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 setting for this tutorial is the north Atlantic Ocean, specifically the area between the Sargasso Sea and (roughly) the latitude of southern Greenland and Iceland. The southern latitude is approximately 35° N, and the northern latitude is approximately 60° N. During the months of spring in the Northern Hemisphere, this region hosts the annual occurrence of the North Atlantic Bloom -- one of the largest seasonal changes in phytoplankton productivity occurring anywhere in the world.
As an introduction to the North Atlantic Bloom, you may wish to read these Web articles:
The North Atlantic Bloom (Science Focus!)
LOCUS Educational Module 5 -- Seasonal Variability
The North Atlantic Bloom is the primarily the result of a combination of increasing sunlight and higher concentrations of nutrients. Nutrient concentrations increase in surface waters over the winter, when phytoplankton growth rates are slow and phytoplankton populations are low. The rapidly increasing concentrations of phytoplankton that signify the bloom are first noted near the southern boundary of the bloom zone in March, and progress northward to the far northern Atlantic in subsequent months.
2. Primary Research Question
A wide variety of questions could and can be asked about the North Atlantic Bloom. For this tutorial, however, we will be examining the North Atlantic Bloom with a new capability in Giovanni - climatological anomaly analysis. Climatological anomaly analysis is a method of examining climate data of many types, with the intention of discovering events or trends that are "out of the ordinary". In scientific terminology, "out of the ordinary" means a significant deviation from the normal (average) state or the mean data value. For remote sensing data, two of the most common applications of anomaly analysis are for sea surface temperature (SST) and sea surface height (SSH) data.
For the North Atlantic Bloom, our primary research question will be:
What are the major patterns of chlorophyll concentration anomalies in the North Atlantic Bloom?
3. Investigation Plan
For this investigation, we will be utilizing the new Giovanni climatological anomaly analysis capability. We will examine the North Atlantic Bloom period (springtime in the Northern Hemisphere) to determine the appearance of chlorophyll concentration anomalies, and we will attempt to determine the nature and cause of the anomalies for a selected time period. Initially, our study area will encompass most of the north Atlantic Ocean, lying between longitudes 10° W and 60° W, and between latitudes 36° and 58° N. After identifying interesting areas to examine, we will reduce the size and scope of the study area.
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. Gionvanni 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 data for the selected area, or Hovmoller plots for the area.
The fundamental requirement for climatological anomaly analysis is a data record that is long enough (and consistent enough) to allow the generation of the climatological mean values that can be compared to a selected period of time. The eight-year long ocean color data record from SeaWiFS now allows such comparisons to be made. The Giovanni climatological anomaly analysis creates data averaged over a selected time period from one month to one year long, and compares the values for that time period to the SeaWiFS climatological mean data values. If the data is greater than the mean value, it generates a positive anomaly. If the data is less than the mean value, it generates a negative anomaly.
To illustrate the anomaly analysis before looking at the North Atlantic Bloom, the image below shows a classic case: the marked reduction of chlorophyll concentration, corresponding to a significant reduction in phytoplankton productivity, in the Peru Upwelling Zone during the famous 1997-1998 El Niño event. The deep purple and dark blue indicate the strong negative chlorophyll concentration anomaly.
5. Preliminary Analysis
For the preliminary analysis, the time period for examination is April-July 2000 and April-July 2001. The spatial coordinates are as listed above: longitudes 10° - 60° W,latitudes 36° -58° N. These selections are entered into the Giovanni interface for SeaWiFS data, the parameter chosen is "chlorophyll a concentration", but rather than choosing "Parameter" in the "Data Type" menu, "Anomaly" is selected, along with the "Area Plot" option. This process generates two chlorophyll concentration anomaly plots, which are shown below. These plots were labeled to show the two areas that will be examined in more detail.
North Atlantic Bloom, Spring 2000
North Atlantic Bloom, Spring 2001
Although there are other patterns of interest (and over the eight-year SeaWiFS mission, there are other interesting patterns in other years), the "North Area" and "South Area" shown above displayed significant changes in the chlorophyll concentration anomaly. Now let's take a closer look at the chlorophyll concentration anomalies in these areas during the North Atlantic Bloom during 2000 and 2001.
In the North Area, there is a strong positive anomaly in 2000, and a somewhat mixed result in 2001, with areas of postive and negative anomaly. In the South Area, there is a very strong negative anomaly in 2000, and a similarly strong positive anomaly in 2001.
6. Refinement of Analysis
Now that we have visualized the chlorophyll concentration anomalies for the selected areas in 2000 and 2001, we can use time plots of the average chlorophyll concentrations to see the seasonal patterns for each month. First, the time-series from 1998 to 2004 iin each area will be generated, and then the time-series for just the years 2000 and 2001.
North Area time-series plot, January 1998 to December 2004
North Area time-series plot, January 2000 to December 2001
For the North Area, the peak chlorophyll concentration values are fairly similar from year-to-year. However the most significant observation here is that the highest chlorophyll concentrations in 2000 occurred in June, while in 2001, the highest chlorophyll concentrations occurred in May. Note that data is frequently missing here in December due to the low sun angle in the winter, which is the reason for the gap in the second time-series plot.
South Area time-series plot, January 1998 to December 2004
South Area time-series plot, January 2000 to December 2001
For the South Area, the peak chlorophyll concentration in 2001 is about twice the value in 2000, and the maximum chlorophyll concentrations for both years occur in the month of April. In the upper time-series plot for 1998-2004, it can be seen that 2000 was the minimum average chlorophyll concentration, and 2001 was the maximum average chlorophyll concentration, for this area. This observation explains why the chlorophyll concentration anomaly was strongly negative in 2000 and strongly positive in 2001.
Now we'll take one more look at these areas and chlorophyll concentrations using Hovmöller plots. Note that a customized color palette was used for these images.
North Area, Hovmöller plot, 2000-2001
It's a little hard to discern the difference in the timing of the peak concentrations in this plot, but if we put the years 2000 and 2001 side-by-side, it's very clear.
These Hovmoller plots clear show that the bloom in 2001 was reduced compared to 2000, and that the peak concentrations occurred in a smaller area a month earlier.
South Area, Hovmöller plot, 2000-2001
For the South Area, we don't have to compare the two years side-by-side; it's clear that the concentrations in 2001 were much higher than in 2000.
7. Statement of Results
In this investigation of chlorophyll concentration anomalies in the North Atlantic Bloom, we have determined, for two selected areas located to the east of Newfoundland (the "North Area") and southeast of Newfoundland (the "South Area"):
- in the North Area, the difference between the strong positive anomaly in 2000 and the mixed pattern of anomalies in 2001 is due to both a shift in the timing of the peak chlorophyll concentrationand the magnitude of the peak chlorophyll concentration values;
- in the South Area, the negative anomaly in 2000 and the positive anomaly in 2001 are due to significant differences in the magnitude of the peak chlorophyll concentration values; for the period 1998 to 2004, 2000 represents the minimum peak concentration value and 2001 represents the maximum peak concentration value.
8. Discussion of Results
This investigation of climatological anomaly analysis with chlorophyll concentrations in the North Atlantic Bloom suggests that significant new insight can be gained with this tool, particularly if it is combined with other types of remote sensing and in situ data. The North Atlantic Bloom is a very complex phenomenon, dependent on several different factors, which include available light levels, cloud cover, nutrient concentrations, current interactions, and water temperatures. An examination of one type of data can indicate interesting aspects of the North Atlantic Bloom, but a full investigation would require a broader scope to the research.
Nonetheless, the results shown here suggest the types of insight that can be gained with further research. The two areas selected for the years 2000 and 2001 showed wide differences in the chlorophyll concentration anomaly between the two years. The proximity of the two areas suggests a possible interconnection between the intensity of phytoplankton production during the North Atlantic Bloom.
In 2000, the South Area displayed a strong negative anomaly, and the chlorophyll concentration values were the minimum observed for the bloom between 1998 and 2004. By contrast, in the North Area in 2000 displayed a strong positive anomaly, and the chlorophyll concentration values were the maximum observed for the period 1998-2004, though this maxima was not as strong as the comparable minima in the South Area.
In 2001, the South Area displayed a strong positive anomaly, and the chlorophyll concentration values were the maximum observed for the bloom in 1998-2004. In the North Area, the peak chlorophyll concentrations occurred one month earlier than in 2000, and both the spatial extent of the maximum concentrations and the average chlorophyll concentration values over the whole area were reduced. For these two years, the chlorophyll concentration anomalies in the two areas were somewhat anti-correlated; i.e., when one anomaly was positive, the other was negative, and vice versa.
This observation is of interest because of the characteristics of current flow in this region of the north Atlantic Ocean. Although Giovanni does not currently have MODIS SST data for 2000 and 2001, it does have SST data for later years. The plot below shows the SST for April-July 2003, with the outlines of the North Area and South Area boxes.
This image shows that both areas encompass oceanic regions where the Gulf Stream and its northern branch, the North Atlantic Current (warmer water to the south) interact with the colder Labrador Current. The Gulf Stream/NAC flow east and northeast, while the Labrador Current flows southeast past Newfoundland and "dives" southward past the island. Areas of current interaction such as these are highly variable, and the convergence zones between such current systems frequently foster enhanced phytoplankton productivity.
One other view of the system, using the Giovanni dataset intercomparison capability, shows both the chlorophyll concentrations (in color) and the SST data (as contour lines). The image below shows that the areas where SST change significantly, which roughly correspond to the locations of the current boundaries, have elevated chlorophyll concentrations. This image is for April-July 2003, the same period as for the SST image shown above.
Because the North Area and the South Area of this investigation are adjacent, the observations suggest that the intensity of the bloom in the South Area, which precedes the bloom in the North Area, could affect the intensity and timing of the bloom in the North Area. If the Gulf Stream/NAC flow supplies nutrients to the North Area during the period of the North Atlantic bloom, a strong bloom in the South Area could reduce the concentration of nutrients flowing to the North Area, which would curtail the bloom in this area. Similarly, a weak bloom in the South Area might mean that the nutrient "flow" to the North Area was increased, leading to a stronger and longer bloom. While this hypothesis is consistent with the observations, considerably more work would be required to determine if it is an actual explanation for the observations.
9. Statement of Conclusions
This investigation indicated that climatological anomaly analysis of chlorophyll concentrations in the North Atlantic Bloom demonstrates the cause of two different kinds of chlorophyll anomaly, one which is primarily due only to differences in chlorophyll concentration, and the other which is due primarily to differences in the timing of the peak chlorophyll concentration. The study areas selected displayed significant differences between the years 2000 and 2001 in both the average value of peak chlorophyll concentrations occurring during the North Atlantic Bloom and the timing of the occurrence of the peak chlorophyll concentrations. The investigation also indicated the possibility that bloom timing and magnitude in the two selected areas is inter-related, with the bloom intensity in the southern area potentially affecting both the timing and intensity of the bloom in the northern area. This relationship is enabled by the flow of the major current systems in the region, the Gulf Stream, North Atlantic Current, and the Labrador Current.
10. Questions for Further Investigation
Clearly, this tutorial investigation suggests several different ideas for further research. The four questions provided below are only examples of topics that could be examined.
- What are the sea surface temperature characteristics of the study region in 2000 and 2001? Do they support the suggested inter-relationship between the timing and magnitude of the bloom in the two selected study areas?
- Do chlorophyll concentration anomalies "migrate" in a given year as the bloom region progresses northward? Can such progress be tracked in the data?
- Can cloud cover data be correlated with the intensity of the bloom in a selected region of the north Atlantic? (Phytoplankton need light for photosynthesis; persistent cloud cover over a particular region might suppress productivity in that region.)
- Is there a relationship between the position of the zones where the major currents interact and the concentration of nutrients in a particular area?