DADB - Seasonal Temperature, Salinity and Density Anomaly |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
15 depth levels |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Camelia Galos, M.S. Thesis, Dartmouth College, 2000 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Observational Data |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Two basic collections of temperature and salinity measurements, namely "Mediterranean Oceanic Data base" (MODB) and "Adriatic Sea Temperature, Oxygen and Salinity data set" (ATOS2), constitute the foundation for the "Dartmouth Adriatic Data Base", herein DADB.
The MODB comprises temperature and salinity observations for the entire Mediterranean Sea (no coastal data) and it was extracted from http://modb.oce.ulg.ac.be/modb/welcome.html for the Adriatic Sea.
* There are no data between 1917 and 1946 Details on the historical datasets: each surface/subsurface
measurement has at least one valid entry - T or S. For the MODB data types, CTD, BT, and BOTTLE represent conductivity-temperature-depth, bathythermograph, and bottle profiles, respectively. |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
DADBraw Construction |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
The merging of "MODB" and "ATOS2" gave rise to two problems: redundancy and uneven space-time distribution of the data. To solve the former, duplicates were eliminated, and only those observations that were independent in space and time were retained. The seasonal grouping of the monthly files was done according to previous studies of the Adriatic Sea (Zore-Armanda 1969, Gacic at al. 1997): winter -- from January to March, spring -- from April to June, summer -- from July to September, and autumn -- from October to December. This seasonally aggregated set of data shows a more uniform distribution over the domain than the monthly data, both in space and in terms of the number of valid measurements for at least one parameter (T or S): 14349, 13523, 16168, 11726 locations for winter, spring, summer and autumn, respectively. The uneven spatial coverage of the data required care in binning these four seasonal data sets. Most of the observations were made at 0, 5, 10, 20, 30, 50, 75, 100, 150, 200, 250, 300, 400, 500, 800, 1000 and 1200 m depth. The fact that the average depth of northern and middle Adriatic is less than 150 m and that 90% of the measurements are in those top 150 m forced the selection of the most meaningful depth levels for aggregation of the data. These levels are selected such that, after averaging temperature and salinity values with the same initial time/space location within each vertical bin, each interval is centered on the depth with the greatest number of measurements (levels listed above). In this way, the redundancy within each vertical bin is minimized by analyzing both the horizontal and vertical spread of the observations. Fifteen "slices" were cut over the whole depth after the data centered at 250 m were distributed to the adjacent layers, and the data for 1000 and 1200 m were grouped together since they are only in the South Adriatic Pit and the temperature/salinity range is the same. This preprocessing was meant to reflect the structure of the observational data, rather than to rely on some preconceived notions of what the vertical layers should be and how the horizontal gaps should be filled. For more information on the DADB construction and climatological results you can order a copy of this M.S. Thesis from Prof. Benoit Cushman-Roisin at Dartmouth College. |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Optimal Linear Estimation (OAX) |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Climatological fields for each season were estimated using four-dimensional optimal linear interpolation on the DADBraw temperature and salinity fields. The OAX5 package used to obtain the DADB seasonal T and S was developed at the Bedford Institute of Oceanography by Ian he, Ross Hendry and Gerry Boudreau. For a complete reference for the algorithm, see the "Technical Report on Objective Analysis Project" here. |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
DADB - Climatological Data |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
The following table contains the DADB compressed files for each season and depth level. They are named "DADB_season_depth.dat.gz", and the format is as follows:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Climatological Fields - Temperature |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
The figures below present the temperature (T), salinity (S) and density anomaly (σ) fields at selected depth levels, obtained after the application of OAX with the preceding parameter values.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||