The Effect of Bottom Water Temperature

The temperature of the ocean water overlying the bottom, known as bottom water temperature (BWT), plays an important role in obtaining a reliable heat flow estimate. Understanding how changes to the BWT can affect a computed heat flow value is very important. If the BWT cools or heats suddenly, then this sends a temperature anomaly propagating by conduction downward into the sediments. Sometimes such changes in the BWT are responsible for large, non-linear temperature variations over the length of the 5-m probe, particularly at shallow water sites. TDI-Brooks recommends that heat flow be conducted in depths greater than 600-800m because of the high risk of a BWT variation in shallow water. Normally in the deep ocean the BWT is fairly stable with depth, and varies only slightly in time. This is conducive to highly accurate and precise heat flow measurements. But BWTs may not be stable at any depth depending on the oceanographic factors.

Bottom water temperature transients have typically been identified after the fact, by recognizing a deviation from linearity in the Bullard plot and then explaining the deviation as a probable transient in the temperature of the overlying water. But it is possible to examine the bottom water temperature output of a real-time ocean model and pro-actively identify potential sites where transients could be expected to be a problem. TDI-Brooks has the capability to classify whether a potential heat flow site may be subject to a transient in the bottom water temperature. The method will correctly classify a heat flow site as stable or unstable with an accuracy exceeding 90%. For example, Figure 1 shows the computed probability of a BWT transient offshore of southern Portugal. This region is influenced by the warm and highly saline Mediterranean Outflow Water flowing out of the Gibraltar Strait and to the northwest over the Portuguese margin. The Outflow occurs between 500 and 1400m and affects the BWT of sites typically considered to be deep. Table 1 provides the specific details found by our method. The column labeled Bullard Plot shows if a transient was detected in the Bullard Plot. The column labeled Computed BWT Probability shows the computed probability of a transient. A probability greater than 30% is highly indicative of a transient. None of the site depths were less than 1100 m, yet all but two were affected by a transient. The method correctly identified eleven of the twelve sites, or 91%.

For more information, or a copy of the technical note (click here to download) further describing the method, please contact Dr. Les Bende at+1 979.693.3446 or lesbender@tdi-bi.com.

Figure 10: Probability of a BWT transient offshore of Portugal. The heat flow sites are denoted with a black triangle.

Table 7. Portugal Alentejo Heat Flow