Data Set (knb-lter-sbc.99.1)

SBC LTER: Reef: Kelp Forest Community Dynamics: Kelp Forest Data to support "Estimating biomass of benthic kelp forest invertebrates from body size and percent cover"

  Summary and Data Links People and Organizations Temporal, Geographic and Taxonomic Coverage Methods and Protocols  

These methods, instrumentation and/or protocols apply to all data in this dataset:

Protocols and/or Procedures
Description:

Field collection

The relationship between individual size and mass or aggregate percent cover and mass were determined for 84 species of benthic invertebrates found on shallow subtidal reefs in the Santa Barbara Channel. Specimens of all species were collected from 11 reefs (4 to 12 m depth) that are long-term study sites for the Santa Barbara Coastal Long Term Ecological Research program (see http://sbc.lternet.edu/sites/sampling/ for detailed site descriptions). Specimens of common taxa were collected throughout the year over a four-year period (April 2010 – May 2014) to account for seasonal and interannual variation in body weight and composition; specimens of uncommon taxa were collected opportunistically. Species displaying large intra annual variation in biomass due to seasonal gonadal development (e.g., sea urchins, crustaceans) were collected during non-spawning periods.

Specimens were collected using SCUBA, brought back to the laboratory in insulated coolers and placed in tanks supplied with running filtered seawater for 1-2 days before processing. This procedure allowed organisms to clear their digestive tract minimizing the contribution of gut contents to biomass. Species that could not be easily maintained in seawater tanks (e.g. sponges, hydroids) were processed immediately upon their arrival to the laboratory.

Measurements of body size and percent cover

Converting data on numerical abundance 101 (i.e. density) to biomass requires information on the relationship between individual size and mass. For solitary species individual size was measured as the length of a morphological trait specific for that species. Because our objective was to develop non-destructive methods for estimating standing biomass from abundance data collected in situ, only traits that were easily measured by divers without damaging the organism were used (e.g., total length, arm diameter, test diameter). Individuals of varying sizes were collected to generate relationships between length and wet mass for 53 species (n = 6 - 207 individuals per species). Percent cover is frequently used as a measure of abundance for many colonial species (e.g., hydrozoans, anthozoans, polychaetes). Developing non-destructive methods for estimating standing biomass for these species thus requires information on the relationship between percent cover and mass. To obtain this information, divers measured the percent cover of a species within 10 cm x 10 cm quadrats using a uniform grid of 20 points. After data on percent cover were recorded, all tissue of the targeted species within the quadrat was collected by carefully removing it from the bottom compound ascidians, bryozoans, sponges) and small aggregating taxa that are often too numerous and indistinct for divers to efficiently count (e.g., hydrozoans, anthozoans, polychaetes). Developing non-destructive methods for estimating standing biomass for these species thus requires information on the relationship between percent cover and mass. To obtain this information, divers measured the percent cover of a species within 10 cm x 10 cm quadrats using a uniform grid of 20 points. After data on percent cover were recorded, all tissue of the targeted species within the quadrat was collected by carefully removing it from the bottom or by collecting the substrate to which the species was attached and removing the tissue in the laboratory. Replicate quadrats containing varying amounts of percent cover were sampled to generate sufficient data for determining the relationship between percent cover and biomass for 31 species (n = 9 - 40 quadrats per species).

Measurements of body mass

Biomass can be reported using a variety of metrics. To facilitate interconversion among these various metrics we estimated body mass as wet mass, dry mass, shell free and decalcified dry mass and ash free dry mass. To minimize effects of water adhesion on wet mass measurements, specimens were removed from holding tanks and blotted dry with a clean paper towel or exposed in air at room temperature and allowed to desiccate for 1-2 minutes prior to being weighed (Dermott and Paterson 1974). Estimates of dry mass were obtained by placing specimens of known wet mass in a drying oven at 60°C for several days until their mass remained constant. Water content was estimated as [1 – (dry mass/ wet mass)] x 100.

After being measured and weighed wet, the calcareous shells of molluscs and the chitonous exoskeletons of crustaceans were separated from the soft tissue, and dried and weighed separately to obtain estimates of dry mass (i.e. dried soft tissue + shell) and shell-free dry mass (i.e. dried soft tissue only). The separation of soft tissue from chitonous exoskeletons of crustaceans was facilitated by microwaving the specimen for 1-2 minutes. Specimens of species with calcified structures such as bryozoans, gorgonians and echinoderms were dried whole to measure dry mass, and then treated with a 5% HCl solution for 3-4 hours to dissolve the calcified structures. Treatment with acid was repeated as necessary to remove all calcification. After full decalcification the remaining soft tissue was carefully separated from the acidic solution, rinsed in deionized water and placed back into the drying oven until the mass remained constant. The dried soft tissue was then reweighed to obtain a measure of decalcified dry mass.

Dry mass samples of species without hard external structures, and shell-free decalcified dry mass samples of species with hard structures were processed to obtain estimates of ash free dry mass. Samples of known mass were placed in aluminum trays and burned in a muffle furnace at 500°C for 4 hours to volatize all organic material (Holme and McIntyre 1984). The weight of the remaining ash was subtracted from the shell free decalcified dry mass to obtain a value for the ash-free dry mass of the sample.

The relationship between length and wet mass was best explained by the power function M = aLb where M is wet mass in grams and L is length of the species-specific morphological trait used to estimate size in mm. Linear regression was performed on log transformed values of length and mass to estimate the slope (b) and intercept (a) for each species. The antilog of the intercept was calculated for use in the power function. Smearing estimates (Duan 1983) were calculated to correct for biased caused by back-transformation of logged parameters, which can result in an underestimate of the response variable (Smith 1993). Residuals from the log-log regression between length and mass for each species were tested for homoscedasticity using White’s General Test (SAS 9.4 Cary, NC, USA). A simple linear regression of the form M = bC was used to describe the relationship between percent cover (C) and wet mass (M). Examination of residuals and graphical inspection showed that percent cover data met the assumptions of linear regression for all species examined.

References

Dermott R, Paterson C (1974) Determining dry weight and percentage dry matter of chironomid larvae. Can J Zool 52:1243-1250.

Duan N (1983) Smearing estimate: a nonparametric retransformation method. J Amer Stat Ass 78: 605-610.

Holme NA, McIntyre AD (1984) Methods for the study of marine benthos. IBP handbook No 16. Blackwell Scientific, Oxford.

Smith R (1993) Logarithmic transformation bias in allometry. Am J Phys Anthr 90:215-228.