Ecosystem Status Report for the Northeast Large Marine Ecosystem
Examination of the individual indicators provided in this report permits a detailed view of trajectories of change in climatic, oceanographic, ecological, and social and economic subsystems of the Northeast U.S. Continental Shelf over the last several decades. However, to identify emergent patterns in this broad suite of indicators, we require ways of synthesizing information from each to discern evidence of systemic change. Construction of an overall index from a collection of indicators is one strategy commonly employed to integrate information from a potentially large number of individual elements. Familiar examples include the well-known Dow-Jones Industrial Average constructed by summing stock prices of 30 major corporations and dividing by an adjustment factor.
We constructed a composite index of many of the variables provided in our Ecosystem Status Report to provide an overall NES Ecosystem Index. We restricted our analysis to variables with a continuous record of observation from 1977 to the present (a number of important ecosystem indicators derived from the NEFSC MARMAP and EcoMon surveys were not available prior to 1977). We used a common multivariate statistical procedure, principal components analysis, to construct the index. This technique involves the construction of a set of mutually independent linear combinations of the original variables. This method has been frequently used as a data reduction technique in analyses of ecological indicators in the NES. This method however does not explicitly consider the time-order of the observations and it cannot be used with variables containing missing values. To directly address the time series nature of the observations and to test for evidence of change-points throughout time period examined, we employed another multivariate approach, chronological cluster analysis which seeks to identify contiguous blocks of time with similar characteristics and points at which, statistically significant change occurs.
This analysis reveals some important changes in the Northeast Continental Shelf over the last several decades. Our first composite index, shown in blue bars in Figure 12.1 suggests that the overall characteristics of the system changed in the late 1980s and remained in a different state until the early 2000s. These change points are consistent with ones shown in other analyses focusing on parts of the ecosystem in more restricted geographical areas, principally in the Gulf of Maine. Our system-wide analysis shows that these changes are much broader in scope and more pervasive throughout the system. The second composite index, shown in the red line, reveals an additional component of change indicating a trend related to factors such as steadily increasing temperatures over the last several decades. Understanding that these ecological changes occur and persist on decadal time scales is important in understanding changes we see in our fisheries over time. In turn, this understanding can help inform management decisions if we detect persistent increases or decreases in productivity over time.
Our NES Ecosystem Index focuses on integrating climate, physical, and ecological indicators. The fishery subsystem responds not only to drivers and pressures related to climate, physical, and ecosystem changes but to management interventions and other factors. Accordingly we conducted a separate analysis on landings by major species groupings. The species categories included (1) principal groundfish, (2) flatfish, (3) pelagic fish, (4) elasmobranchs (dogfish and skates) (5) other fish, (6) molluscs, and (7) crustaceans. In this case, we could extend the time series back to 1964 to provide a longer term perspective on change in the fishery. We again constructed composite indices using principal components analysis as a way to summarize the information in the landings series and to tests for patterns in the data.
The first two composite scores derived from the principal components analysis are shown in Figure 12.2. In this case, the first composite score is shown in the blue line. It indicates a positive overall trend over time. The first composite score reflects increases in landings of crustaceans and molluscs in particular and, to a lesser extent, increases in elasmobranch landings. Interestingly, the second composite score (shown in red bars) shows a more periodic pattern. Visual inspection of this second index shows break points in the same locations as the shorter ecosystem index (1989 and 2002) as well as two additional changes in the earlier part of the series (1968 and 1976). However the more formal test for change points in the series using chronological cluster analysis reveals a more complex picture. In this case, the analysis involves not only the information contained in the first and second composite scores but the complete series of landings data. The vertical lines in Figure 12.2 show the change points revealed by the chronological cluster analysis. A sequence of changes occurred in 1968, 1976, 1982, 1989, 1994 and 2001. The first change point corresponds to an escalation of fishing by the distant water fleet after an initial developmental phase. The second point in 1976 can be connected to the phasing out of distant water fleet operations as extended jurisdiction under the Magnuson-Stevens Act was about to be implemented. The change point in 1982 corresponds to the switch from the quota-based NEFMC groundfish management system in place from the inception of extended jurisdiction to more qualitative management measures based on mesh size and other regulations that remained in place through 1993. In 1994, changes in the fishery reporting system and further modifications to groundfish management regulations were implemented. The NEFMC groundfish management system changed to one incorporating a phased implementation of reductions in days-at-sea (which continued to increase until 2001 and then declined). The indicated change in 1989 may be related to further management changes implemented under the Multispecies Groundfish Management plan, and possibly, changes in gear technology associated with the development of rock-hopper gear. The analysis did not detect the change related to the groundfish rebuilding plan initiatives starting in 2004 or the sequence of further reductions in days-at-sea for groundfish vessels in 2006 and 2008. Because the data series incorporated in this analysis extended only until 2012, the important change to sector management and catch shares implemented in 2010 was not detected. Because the landings series is not limited to groundfish and includes important fishery components from species managed by the Mid-Atlantic Fishery Management Council (MAFMC) as well as NEFMC, the apparent signature of NEFMC groundfish management actions in the overall landings series is extremely interesting.
Considering the indications of system-wide change from the NES Ecosystem Index and the NES Fishery Index and the change points observed in the fish condition and recruitment indices (see Figures 6.7 and 6.8), which were not incorporated in the composite index values, a convincing case can be made for the occurrence of ecosystem changes occurring on decadal time scales on the Northeast Continental Shelf. These changes, in concert with management interventions appear to collectively exert strong influence on fishery performance.