Ecological Performance - Rigs to Reefs

ECOLOGICAL PERFORMANCE AND TROPHIC LINKS: COMPARISONS AMONG PLATFORMS AND NATURAL REEFS FOR SELECTED FISH AND THEIR PREY

Principal Investigators: Mark Page (UCSB), Jenifer Dugan (UCSB), Milton Love

(UCSB), and Hunter Lenihan (UCSB)

Summary of Research

One of the major issues in the disposition of decommissioned oil platforms is the need for research that assesses the quality of platforms as habitat for ecologically and commercially important invertebrates and fishes. The primary goal of this project is to explore whether indices of ecological performance (e.g., nutritional condition, size, individual growth) of a model fish and its invertebrate prey differ between platforms and natural reefs. We selected the painted greenling (Oxylebius pictus) as our model fish. It is one of the few fish species found on every surveyed oil platform and natural reef in the Santa Barbara Channel region. They are territorial and thus have limited movement among locations. Invertebrate prey of painted greenling consist primarily of small amphipod crustaceans, but may also include small gastropods, polychaetes, and other invertebrates (see below). During the study, we sampled potential invertebrate prey monthly at two natural reefs (Naples, Mohawk) and two offshore oil platforms (Holly, Houchin). Painted greenling were sampled from these locations in September 2003 and April 2004.

Samples of invertebrates are collected within randomly placed 20 x 20 cm quadrats along transect lines at a depth of ~ 9 m using a combination of scraping and vacuum sampling. On return to the laboratory, prey items are separated from non-prey material that commonly includes turf forming algae and, from the platforms, mussels and other macroinvertebrates. The major non-prey components of the benthic habitat in each sample, such as the benthic algae, are also quantified in terms of weight or volume. Following this coarse processing, prey items are identified, counted and weighed, and for some taxa measured. These data will allow us to compare the standing crop, composition, and perhaps production of potential prey among locations. To date, we have processed 625 samples of invertebrates taken at the four locations. In addition to the field sampling, we are comparing the “instantaneous” molt rate and molt increments of caprellid amphipods, important prey items of painted greenling, among locations. Caprellid amphipods are maintained individually in the laboratory following sampling, and the number of individuals that molt from each location is recorded over three consecutive days. The molted exuvia and newly molted animal are measured to determine the molt increment of each individual that molts. These data may permit a comparison of individual growth, one component of production, of these amphipods among locations.

Our data to date indicate that the total density and taxonomic composition of potential invertebrate prey of painted greenling differ greatly among platform and reef locations (Fig. 1). Total densities of small invertebrates were more than an order of magnitude higher at Platform Holly than the other locations and were frequently the lowest at Mohawk Reef. The invertebrate assemblage at the platforms and Naples Reef is dominated numerically by gammarid and caprellid amphipod crustaceans. The species composition of caprellid amphipods, important prey items of painted greenling, differ among locations with an large invasive species (up to 18 mm) Caprella mutica comprising 61% of the assemblage at Holly, but <10% at the other locations (Table 1).

Figure 1. Mean density (±1SE) of major groups of organisms over time at platform and natural reef study locations. n=8 to 12 quadrats/location in each month. Note variable scales on y-axis.

The ecological performance of painted greenling is being compared across locations through measurements of the quantity and quality of prey items consumed (number and types of prey in the stomach and overall weight of stomach contents) and the nutritional state of the fish (condition factor, K). Painted greenling are collected by hand in the same area that invertebrates were sampled, injected through the mouth with formalin to preserve the stomach contents, and returned to the laboratory for processing. In the laboratory, each fish is measured (standard length) and weighed. The stomach from each fish is removed and the contents weighed, identified, and counted.

Preliminary data from fish sampled during September 2003 show that painted greenling at all locations were feeding primarily on gammarid and caprellid amphipods (Figure 2). There also appears to be the selective ingestion of caprellid amphipods over other types of prey; caprellid amphipods were present in stomach contents at disproportionately higher frequencies than their relative abundance at each location. This pattern was particularly evident at Mohawk Reef and Platform Holly. The condition of painted greenling (estimated as K) varied among locations and appeared related, at least in part, to the availability of amphipod prey; values of K were

Table 1. Species composition of caprellid amphipods (Caprella spp. as percent of total) at the platform and natural reef locations. Data are the % contributed by each species to the total weight of fish prey sampled on each platform (these data are from samples processed to date).

Caprellid Species

Location	C. californica	C. verrucosa	C. equilibra	C. mutica
Naples Reef	28.0	22.2	46.6	2.7
Platform Holly	1.0	<1.0	37.1	61.0
Mohawk Reef	62.7	18.7	17.6	0.5
Platform Houchin	4.2	5.3	81.0	8.1

substantially higher for fish at Platform Holly compared to fish from the other locations (Figure 3). This site also had the highest abundance of prey species, indicating K may be influenced by food availability for the model fish. However, an unknown “location effect” may be present as well since K is similar at Mohawk and Platform Houchin, but prey availability is higher at the platform. We have initiated a laboratory experiment to directly test the effects of prey abundance, species composition, and microhabitat structure (which also differs among our sites) on fish feeding rates and condition index. This experiment will help us better understand the variation observed in K-value of fish and its relationship to prey population dynamics across POCS platforms and natural reefs.

To complete the study, we will develop estimates of standing crop and production of these prey taxa at each location, process samples of painted greenling taken in April 2004 for stomach contents and condition factor, and conduct experiments on fish feeding behavior to link prey and predator performance.

Figure 2. a) Density of invertebrates in August 2004 at Naples and Mohawk Reefs and Platforms Holly and Houchin by taxa and b) proportion of prey items by taxa in the stomach contents of painted greenling sampled at these same locations in September 2004. Key: gamm=gammarid amphipod, capr=caprellid amphipod, isop=isopod, amph=unk amphipod, tana=tanaidacean, poly=polychaete, biva=bivalve, shri=shrimp, ostr=ostracod, deca=decapod, gast=gastropod, nema=nematode, bryo=bryozoan, cuma=cumacean

Figure 3. Condition factor (“K”)of painted greenling versus the log density of gammarid and caprellid prey at Platforms Houchin (hou) and Holly (hol), and Mohawk (mo) and Naples (na) Reefs.

Another objective of our study is to estimate the total possible production of fish and invertebrate prey on POCS platforms and rocky reefs using the ecosystem-trophic interaction software provided by ECOPATH/ECOSIM. The general goal of our modeling is to estimate how much fish biomass can be produced on POCS oil platforms versus natural rocky reefs through predation on populations of common fish prey species. We will use the fish and prey species abundance data collected in our monthly surveys, growth rates of painted greenling approximated from our field sampling data, growth rates of prey species determined in our laboratory experiments, and feeding rates recorded from our laboratory experiments in our models once these results are available.