Environ Monit Assess (2015) 187:234 DOI 10.1007/s10661-015-4431-8
Measuring coral size-frequency distribution using stereo video technology, a comparison with in situ measurements Joseph A. Turner & Nicholas V. C. Polunin & Stuart N. Field & Shaun K. Wilson
Received: 26 July 2014 / Accepted: 12 March 2015 # Springer International Publishing Switzerland 2015
Abstract Coral colony size-frequency distribution data offer valuable information about the ecological status of coral reefs. Such data are usually collected by divers in situ, but stereo video is being increasingly used for monitoring benthic marine communities and may be used to collect size information for coral colonies. This study compared the size-frequency distributions of coral colonies obtained by divers measuring colonies ‘in situ’ with digital video imagery collected using stereo video and later processed using computer software. The sizefrequency distributions of the two methods were similar for corymbose colonies, although distributions were
J. A. Turner (*) : N. V. C. Polunin School of Marine Science and Technology, University of Newcastle upon Tyne, Ridley Building, Newcastle upon Tyne NE1 7RU, UK e-mail: [email protected] J. A. Turner Joint Nature Conservation Committee, Monkstone House, Peterborough PE1 1JY, UK S. N. Field : S. K. Wilson Marine Science Program, Department of Parks and Wildlife, Kensington, WA 6151, Australia
different for massive, branching and all colonies combined. The differences are mainly driven by greater abundance of colonies >50 cm and fewer colonies 5 cm and was able to record measurements on 87 % of the colonies detected. However, stereo video only detected 57 % of marked colonies 5 cm across) and smaller coral colonies (5 cm were measured at all sites. At two of the sites, corals 5- and 5 cm within a 1-m belt and 62 colonies 5 cm. For colonies 5 cm, individual colonies were separated into five size categories (5–10, 11–20, 21–50, 51–100 and >100 cm, adapted from Mundy 1996). For analysis of size distribution and measurement ability, genera were separated by morphology: corymbose Acropora spp. (not including tabular or arborescent growth forms), massive Favia spp., Favites spp., Goniastrea spp., Lobophyllia spp., and Platygyra spp., and other branching Pocillopora spp., Porites spp., Seriatopora spp., and Stylophora spp. For corals 100
Total
In situ
135 (2.87)
386 (0.33)
586 (0.01)
173 (1.37)
87 (3.33)
1367
Stereo video
164 (3.14)
375 (0.37)
531 (0.01)
130 (1.50)
50 (3.64)
1250
All corals
χ2 =16.57, p100 cm) colonies represented only ~10 and ~5 % of the samples respectively (Fig. 2a). The size frequency distributions for each of the methods (data from all sites and taxa/morphologies) were significantly different (χ2 = 16.57, df=4, p100 cm) colonies and fewer small colonies (5–10 cm) being observed in situ (Table 1, Fig. 2a). For the corymbose acroporid colonies, distributions measured using the stereo video and in situ methodologies were very similar (Table 1, Fig. 2b). However, there were significant differences in size distribution between the two methods for both massive and other branching colonies (Table 1), primarily driven by the in situ method detecting more colonies in the 51–100-cm size class for both morphologies (Table 1; Fig. 2c, d). Although there were major differences in the composition of corals among sites, the in situ and stereo video methods described a similar community with respect to taxa and colony size (PERMANOVA, pseudo-f =
Fig. 2 Mean proportion (+S.E.) of a total (n=8), b corymbose (n=6), c massive (n=4) and d branching (n=4) colonies in each size category for each of the techniques. n5 cm, 190 (92.6 %) were identified using the stereo video methodology (visible in at least one camera). Of these identified colonies, 178 (86.8 % of all marked colonies) could be measured using the stereo video methodology (visible in both cameras). The in situ methodology consistently recorded larger diameters than the stereo video method, particularly for massive and branching growth forms (Fig. 4). Regression analysis highlighted significant relationships for measurements of corymbose acroporids (F1,78 =2133.4, p
Measuring coral size-frequency distribution using stereo video technology, a comparison with in situ measurements Joseph A. Turner & Nicholas V. C. Polunin & Stuart N. Field & Shaun K. Wilson
Received: 26 July 2014 / Accepted: 12 March 2015 # Springer International Publishing Switzerland 2015
Abstract Coral colony size-frequency distribution data offer valuable information about the ecological status of coral reefs. Such data are usually collected by divers in situ, but stereo video is being increasingly used for monitoring benthic marine communities and may be used to collect size information for coral colonies. This study compared the size-frequency distributions of coral colonies obtained by divers measuring colonies ‘in situ’ with digital video imagery collected using stereo video and later processed using computer software. The sizefrequency distributions of the two methods were similar for corymbose colonies, although distributions were
J. A. Turner (*) : N. V. C. Polunin School of Marine Science and Technology, University of Newcastle upon Tyne, Ridley Building, Newcastle upon Tyne NE1 7RU, UK e-mail: [email protected] J. A. Turner Joint Nature Conservation Committee, Monkstone House, Peterborough PE1 1JY, UK S. N. Field : S. K. Wilson Marine Science Program, Department of Parks and Wildlife, Kensington, WA 6151, Australia
different for massive, branching and all colonies combined. The differences are mainly driven by greater abundance of colonies >50 cm and fewer colonies 5 cm and was able to record measurements on 87 % of the colonies detected. However, stereo video only detected 57 % of marked colonies 5 cm across) and smaller coral colonies (5 cm were measured at all sites. At two of the sites, corals 5- and 5 cm within a 1-m belt and 62 colonies 5 cm. For colonies 5 cm, individual colonies were separated into five size categories (5–10, 11–20, 21–50, 51–100 and >100 cm, adapted from Mundy 1996). For analysis of size distribution and measurement ability, genera were separated by morphology: corymbose Acropora spp. (not including tabular or arborescent growth forms), massive Favia spp., Favites spp., Goniastrea spp., Lobophyllia spp., and Platygyra spp., and other branching Pocillopora spp., Porites spp., Seriatopora spp., and Stylophora spp. For corals 100
Total
In situ
135 (2.87)
386 (0.33)
586 (0.01)
173 (1.37)
87 (3.33)
1367
Stereo video
164 (3.14)
375 (0.37)
531 (0.01)
130 (1.50)
50 (3.64)
1250
All corals
χ2 =16.57, p100 cm) colonies represented only ~10 and ~5 % of the samples respectively (Fig. 2a). The size frequency distributions for each of the methods (data from all sites and taxa/morphologies) were significantly different (χ2 = 16.57, df=4, p100 cm) colonies and fewer small colonies (5–10 cm) being observed in situ (Table 1, Fig. 2a). For the corymbose acroporid colonies, distributions measured using the stereo video and in situ methodologies were very similar (Table 1, Fig. 2b). However, there were significant differences in size distribution between the two methods for both massive and other branching colonies (Table 1), primarily driven by the in situ method detecting more colonies in the 51–100-cm size class for both morphologies (Table 1; Fig. 2c, d). Although there were major differences in the composition of corals among sites, the in situ and stereo video methods described a similar community with respect to taxa and colony size (PERMANOVA, pseudo-f =
Fig. 2 Mean proportion (+S.E.) of a total (n=8), b corymbose (n=6), c massive (n=4) and d branching (n=4) colonies in each size category for each of the techniques. n5 cm, 190 (92.6 %) were identified using the stereo video methodology (visible in at least one camera). Of these identified colonies, 178 (86.8 % of all marked colonies) could be measured using the stereo video methodology (visible in both cameras). The in situ methodology consistently recorded larger diameters than the stereo video method, particularly for massive and branching growth forms (Fig. 4). Regression analysis highlighted significant relationships for measurements of corymbose acroporids (F1,78 =2133.4, p
