172 



MARSH 



1 g C m~^ d~^ and 0.5 kg CaCO^ m~^ y~\ respectively. 

 The first mode is itself regarded as a composite and can 

 be further subdivided. In particular, it may include areas of 

 continuous coralgal cover and discrete heads with 

 P = 20 and G = 10, where water depth and circula- 

 tion are adequate. Kinsey proposed a "standard" reef flat 

 with a gross P of 7 ± 1 g C m~^ d"\ a P:R ratio of 

 1 ± 0.1, and a G of 4 ± 1 kg CaCOs m"^ yr"\ 

 where "reef flat" conforms to the concept of a fully 

 developed (at or near present-day sea level), areally exten- 

 sive (at least 100 m across), high-activity zone of the 

 coralgal type. 



Kinsey (1982) also considered comparative aspects of 

 calcification rates and reef growth (accretion) between 

 Pacific and Caribbean reefs and attempted to resolve what 

 seemed to be discrepancies between the two oceans. He 

 noted that the apparently higher rates in the Caribbean 

 were based primarily on long-term accretion rates deter- 

 mined from stratigraphic methods and that most estimates 

 in the Pacific were derived from short-term chemical 

 changes in resident water masses. He concluded that there 

 probably was a faster growth of Caribbean reefs during the 

 Holocenc epoch, with major factors being differences in 

 sea level, tectonics, and wave energy. The particular com- 

 bination of these factors in the Caribbean led to diminutive 

 surface features and proportionately greater seaward 

 slopes there, with wider expanses of reef flat and propor- 

 tionately smaller seaward-slope areas in the Pacific. Kinsey 

 further concluded that any interocean differences in the 

 calcifying capacity of reef communities are small. Hence, 

 additional research has served to put earlier Enewetak 

 work into a broader context but has not drastically altered 

 earlier conclusions resulting from Enewetak work. 



One earlier impression probably arising from Enewetak 

 research (e.g., Odum and Odum, 1955) should be modi- 

 fied. As pointed out by both Smith (1983) and Kinsey 

 (1983), initial reports of high productivity of reef-flat com- 

 munities led to the tendency to regard whole-reef systems 

 or "coral reefs" as being one of the world's most produc- 

 tive ecosystems. However, if the complete system, particu- 

 larly including the lagoon and extensive sand/rubble areas, 

 is considered, the rates of production are much more mod- 

 est. Kinsey (1983) tabulated published values of commu- 

 nity metabolism for complete reef ecosystems and showed 

 P (gross) ranging from 2.3 to 6.0 g C m~^ d~\G ranging 

 from 0.5 to 1.8 kg CaCO m"^ yr^', and P:R ratios hold- 

 ing constant at 1 for the five studies. The distinction 

 between particular reef communities and whole reef 

 ecosystems is one which must be more carefully drawn in 

 future studies. 



Somewhat related to this point is the question of the 

 metabolism of sediment communities, which comprise a 

 substantial proportion of the whole system. Harrison 

 (1983) studied this question by placing plastic domes over 

 such communities at Enewetak and monitoring O2 and 

 CO2 fluxes. He derived empirical respiratory quotient 

 values of 0.8 and repwrted that more carbon is respired by 

 the sediment community than is produced. He calculated 



that excess production exported from the windward reef 

 flats was sufficient to support the metabolsim of these dis- 

 tinctly heterotrophic sediment communities. Both produc- 

 tion and respiration showed a decline with depth. Accord- 

 ing to Harrison, "Biotic and functional comparisons 

 between Enewetak and Kaneohe Bay, Hawaii, suggest 

 metabolic and structural similarities between these physio- 

 graphically disparate coral reef ecosystems." Thus, there is 

 a recurring theme of similarities between reef processes at 

 Enewetak and those in other seemingly different reef 

 ecosystems. 



Several important insights about nutrient availability 

 had their origins at Enewetak but have been sharpened 

 and extended by work at other localities. After the initial 

 measurements of nitrogen fixation at Enewetak (Webb et 

 al., 1975; Wiebe et al., 1975), there followed a number of 

 other studies of this process at other localities (e.g., Cross- 

 land and Barnes, 1976; Burris, 1976; Capone et al., 

 1977). However, Szmant-Froelich (1983) pointed out that 

 such measurements have generally been restricted to reef 

 flats or back-reef areas and that denitrification (conversion 

 of N03^ to N2) has not been adequately measured in any 

 reef environments. 



Entsch et al. (1983), working on the Great Barrier 

 Reef, also conducted research on nutrient availability. They 

 found a large pool of nitrogen and phosphorus in car- 

 bonate sediments and in the interstitial waters of the sur- 

 face layers of sediments. Nutrient concentrations were con- 

 sidered to be sufficient to allow high rates of uptake by 

 epilithic algae. This is apparently an imfxsrtant recycling 

 mechanism in reef systems. 



Andrews and Muller (1983), building upxDn DiSalvo's 

 (1971, 1974) idea of regenerative spaces, measured 

 nutrients in a lagoonal patch reef of the Great Barrier Reef 

 complex and studied rates of water percolation through 

 the reef. Concentrations of NOa" and P04~"' in cavities on 

 the vertical face of the reef were found to be significantly 

 higher than in the surrounding water. Nitrogen export 

 through tidal flushing of their patch reef was reported to 

 be of the same order of magnitude as export from the 

 Enewetak reef flat studied by Webb et al. (1975); this 

 export was presumably supported by nitrogen fixation. The 

 molar ratio of nutrient regeneration rates was calculated to 

 be 140:2:7 for NO3 :N02":P04"^; if NOs* and NO2" were 

 lumped, the N:P regeneration ratio approximated 20:1. 



Smith (1983) further sharpened our understanding of 

 productivity and nutrient relationships in reef ecosystems 

 by pointing out that the net productivity of the whole-reef 

 system (rather than simply the reef-flat portion of the 

 system) is low. He estimated it to be less than 100 mgC 

 m~^ d~\ or within about the same range as "new" pro- 

 duction in open-ocean planktonic systems. As he and 

 Szmant-Froelich (1983) pointed out, any net production in 

 the ecosystem as a whole requires an input of new 

 nutrients. Recycling of nutrients already in the system can 

 support high gross production if the P:R ratio is exactly 1 

 and if the recycling is efficient. If recycling processes are 

 not efficient then there must be an input of new nutrients 



