REEF PROCESSES 



163 



on the alqal portion and a net release on the coral portion. 

 There were no significant day-night differences, but there 

 was greater export of NH4^, DON, and total N during 

 noon-to-midnight periods than during midnight-to-noon 

 periods. The C:N ratio decreased progressively from 

 offshore (15; 1) to the lagoon (6.6:1). There was a major 

 input of organic nitrogen to downstream portions of the 

 coralgal transect, mainly in the form of algal fragments bro- 

 ken off from the surf zone. The blue-green alga Calothrix 

 Crustacea seemed to be a major nitrogen fixer; hetero- 

 trophic bacteria were apparently not important in this pro- 

 cess since fixation was strongly light-dependent. The 

 increase in POC as water crossed the reef flat was propor- 

 tionately less than that of total nitrogen. Webb et al. calcu- 

 lated that 1000 kg N ha"^ yr~' was exported from the 

 reef flat, a high value that falls at the upper end of the 

 range of nitrogen fixation values for managed agricultural 

 plots. As we have already seen, this process of nitrogen 

 fixation was invoked by Johannes et al. as an important 

 part of the explanation of how reef ecosystems have high 

 productivities in the midst of nutrient-poor oceanic waters. 

 Nitrogen metabolism, like other metabolic processes, was 

 clearly found not to be dominated by corals. 



Webb and Wiebe (1975) made additional observations 

 on the nitrification processes on a reef; they reported on 

 in-situ and in-vivo incubations with and without an 

 ammonium oxidase inhibitor and concluded that an auto- 

 trophic pathway involving two separate organisms was 

 operating in the oxidation of ammonium to nitrate. The 

 bacterium Nitrobacter agilis was found to be at least one 

 organism responsible for the terminal oxidation of NO2 to 

 NO3-. 



A later paper, by Wiebe et al. (1975), also considered 

 aspects of nitrogen fixation in a coral reef community. It 

 suggested that, since algal flats fix nitrogen at rates com- 

 parable to those in managed agriculture, and this fixation 

 contributes to high productivity of adjacent reefs and 

 lagoons, algal flats should receive increased conservation 

 priority. They further observed that Calothrix Crustacea, 

 the dominant nitrogen fixer, grows in two forms. One of 

 these forms is a thin, yellow-brown, often almost 

 unispecific film covering large portions of the intertidal reef 

 flat. Most of this algal film remains moist at low tide; at 

 high tide herbivorous fish (especially acanthurids and 

 scarids) graze it extensively. Another growth form of 

 C. Crustacea occurs along the upper intertidal bench zone 

 as a black, felt-like mat up to 5 mm thick; this mostly dries 

 out at low tide and is not heavily grazed by herbivorous 

 fish at high tide because the water is too shallow. The 

 nitrogen fixation rates of moist samples of the upper inter- 

 tidal form averaged only 60% of those of the reef flat (34 

 versus 55 X 10~^ moles h~^ cm~^). However, per unit 

 of horizontal map area, the actual surface area of coral 

 and reef rubble is much greater than that on the algal flat 

 and may, therefore, lead to comparable rates of nitrogen 

 fixation for the two habitats, normalized to square meters 

 of map area. Wiebe et al. further stated that the nitrogen 

 fixed by Calothrix may enter the reef trophic web directly 



by grazing, through broken-off fragments, or through 

 release into solution. 



Wiebe (1976) summarized the above studies and 

 further pointed out that salinities ranging from 2 to 45 ppt 

 had no detectable effect on the rate of nitrogen fixation. 

 Furthermore, the rate was temperature-dependent and 

 approximately doubled between 27°C and 36°C, was at 

 24°C, and increased for 2 hours then ceased at 39°C. 

 The greatest upstream-downstream increase in concen- 

 tration of nitrogen species was for DON, followed by NH3 

 and NOs^ in about equal concentrations. Not detected in 

 the flow studies was NO2 ■ 



Phosphorus Cycling 



There are fewer studies of phosphorus cycling in the 

 reef ecosystem as a whole. Odum and Odum (1955) made 

 a few measurements of reactive phosphorus in waters flow- 

 ing over the intcrisland reef, in waters entering the wide 

 passage not far from their reef transect (presumed to be 

 representative of oceanic waters outside the atoll) and in 

 the lagoon. They reported levels of 0.26 to 0.64 ^g atoms 

 1~' and concluded that there was a tight cycling of this ele- 

 ment internally in the reef-flat community. Gilmartin (1960) 

 likewise made a few measurements in lagoon waters and 

 found generally low levels of the same order of magnitude 

 as those reported by the Odums. Measurements by 

 Pomeroy and Kuenzler (1967) were made incidentally to 

 work with individual populations; their work is discussed 

 later. 



The first extensive measurements of changes in waters 

 flowing across the reef flats were made by Pilson and 

 Betzer (Johannes et al., 1972; Pilson and Betzer, 1973). 

 They reported that the concentrations of reactive and 

 organic phosphorus did not show detectable change in 

 waters flowing across a coralgal transect but that there was 

 a slight decrease in reactive P and a slight increase in 

 organic P across a strictly algal transect. In particular, they 

 found that concentrations did not vary in proportion to 

 photosynthesis and respiration rates of the whole commu- 

 nity, despite the fact that their ability to detect changes 

 was at least two orders of magnitude more sensitive than 

 would have been required to detect such changes if the 

 Redfield (atomic) oxygen:phosphorus ratio of 138:1 was 

 applicable in this system. Pilson and Betzer also found no 

 diurnal variations in concentrations as waters flowed across 

 the repf. This remarkable constancy suggested to them 

 that the plants were taking up phosphorus at a nearly con- 

 stant rate, regardless of the magnitude of photosynthetic 

 activity. Their mean concentrations of P in flowing waters 

 were 326 nmoles total P, 172 nmol reactive P, and 154 

 nmol organic P. 



The conclusions of Pilson and Betzer were later chal- 

 lenged by Atkinson (1981), who worked primarily in 

 Kaneohe Bay, Hawaii, but also made some observations at 

 Enewetak. He found that exchange rates between reef 

 benthos and the water column did not fit the Redfield ratio 

 and concluded that changes in phosphate concentration of 



