TY - JOUR
T1 - Determinants of the steep species-area relationship of coral reef fishes
AU - Belmaker, J.
AU - Ben-Moshe, N.
AU - Ziv, Y.
AU - Shashar, N.
N1 - Funding Information:
Acknowledgments Special thanks to M.L. Rosenzweig, M. Kiflawi, Z. Abramsky, O. Ovadia and several anonymous reviewers for fruitful discussion and constructive comments, and to O. Ben Shafrut for help with diving. This study was partly supported by the PADI foundation, the Lerner-Gray Memorial Fund and the Kreitman foundation.
PY - 2007/3/1
Y1 - 2007/3/1
N2 - The increase in species richness with area is known as the species-area relationship (SPAR). Although several mutually non-exclusive processes may produce the SPAR, the null, often ignored, hypothesis states that a SPAR can be generated by random placement alone. The log-log-transformed SPAR of coral reef fishes on small patch-reefs revealed a steep slope of 0.55. However, this slope was dependent on the cumulative area of the reef examined and was therefore affected by random placement. After statistically removing the contribution of random placement from the SPAR, the slope was estimated to be 0.21. This is consistent with estimates from other, mostly terrestrial, systems. Furthermore, a randomization procedure, where the probability of fishes to reach a patch was proportional to reef area, showed that the field measured SPAR did not differ from random placement. In addition, fish assemblages on species poor reefs did not form subsets of species rich reefs (i.e., no nestedness) beyond that expected from random placement. Steep log-log-transformed SPARs can be formed by random placement alone, indicating that caution should be used when assigning an ecological meaning to SPARs generated from small spatial scales.
AB - The increase in species richness with area is known as the species-area relationship (SPAR). Although several mutually non-exclusive processes may produce the SPAR, the null, often ignored, hypothesis states that a SPAR can be generated by random placement alone. The log-log-transformed SPAR of coral reef fishes on small patch-reefs revealed a steep slope of 0.55. However, this slope was dependent on the cumulative area of the reef examined and was therefore affected by random placement. After statistically removing the contribution of random placement from the SPAR, the slope was estimated to be 0.21. This is consistent with estimates from other, mostly terrestrial, systems. Furthermore, a randomization procedure, where the probability of fishes to reach a patch was proportional to reef area, showed that the field measured SPAR did not differ from random placement. In addition, fish assemblages on species poor reefs did not form subsets of species rich reefs (i.e., no nestedness) beyond that expected from random placement. Steep log-log-transformed SPARs can be formed by random placement alone, indicating that caution should be used when assigning an ecological meaning to SPARs generated from small spatial scales.
KW - Nested subset
KW - Red Sea
KW - SPAR
KW - Scale
KW - Spatial distribution
UR - http://www.scopus.com/inward/record.url?scp=33947692166&partnerID=8YFLogxK
U2 - 10.1007/s00338-006-0162-1
DO - 10.1007/s00338-006-0162-1
M3 - Article
AN - SCOPUS:33947692166
SN - 0722-4028
VL - 26
SP - 103
EP - 112
JO - Coral Reefs
JF - Coral Reefs
IS - 1
ER -