The point of this page is to share information on the construction of an Ocean Bottom Properties (OBT) database. By far, temperature is the most frequently recorded variable at ocean depth - there are many different ways to record temperature (bolltes, CTDs, XBTs, reversing thermometers, etc.). Second is salinity, measured in bottle collections or via CTD. The other measured variable in this database are dissolved oxygen, nitrate, nitrite, inorganic phosphate, ph, and silicate. The impetus for this work is the complete lack of a database equivalent to the COADS (Comprehensive Ocean Atmosphere Data Set) for the ocean bottom. I wanted such a database for my work in modeling historical changes in halibut (a demersal groundfish) size at age. Seems to me that scientists who work with other species of groundfish might also appreciate historical data on oceanographic properties of the water in which these fish reside.
I am NOT contructing a database for the world's oceans, only the Alaskan and West Coast continental shelf. The database has the following boundary conditions
All data - all kinds, all locations, all times
NODC bottle data from the kind folks out at the National Oceanographic Data Center. It took a long time to get this data, but it was worth it.
NODC mbt data from the same folks. MBT = mechanical bathythermograph, an old fashioned way to get temperature at depth
NODC xbt data from the same folks. XBT = expendable bathythermograph, a new fashioned way to get temperature at depth
NODC ctd data from the same folks. CTD = conductivity, temperature, depth. These data are collected by instruments that measure temperature by thermistor, salinity by conductivity, and depth by pressure.
Ocean bottom temperature data from NOAA/NMFS/RACE research surveys
Ocean bottom temperature data from International Pacific Halibut Commission (chez moi!)
Ocean
bottom temperature data from Russian bottom trawl surveys
Using all the data sources, here are temperature data locations by decade
(jpg files)
1930s
| 1940s
| 1950s
| 1960s
| 1970s
| 1980s
| 1990s
Here are more detailed plots of data by month from 1950-1997 (ps files)
1950-52
| 1953-55
| 1956-58
| 1959-61
| 1962-64
| 1965-67
| 1968-70
| 1971-73
1974-76
| 1977-79
| 1980-82
| 1983-85
| 1986-88
| 1989-91
| 1992-94
| 1995-97
Many different analyses can be conducted. Here a few so far.
First, some average temperature plots.
Monthly
averages for the 1950-96 period (ps file)
Long-term
average for 1950-96, averaged across months May-Sept.
.
Now, some multiple plots using differencing to identify
temporal changes in bottom temperatures. The following graphics each
have three panels. The bottom panel show the average bottom temperature
for the five year period (averaged across months May-Sept.). the
middle panel shows the change in bottom temperature from the previous pentad
(i.e. a difference map where the early pentad tempature at each location
is subtracted from the later pentad temperature). Positive temepratures
indicate a warming from one pentad to the next. The third panel shows
the difference between the pentad bottom temperatures and the long term
mean (1950-96, May-Oct.). Positive temperatures indicate pentad temperatures
are warmer than the long term mean.
1960-64
| 1965-69
| 1970-74
| 1975-79
| 1980-84
| 1985-89
| 1990-94
| 1995-99
To begin to quantify interannual/decadal variability I
made some annual areal averages. Initially, I selected five regions
with the coordinates listed below. For this attempt, I averaged all
observation for months May-September within the region to obtain an average
for each year. If there were less than 10 observations for a given
year, no average value was computed. The linked graphics are jpg
files.
NW
Bering Shelf: lat:60-64, lon: 184-192
SE
Bering shelf: lat: 56-58, lon: 186-194
Western
Gulf: lat: 56-60, lat: 204-212
Eastern
Gulf: lat: 54-60 long: 220-228
British
Columbia: lat: 48-54, lon: 228-234