Appendix A. Supplemental methods.
The community dynamics of the dominant mobile epibenthic megafauna were examined at an abyssal site in the NE Pacific as part of a long-term ecological research program. Abundance and body size data collected from 1989–2004 were used to address variation in rank abundance distributions (RADs), Pielou’s evenness, species composition, and interspecific body size vs. abundance relationships. The community descriptors were then related to the habitat’s primary fluctuating resource, particulate organic carbon (POC) flux to the seafloor. Below are supplemental comments on the use of the 50 meter above bottom (mab) trap data to represent abyssal food supplies and the methods used for analysis of community change.
It should be noted that although the 50 mab trap typically has a higher mass flux than the 600 mab trap due to resuspension and/ or lateral transport, the POC flux values have not been shown to be significantly different (Baldwin et al. 1998). This lack of difference could be due to the relatively old and refractory nature of resuspended carbon at the site (Bianchi et al. 1998, Druffel et al. 1998, Smith et al. 2001). A 50-km radius integration was used here to estimate export flux from satellite data. Surface export flux at Station M has been estimated to be similar for areas extending hundreds of kilometers north and south of the site along the California coast (Smith et al. 2006). Figure A1 illustrates the data sets that make up the POC flux composite. The composite used in the study utilized empirical estimations of POC flux (Smith et al. 2006) where no in situ measurements were available.
The method of comparing similarity between samples as a function of the temporal lag between the samples has been used previously (e.g., Collins et al. 2000, Thibault et al. 2004, Venrick 1990). Empirical data and simulations have determined the technique was rigorous over a range of scales (Collins et al. 2000). This method can be used to examine relative rates of change in a community and can detect basic nonlinear trends as well (Fig. A2). Caution should still be used when interpreting such plots, however, since coherent cycles of divergence and convergence can happen at smaller scales than might be resolved. This is clear when the more cyclic MDS x-ordinate of RAD similarity (Fig. A2a) and the time lag plot (Fig. A1a) are compared. Records of sufficient time averaging or with several cyclical shifts present may, for example, show no apparent overall change. The Bray-Curtis similarity utilized here has been shown to be superior to Euclidean distance in estimating the similarity of biological communities (Bloom 1981). The significance of each test was evaluated independently to avoid subjective decisions about what makes up a test or group of tests (Feise 2002). Serial autocorrelation corrections were not used due to the uneven and relatively infrequent timing of the megafauna sampling. The overall conclusions were made within the context of an analysis utilizing multiple community descriptors and a rational assessment based on the results of previous studies at the site.
|
|
| FIG. A1. The composite presented in the main manuscript Fig. 2d is a composite of POC flux estimates from 50 mab (black) and 600 mab (green) sediment traps and model-estimated flux (blue). The red dashed line is unincorporated model data. The difference between the red dashed lined and the black line indicates how well the model estimates correspond to measured 50 mab POC flux values. |
|
| FIG. A2. Several potential outcomes of examining the similarity of temporally different samples are shown in Figure 1a and b. This is a modified version of Fig. 2 that appeared in Collins et al. (2000). |
LITERATURE CITED
Baldwin, R. J., R. C. Glatts, and K. L. Smith, Jr. 1998. Particulate matter fluxes into the benthic boundary layer at a long time-series station in the abyssal NE Pacific: composition and fluxes. Deep-Sea Research II. 45:643–666.
Bianchi, T. S., J. E. Bauer, E. R. M. Druffel, and C. D. Lambert. 1998. Pyrophaeophorbide-a as a tracer of suspended particulate organic matter from the NE Pacific continental margin. Deep-Sea Research II 45:715–731.
Bloom, S. A. 1981. Similarity indices in community studies: potential pitfalls. Marine Ecology Progress Series 5:125–128.
Collins, S. L., F. Micheli, and L. Hartt. 2000. A method to determine rates and patterns of variability in ecological communities. Oikos 91:285–293.
Druffel, E. R. M., S. Griffin, J. E. Bauer, D. M. Wolgast, and X-C. Wang. 1998. Distribution of particulate organic carbon and radiocarbon in the water column from the upper slope to the abyssal NE Pacific Ocean. Deep-Sea Research II 45:667–687.
Feise, R. J. 2002. Do multiple outcome measures require p-value adjustment? BMC Medical Research Methodology 2:8–11.
Smith, K. L., Jr., R. J. Baldwin, H. A. Ruhl, M. Kahru, B. G. Mitchell, and R. S. Kaufmann. 2006. Climate effect on food supply to depths greater than 4000 m in the northeast Pacific. Limnology and Oceanography 51:166–167.
Smith, K. L., Jr., R. S. Kaufmann, R. J. Baldwin, and A. F. Carlucci. 2001. Pelagic-benthic coupling in the abyssal eastern North Pacific: An eight-year time-series study of food supply and demand. Limnology and Oceanography 46:543–556.
Thibault, K. M., E. P. White, and S. K. M. Ernest. 2004. Temporal dynamics in the structure and composition of a desert rodent community. Ecology 85:2649–2655.
Venrick, E. L. 1990. Phytoplankton in an oligotrophic ocean: species structure and interannual variability. Ecology 71:1547–1563.