea.

So is this a real feature, or is it erroneous data? In the wondrous World Wide Web, a lot of information can be discovered, but first it was necessary to determine where exactly this was. After consulting several maps, it was determined that the small peninsula of land north of the feature is most commonly spelled "Ras Baranis", and an apparently very small coastal town here is Port Berenice. The area south of Ras Baranis is noted for coral reefs, and is called Foul Bay. The largest reef in the area is St. John's Reef, and is a prime scuba diving destination for diving expeditions to this area (and not easily reached!) Most of this information was found on scuba diving Web sites. The map below provides a good picture of the area.

At the southern edge of this map, a small island called "Mirear Island" can be seen. Searching on the phrases "Foul Bay" and "Mirear Island" found a few descriptions of the area provided by adventurous sailors. Mirear Island is a small sandy island within a large reef complex that is connected to the shore, by description characterized by numerous large coral heads and difficult small channels that must be navigated with considerable care to reach safe anchorage. Notably, Mirear Island and the reef complex are located right at 23.3 degrees North.

It was also possible to find a Space Shuttle photograph of Ras Baranis and the Mirear Island reef:

This view is looking toward the south. The Mirear Island reef resembles a fan south of Ras Baranis.

So what is happening here? Based on the characteristics of other reef complexes like this, a good supposition is that the reef complex is generating a large amount of sediment: pieces of coral, stony fragments of the "leaves" of coralline algae, which grow on the bottom of reef lagoons, and most importantly, benthic foraminifera. Benthic foraminifera are one of two types of foraminifera. One type of foraminifera is a free-floating phytoplankton that is found in the surface waters of the oceans around the world. Benthic (from benthos, "bottom") foraminifera are larger foraminifera that live on the sea (or reef) floor. Both types of foraminifera form hard shells (also called tests) out of calcium carbonate, which also forms the skeletons of coral and the leaves of coralline algae.

Benthic foraminifera are important because they contain chlorophyll; coral and coralline algae also contain chlorophyll. In coral, the chlorophyll is found in algae that live in a symbiotic relationship with the coral polyps. (A large amount of information can be found regarding symbiotic algae on the Web.)

The proposed explanation for the higher chlorophyll concentrations around Mirear Island is that winds are transporting sediments that contain chlorophyll (composed of coral, coralline algae, and benthic foraminifera) from the Mirear Island reef just far enough offshore to be detected by SeaWiFS. SeaWiFS data processing will exclude shallow waters close to shore, but in deeper waters chlorophyll concentration data can be acquired. The Red Sea has a very small shallow coastal area -- the waters get very deep quite close to shore (many of the reefs are in water that is very deep for scuba diving because the water is so clear).

Wind is implicated because of the seasonal pattern, where the higher chlorophyll concentrations are absent in the desert summer. In spring and fall, warm and cold fronts produce wind (the spring dust storms that were encountered by the U.S. military during the war with Iraq illustrate this point quite well). However, in the heat of summer, the desert becomes calm and blisteringly hot. This climate pattern could explain why sediments containing chlorophyll could be transported far enough from the reef complex to be observed by SeaWiFS -- and thus the 'spike' in the data plot is probably not erroneous data, but is actually indicates an interesting interaction between the land, the reef, and the Red Sea.

9. Statement of Conclusions

• This investigation determined that there are distinct seasonal patterns in the SeaWiFS chlorophyll concentrations in the Red Sea. The northern Red Sea has a spring bloom pattern of increased chlorophyll concentration that occurs in February, March, and April. The southern Red Sea has a seasonal pattern of higher chlorophyll concentrations in the summer months, coinciding with the summer monsoon.
• Chlorophyll concentrations in the Red Sea are generally low. Chlorophyll concentrations are significantly lower in the northern Red Sea compared to the southern Red Sea. The lack of nutrients from rivers is likely a primary reason for the low chlorophyll concentrations in the Red Sea.
• Aerosol optical thickness over the southern Red Sea is highest in the summer months, coinciding with the summer monsoon.
• A small area of high chlorophyll concentration discovered in these analyses appears to be related to a coastal complex of coral reefs.

10. Questions for Further Investigation

The "open-ended" research question that initiated this tutorial shows that a fairly simple initiating question can foster a number of additional realms for inquiry. In section 8, one of these additional questions was investigated to an extent, but it is possible that still more could be learned. So here is a short list of several additional questions that are left unanswered -- questions that could lead to even more investigation and discovery.

Northern Red Sea

- Can any additional information be found regarding the apparent "spring bloom" pattern of productivity in the Red Sea? (An example of an answer: searching the Web using the phrase "Red Sea" and the word "seasonality" provided the following link: Ben-David-Zaslow et al., Marine Biology, April 1999, pages 553-559.)

- How are nutrients generated in the deeper waters of the Red Sea (or in the deeper waters of the ocean)? Why do nutrient concentrations increase in surface waters during spring?

- What explanations can be proposed for the significantly lower chlorophyll concentrations observed for the spring blooms in 1998 and 1999 in the Red Sea? Can support be found for any of these explanations?

Southern Red Sea

- Can additional information be found regarding the influence of the monsoon on the marine biology of the southern Red Sea?

- Does water exchange between the Gulf of Aden and the Red Sea through the Bab al Mandeb, or is the flow exclusively in one direction? If so, in which direction does the water flow? Is the Gulf of Aden a possible nutrient source for the southern Red Sea?

- Do aerosol dust concentrations affect the accuracy of the chlorophyll concentration data in this region?

Mirear Island/Foul Bay

- What other explanations can be proposed for the higher chlorophyll concentrations observed in this area?

- What other data sources could be accessed to determine if winds are related to the higher chlorophyll concentrations here?

- Have there been any scientific studies of this interesting reef complex?