Project: Evaluate the impact of harvesting precious corals on resident deepwater fish assemblages and the monk seal forage base.

Principal Investigator: Frank Parrish

Summary

The primary objective of this work was to evaluate the ecological effect of commercial harvesting of precious corals on deep water fish assemblages that comprise a portion of the endangered Hawaiian monk seals forage base. The intent was to survey sites where commercial coral harvesting had expanded in 2001 and compare the fish and coral data with baseline information collected at those sites in 2000 using a before-after analysis. However, the harvesters never expanded out of the Makapuu coral bed as they had planned and because of economic concerns suspended their operations indefinitely. Consequently results reported are limited to precious coral stock assessments and general observations on the associated deepwater fish community.

Background

The link between seals and precious corals was established in 1998 using submersibles to inspect sites at Brooks Bank and French Frigate Shoals (FFS) where endangered Hawaiian monk seals fitted with satellite tags focused their foraging. The hypothesis was that the cluster of satellite positions were centered on a specific, habitat type the seals were targeting. Surveys using the Hawaii Undersea Research Laboratory's Pisces V revealed that seals were visiting beds of precious corals and that this was a minority habitat type over the wider region. Results of the HURL surveys were recently published in the January 2002 issue of Marine Mammal Science (Parrish et al. 2002).

The reemergence of the Hawaiian precious coral fishery from decades of dormancy prompted additional HURL surveys in 2000 to collect baseline information at all known Hawaiian precious coral beds (Fig. 1) before the fishery expanded and a baseline reference point lost. Coral and fish data were collected at FFS, WestPac Bank, Kaena Point, Cross Seamount, and Keahole Point. The Brooks Bank survey was lost to mechanical problems, and poor weather prevented the survey for Makapuu Point where commercial harvesting of coral was already in progress. Besides Makapuu, some harvesting had already been conducted at Keahole before the HURL 2000 survey. On all dives the number and size of deepwater fish and corals were recorded, and temperature monitors were deployed at FFS and Cross Seamount.

Figure 1. All known Hawaiian precious coral beds.

2001 Surveys

In 2001 dives were awarded to resurvey sites commercially harvested since the 2000 HURL surveys. However, commercial harvesting remained focused at the Makapuu bed where weather had prevented the HURL survey of the site in 2000. Sites surveyed in 2001 include Brooks, FFS, Keahole point, Cross Seamount, and Makapuu Point. At all sites submarine transects covered depths between 350 and 500 m.

Because coral harvesters suspended their operations without expanding to areas where baseline data had been collected the intended comparison of deepwater fish communities in recently harvested coral beds to their pre-harvest status (baseline data) was not possible. Observations on the standing stock of precious coral beds and associated fish assemblages are discussed in the sections below.

Precious coral stock

The bulk of coral harvesting has always been focused on pink corals (Corallium secundum) at the Makapuu bed, and the 2001 surveys of the site found some evidence of the recent coral harvesting such as stumps and broken fragments of coral colonies. Co-investigator, Dr. Richard Grigg, collected coral size data on these surveys and incorporated the data with prior surveys, creating an age structure of the coral stock dating back 30 years to pre-industry days in the early 1970s (Fig. 2). The resurvey conducted in 2001 demonstrates several important points: first, the lack of significant impact to the coral stock associated with recent harvesting in 1999 and 2000 and second, a continuing pattern of coral recovery since 1983. A "movement" of the strong 10-15 year class in 1983 to older year class groups in 1985 and 2001 is evident. This is consistent with current estimates of pink coral growth rate of 1 cm/yr increase in height. It is also possible to see the consistent recovery in 2001 of year class groups from 20-25, 25-30, 30-35, 35-40, and 40-45 years.

Figure 2. An age structure of the coral stock dating back 30 years to pre-industry days in the early 1970s created by Dr. Richard Grigg, who collected coral size data on these surveys and incorporated the data with prior surveys.

Some harvesting was conducted at the Keahole bed to collect red (Corallium regale) and gold (Gerardia sp.) coral. The size/age structure of the data collected at Keahole Pt. in 2001, when compared to earlier data collected by Grigg, confirms current estimates of MSY and showed no impact of recent harvesting on red or gold coral stocks. Finally, the spatial extent of coral assemblages at these and the other locations was better defined in the 2001 surveys.

Deep water fish assemblages

Fish counts were lumped by 5-min survey intervals as the submarine/ROV proceeded over the bottom. The length of each fish observed was estimated to within 5 cm generating a database of more than 13,000 individuals. Depth and substrate information were collected for each survey interval. Since changes in fish assemblages associated with harvesting cannot be addressed, the data collected will be best used to examine differences in fish assemblages by latitude, substrate type, and proximity to monk seal colonies. Preliminary analysis of the data indicates standardized fish densities across all sites varied to as much as five times the minimum fish densities observed. Length data will be used to generate fish biomass values and provide a more suitable basis with which to compare the size of fish assemblages across sites and habitats.

A total of 47 species were encountered, including one that is a new record for the Hawaiian Archipelago and may be a new species (Fig. 3). Dividing the fish species into behavioral functional groups that may reflect differences in the prey types targeted by foraging seals reveals that nearly half of all fish seen were bottom hoverers (Fig. 4). Together with benthic searchers and ambush predators that stay on the bottom, the benthic fraction makes up 91 percent of the fish assemblage, leaving a small percentage of demersal fish types. Future analysis will address the degree of association of fish with coral colonies. A few taxa in the midwater hoverer group such as Antigonia sp. and Hollardia goslinei clearly associated with structural relief, including deepwater corals. Other taxa including eels, morids and Polymixia sp. are less clearly associated and will require an analysis using substrate type as a covariate.

Figure 3. Possible new fish species.

Figure 4. Fish species divided into behavioral functional groups that may reflect differences in the prey types targeted by foraging seals.

Ancillary survey work

At each site bathymetry data were collected using HURL's Seabeam mapper on a time-available basis. Sizable swaths of bottom were mapped at FFS, Brooks Bank, and Cross Seamount. The survey of Cross Seamount revealed a depression larger than a square kilometer and 100 m deep on the summit with sheer cliff walls to the north and east, which supported colonies of gold coral (Fig. 5). At FFS the side scanning sonar of the Pisces V submersible was used to run north, southeast, and west bound tracts to document the limits of the gold coral bed. Gold corals were found to be exclusively associated with the 10-m-high basalt pinnacles that were evident on the sonar images in the form of distinct reflections and dark shadows.

Figure 5. R/V KoK SeaBeam survey of the Cross Seamount summit from data collected in the years 2000 and 2001. Depth contour level is 10 m. The steep slopes of seamount are dark. The walls of the central depression are also dark.

Since the 2001 dives revisited the same sites inspected in 2000 some additional operations were included. Gold coral colonies were measured and marked in 2000 and remeasured in 2001 at FFS and Cross Seamount to help validate current notions as to the growth rate of gold coral. These data have yet to be examined. Also at FFS and Cross Seamount, temperature recorders were deployed for a year to look for differences in seasonal temperature profiles. The FFS profile exhibited cyclical oscillation from 9.5° to 12° C with a single warm spike to 14° C. In contrast Cross Seamount was less cyclical and oscillated at a colder range from 7.5° to 9.5° C.

Future work

Additional submarine time has been requested in 2002 as part of the HURL's Ocean Exploration Initiative to survey seamounts at the northern extent of the archipelago. The additional data from coral and fish surveys conducted at these locations will greatly improve the statistical power with which to evaluate the importance of corals to the deepwater fish community. In particular being able to look at deepwater assemblages adjacent to a number of seal colonies other than just FFS could reveal unanticipated patterns.

Parrish F. A., K. Abernathy, G.J. Marshall and B.M. Buhleier (2002) Hawaiian monk seals (Monachus schauinslandi) foraging in deep-water coral beds. Marine Mammal Science 18(1):244-258.