species zonation does not appear to repre- 

 sent serai stages of succession except, 

 perhaps, for locations of recent coloniza- 

 tion or where sediment is accumulating 

 rapidly. The role of mangroves in 

 land-bui'lding seems more passive than 



active. Geomorphological and hydrological 

 processes appear to be the dominant forces 

 in determining whether mangrove shorelines 

 recede or grow. The role of mangroves is 

 to stabilize sediments which have been 

 deposited by physical processes. 



3.3 NUTRIENT CYCLING 



Current understanding of nutrient 

 cycles in mangrove ecosystems is far from 

 satisfactory. Sporadic field measurements 

 have been made, but a complete nutrient 

 budget has not been published for any 

 mangrove ecosystem in the world. 



Several pioneering field studies were 

 conducted in Florida (Carter et al. 1973; 

 Snedaker and Lugo 1973; Onuf et al. 1977) 

 and one simulation model of mangrove nu- 

 trient cycling has been published (Lugo et 

 al. 1976). Preliminary measurements of 

 nitrogen fixation were made (Zuberer and 

 Silver 1975; Gotto and Taylor 1976; 

 Zuberer and Silver 1978; Gotto et al. 

 1981). Based on these studies, we present 

 the following preliminary conclusions. 



Mangrove ecosystems tend to act as a 

 sink (net accumulator) for various ele- 

 ments including macro nutrients such as 

 nitrogen and phosphorus, trace elements, 

 and heavy metals. As we have discussed in 

 section 1.7, these elements are removed 

 from waters flowing through mangrove 

 swamps by the concerted action of the 

 mangrove prop roots, prop root algae, the 

 associated sediments, the fine root system 

 of the mangrove trees, and the host of 

 small invertebrates and microorganisms 

 attached to all of these surfaces. Al- 

 though the turnover times for these ele- 

 ments in mangrove swamps are not known, it 

 appears that at least a portion may be 

 stored or tied up in wood, sediments, and 

 peat for many years. 



Although mangrove ecosystems may tend 

 to accumulate nutrients, there is a con- 

 tinual loss through export of particulate 

 and dissolved substances. If significant 

 nutrient storage and resultant high pri- 

 mary production are to occur, there must 

 be a continual input of nutrients to the 

 mangrove forest from outside the system 

 (Figure 7). Where nutrient influx to the 

 mangrove ecosystem is approximately 

 balanced by nutrient loss in exported 

 organic matter, then nutrient storage will 

 be minimal and mangrove net primary pro- 

 duction will be low. This appears to 

 occur in the scrub mangrove community type 

 and to a lesser extent in the basin and 

 hammock community types. 



Carter et al. (1973) and Snedaker and 

 Lugo (1973) have hypothesized that the 

 greatest natural nutrient inputs for man- 

 grove swamps come from upland and terres- 

 trial sources. Apparently for this rea- 

 son, the most luxuriant and productive 

 mangrove forests in south Florida occur in 

 riverine locations or adjacent to signifi- 

 cant upland drainage. 



Localized sources of nutrients, such 

 as bird rookeries, can result in greater 

 nutrient storage and higher mangrove pro- 

 ductivity (Onuf et al. 1977). If however, 

 large bird rookeries (or artificial nu- 

 trient inputs) occur in poorly flushed 

 sections of mangrove ecosystems, resultant 

 high nutrient levels may inhibit mangrove 

 growth (R. R. Lewis, III, Hillsborough 

 Community College, Tampa, Fla.; personal 

 communication 1981). 



The output from the simulation model 

 of Lugo et al. (1976) suggests that if 

 nutrient input to a mangrove ecosystem is 

 reduced, then nutrient storage levels 

 within the mangrove ecosystem will be 

 reduced and mangrove biomass and produc- 

 tivity will decline. To our knowledge 

 this hypothesis has not been tested in the 

 field. 



Nitrogen fixation occurs in mangrove 

 swamps at rates comparable to those 

 measured in other shallow, tropical marine 

 areas (Gotto et al. 1981). Nitrogen 



30 



