All of the questions below are linked to descriptive answers in this chapter of the Giovanni-3 Online Users Manual regarding the Giovanni-3 Oceans Monthly instance. Click the question of interest.
- How does the SeaWiFS climatology and anomaly function work?
- Why are there two SeaWiFS data sets: SeaWiFS.R5.2 and SeaWiFS.R2009?
What are MODIS-Terra, MODIS-Aqua, and SeaWiFS?
MODIS stands for the Moderate Resolution Imaging Spectroradiometer. There are two MODIS instruments in orbit; MODIS-Terra is on the Terra satellite, which observes the Earth (in daytime) during the morning. This should allow better surface observations due to reduced cloud cover. MODIS-Aqua is on the Aqua satellite, which observes the Earth (during the daytime) in the early afternoon. This timing allows for improved observations of clouds and other atmospheric variables, which other instruments on Aqua are intended to measure. SeaWiFS stands for the Sea-viewing Wide Field-of-view Sensor, an instrument which primarily observes the oceans, but also provides data and imagery from land. The two MODIS instruments have a large number of spectral bands allowing measurements of land, ocean, and atmospheric variables; SeaWiFS has only eight bands used for ocean observation and atmospheric correction.
What are the current versions of MODIS-Aqua and SeaWiFS data available in the Giovanni-3 Oceans Monthly instance?
The current version of SeaWiFS data is Reprocessing 2009 (SeaWiFS.R2009). We have also kept the previous version of SeaWiFS data, Reprocessing 5.2, (SeaWiFS.5.2) for comparison purposes, which was completed in July 2007. Reprocessing 2009 replaced normalized water-leaving radiances (nLw) with remote-sensing reflectances (Rrs), so this allows a way to see how the values are related. Because Reprocessing 2009 also made calibration adjustments and other data quality changes, the values of nLw and Rrs will not correspond in all places, i.e., using Rrs from the SeaWiFS.R2009 data set to calculate nLw will not necessarily produce the values of nLw in the SeaWiFs.R.5.2 (or vice versa, using nLw in SeaWiFS.R.5.2 to calculate Rrs will not necesarily produce the values of Rrs in SeaWiFS.R2009).
The current version of MODIS-Aqua data is Reprocessing 1.1, which was completed in August 2005.
What MODIS-Aqua and SeaWiFS data types are available in this instance? What are GSM data?
All MODIS-Aqua and SeaWiFS data in Giovanni-3 are Level 3 monthly data at 9 km resolution, to allow intercomparison. MODIS-Aqua Level 3 data is also available at 4km resolution from the Ocean Biology Processing Group (OBPG). Level 3 data are global mapped data products; the monthly data has been binned (collected and averaged) spatially and temporally over an entire month to provide mean monthly data values. GSM data are ocean optical products derived from SeaWiFS and MODIS-Aqua data, and sometimes both, as for the "Merged" data products. GSM stands for Garver-Siegel-Maritorena, the last names of the researchers who have provided the data products.
What MODIS-Aqua, SeaWiFS, and GSM data parameters are available for analysis?
The currently available data parameters are listed below. Because MODIS-Aqua have different band locations, the parameter definition is slightly different. The MODIS-Aqua values, when different, are shown in brackets.
How are data parameters selected for analysis?
Individual parameters are selected by clicking in the check boxes. An entire group may be selected by clicking the group box (leftmost column heading bar, next to the collection name and date range information). If this selection is performed, individual parameters may then be deselected by clicking in the appropriate check box(es). Note that at least two parameters must be selected for multi-parameter services such as Comparison Plot.
What are the current data analysis and visualization capabilities in the Oceans Monthly instance?
There are currently 11 data analysis and visualization options for the Oceans Monthly data.
How is the spatial area of interest selected?
The spatial area of interest can be selected by either of the two methods described below
Java applet map: The Java applet map can be utilized to "click and drag" a selection box to designate the spatial area of interest.
Geographic coordinates: The geographic coordinates of the corners of the spatial area box can be input individually. This method can be used if Java is not enabled. Note that decimal and not degree (seconds and minutes) coordinates must be entered.
How is the time range of interest selected?
The starting date (year/month/day), and the ending date (year/month/day) for the analysis are selected using the calendrical menus on the Giovanni interface page. Error messages will be generated if the beginning month or the end month are out of the data set time range, or if the end month is before the beginning month. Note that the datasets have different time ranges, and this can affect the output for multi-parameter services.
Are there any limits on the spatial area or time range than can be selected?
Longer time ranges and larger area selections will take considerably more time for initial data acquiisition from the archive (fetching) and a longer period of elapsed time for processing. Note that the Ocean data is currently the highest spatial resolution (9 km) of any data in the Giovanni system. This higher resolution may require somewhat longer time periods for computations than for coarser resolution data.
If time ranges are selected that are not concurrent for all the datasets used in a multi-parameter service, an error message may be generated. In some cases, gaps will occur in the data output, such as for the Time-Series services. In other cases, the service will not be able to generate the desired output.
What are the plotting preference options for the data visualizations in the Oceans Monthly instance?
The plotting preference options are described in the Plot Preferences section of the Online Users Manual.
After the initial output has been generated, the temporal and spatial ranges (constraints) and plotting preferences can be changed, and a new output generated, using the "Submit Refinements" button.
How do I select data parameters for multi-parameter visualization options?
Select at least two parameters. Multi-parameter maps can only accept two parameters; time-series can take more. Then generate the visualization. If you choose data from two different data sets, the map functions will only work if the data is in a time period with data from both data sets. If there is missing data for time-series, there will appear to be gaps in the time-series plot.
To display multiple time-series plots with shared Y-axes and different colors for the data and data axis labels, first generate the time-series of each parameter by selecting the "Time-Series" visualization option. This willgenerate a separate time-series for each parameter. Then click "Yes" for the Overlay Flag option, and "Submit Refinements". If the units are the same, the Y-axis will be adjusted to display both time-series on the same plot. If the units are different, then separate colors and axes will be displayed corresponding to each parameter.
What are the data output (download) options that I can select?
The current options for data download are HDF (HDF4), netCDF, and ASCII (text). Some map visualizations also include the KMZ file format option for Google Earth display. At present, KMZ files are limited to single parameter maps only and the default color palette only.
Why isn't there any chlorophyll data from MODIS-Terra?
MODIS-Terra experienced significant difficulties in establishing and maintaining sufficient radiometric accuracy and consistent caibration. While most of the other data types from MODIS-Terra are deemed scientifically accurate, the stringent requirements of ocean color radiometry prevented the generation scientifically valid ocean color data products. The OBPG is still working with this data in an effort to produce a useful data set.
Why are there three sea surface temperature data products? Which one is the best one to use?
The standard SST measurements were made at 11 and 12 µm. This is consistent with SST measurements from the Advanced Very High Resolution Radiometer (AVHRR). However, there is significant water vapor absorption at these wavelengths, so that over humid regions, such as the tropics, SST atmospheric correction is less certain. The 4 µm band is less sensitive to water vapor.
The Sun heats up the surface layer of the ocean during the day. Therefore, SST observed during the day may be somewhat warmer than "actual" SST in the top meter of the ocean, which is the traditional definition of SST. This diurnal variation will be more prominent where solar radiation is stronger.
It is difficult to state which product is the best to use. The 11 µm products can be directly compared to AVHRR SST data. However, the 4 µm product may be slightly more accurate, especially for tropical waters.
The document linked below provides additional information.
Sea Surface Temperature Measurements of the MODIS and AIRS Instruments Onboard of Aqua Satellite
Why are there so many atmospheric data products? I thought this was data for the oceans.
The complex atmospheric correction process for ocean color radiometric data requires the calculation of several atmospheric components, particularly aerosols. The Oceans data set therefore includes these atmospheric parameters.
How does the SeaWiFS climatology and anomaly function work?
The SeaWiFS monthly climatology data used in the Ocean Month Giovanni instance are products of the Ocean Biology Processing Group (OBPG), downloaded from the ocean color ftp site. The current climatology data are derived by time-binning the monthly binned files from September 1997 through August 2007, mapped to an Equidistant Cylindrical projection, with 9km resolution at the equator, similar to monthly Standard Mapped Image (SMI) products. For example, the January Climatology data are time-binned for each January from 1998 to 2007. Weighting of 1/sqrt(n) is applied for time-binning, where n is the number of valid pixels within a binning area. The same time binning software used for generating 8-day or monthly products is applied to climatologies ( SeaWiFS Prelaunch TM 32, appendix B). For these climatologies, each month is derived from 9 years of monthly bins. Because the SeaWiFS sensor was inactive for much of the first part of 2008, the base period has not been extended to include 2008 (12/01/2008).
The SeaWiFS monthly anomaly output is generated simply by determining the difference between each data point in the monthly SMI data product used in Giovanni (such as chlorophyll a)and the monthly climatology SMI for that product.
Why are there two SeaWiFS data sets: SeaWiFS.R5.2 and SeaWiFS.R2009?
One of the main changes between SeaWiFS.R.5.2 and SeaWiFS.R2009 was the change from normalized water-leaving radiance (nLw) values for the 6 observational SeaWiFS bands (bands 7 and 8 are used for atmospheric correction) to values of remote sensing reflectance (Rrs). We have kept the previous data set to allow comparison of these values. Also, all of the changes described for Reprocessing 2009 were performed to improve the values of the geophysical products, with chlorophyll a concentration being one of the most familiar and recognized. Because users may wish to examine how much the chlorophyll a product values have changed between the two data sets, we have kept the SeaWiFS.R5.2 data set to allow this comparison. The most accurate chlorophyll a data from SeaWiFS is now be the SeaWiFS.R2009 data, and it is these data that should be used for research and applications. For example, Giovanni now allows plotting of comparative time-series or latitude-longitude difference maps for the two SeaWiFS data sets, which may have instructional and research value.