160
              
            
            
              
                D. M. King and M. N. Tamburri
              
            
            
              water upon discharge (Option 3c in Table 1 and Figure 2). This could be highly effective
            
            
              at detecting ballast water discharge violations, but would be extremely costly in terms of
            
            
              time and money. For example, current shipboard evaluations of biological efficacy during
            
            
              BWTS testing often involve teams of four to six specially trained technicians, several
            
            
              hours of sample collection time during ballast water discharge, dozens of hours of sample
            
            
              analyses, and often costs over
            
            
              $
            
            
              100,000 per sampling event. Sampling from a single vessel
            
            
              to achieve even 50% or 75% confidence that ballast water being discharged meets or
            
            
              does not meet standards (Options 3a and 3b, respectively, in Figure 2) could cost several
            
            
              hundred thousand dollars. Using available sampling and analytical methods, the cost per
            
            
              vessel to achieve closer to 90% or 100% confidence would be significantly higher. Of
            
            
              course, confidence in the overall verification system depends on the number of vessels
            
            
              sampled in addition to the intensity of sampling from individual vessels. This means that
            
            
              achieving high confidence by intensively sampling individual vessels is likely to make an
            
            
              extensive sampling program prohibitively costly and could result in relatively low overall
            
            
              confidence levels.
            
            
              Between these two extremes on the preliminary cost-effectiveness curve are a range
            
            
              of alternatives that involve applications of indirect measures, such as sensors that monitor
            
            
              conditions within ballast tanks or in ballast piping during uptake and discharge of ballast
            
            
              water to validate that they are (and have been) consistent with proven operational param-
            
            
              eters known to remove or kill planktonic organisms. For example, commercially available
            
            
              industrial or environmental sensors that quantify dissolved oxygen (for BWTS based on
            
            
              deoxygenation) or total chlorine (for BWTS based on electrochlorination) generate data
            
            
              that can serve as reliable proxies to establish with high confidence that a particular treatment
            
            
              system maintained conditions that have been proven to effectively and consistently meet
            
            
              discharge standards. The same level of rigorous and independent performance evaluation
            
            
              and validation required for BWTS certification needs to be applied to sensors that are
            
            
              used in compliance monitoring. However, there are already established programs, such as
            
            
              the Alliance for Coastal Technologies in the United States,
            
            
              27
            
            
              that currently conduct such
            
            
              independent sensor testing.
            
            
              At this preliminary stage of analysis, Figure 2 provides only a basis for posing an op-
            
            
              erational hypothesis that indirect sensor-based measurements are more cost effective than
            
            
              any direct ballast water sampling alternative that is not prohibitively costly. Critically im-
            
            
              portant questions still need to be addressed regarding how emerging technologies involving
            
            
              sensors and sampling might change the relative position of alternatives in Figure 2. From a
            
            
              cost-effectiveness and regulatory impact perspective, there are equally important questions
            
            
              that need to be addressed soon regarding whether spending to increase the volume and sta-
            
            
              tistical accuracy of direct ballast water sampling will result in more or less favorable shifts
            
            
              in the cost-effectiveness curve than similar amounts of spending to improve the precision
            
            
              and reliability of indirect measures of ballast water discharge using sensors.
            
            
              
                Results
              
            
            
              
                
                  Benefits of Noncompliance
                
              
            
            
              Since most vessels planning to visit U.S. ports will install a certified BWTS, the main
            
            
              compliance issue will involve whether ship operators are using them and, if so, properly
            
            
              operating and maintaining them.
            
            
              28
            
            
              Preliminary economic research has indicated that the cost
            
            
              of purchasing and installing a typical BWTS is in the range of
            
            
              $
            
            
              600,000 to
            
            
              $
            
            
              1.2 million, and
            
            
              the annual cost of maintaining and operating it ranges from
            
            
              $
            
            
              15,000 to
            
            
              $
            
            
              125,000.
            
            
              29
            
            
              A rough
            
            
              Downloaded By: [Cantrell, Joyce] At: 17:14 26 July 2010