Phosphorus released from aerobic sediment surfaces made up the major fraction of the total P-load to the trophogenic zone in four shallow Danish lakes in 1987. Gross release rates were 15, 21, 33, and 100 mg P m-2 d-l (average summer values). In three of the lakes, water temperature alone explained -70% of the seasonal variation in sediment P release. Long-term experiments with undisturbed sediment cores revealed that the P release was significantly influenced by temperature and NO,-in all four lakes, but by pH only in one lake when these three parameters varied within the normal seasonal range (temperature: O"-21°C; NO,-: O-200 PM; pH: 7.5-10.5). Q10 values for the temperature effect were between 4.1 and 6.8 in the three lakes with large proportions of Fe-bound P in the sediment, but only 3.5 in the lake in which the P pool was relatively small. The thickness of the oxidized layer at the sediment surface varied from 3 to 15 mm and decreased when temperature increased. High N03-concentrations increased the thickness throughout the season and accordingly sediment P release was reduced in winter and early summer. Meanwhile, in late summer when inorganic N was depleted in lake water, N03-additions increased the P release from sediment, probably by stimulating the mineralization process.In the past 15-20 yr it has become evident that significant amounts of P may be released from aerobic sediment surfaces of shallow temperate lakes. The P release often constitutes a major fraction of the total P load to the trophogenic zone in summer (Andersen 1982; Ryding 198 5). This internal P loading has been identified as an important mechanism in delaying recovery of shallow lakes following reduced external P loading (Ryding 1985), although homeostasis in the biological structure may be important too (e.g. Jeppesen et al. 199 1). Several physical, chemical, and biological factors influence the P exchange across aerobic sediment surfaces. Lake water temperature, N03-, and pH have been proposed as important steering factors for the season-' Present address:

Seasonal variation in P sedimentation, sediment P release, and sediment P pools was studied in a coastal marine sediment at 16-m water depth. Net sedimentation of P amounted to 5 l-63 mmol m-2 yr-l, compared to a sediment P release of 34 mmol m-2 yr-l. The resulting deficit corresponded with a burial flux of 18 mmol P m-2 yr-I. Iron-bound P was the most dynamic P pool in the mixed surface sediment, where it made up 175 mmol m-2 of a total of 530 mmol P m-2. Iron-bound P decreased rapidly with depth and contributed only 3.5% to the burial flux. P sedimented in spring was retained in the pool of iron-bound P until SeptemberOctober, when half of the annual P release occurred. The autumn maximum in P release seemed to be linked to sulfate reduction by two mechanisms: sulfide was primarily responsible for the reduction of ferric oxyhydroxides (FeOOH) and hence the release of iron-bound P, and precipitation of FeS and FeS, restricted the upward migration of Fe 2+ but not of dissolved reactive phosphate (DRP) and resulted in a saturation of sorption sites for DRP on FeOOH in the sediment surface layer. The saturation of sorption sites was reflected by a minimum ratio of FeOOH to iron-bound P in surface sediment measured in October. Seasonal changes in this ratio provided the best correlation with the DRP efflux (R = -0.74), indicating that adsorption onto FeOOH is probably the most important factor controlling sediment P release.

In the sediment of the shallow, hypertrophic Lake Sønderby, Denmark, potentially mobile phosphorus (Pmobile) was determined by a sequential extraction technique as the sum of porewater P, iron-bound P, and nonreactive P (i.e., polyphosphates and organic P). A good agreement was observed between loss rates of Pmobile in the top 10 cm of the sediment from winter to summer, P release rates measured in undisturbed sediment cores, and rates of P accumulation in the lake water from winter to summer (22, 32, and 30 mg of P m(-2) day(-1), respectively). This suggests that the operationally defined Pmobile was the sediment P fraction responsible for the internal loading in the lake. In autumn 2001, 11 mg of aluminum (Al) L(-1), equivalent to 31 g of Al m(-2), was added to the lake water. This dosage represented a 4:1 molar ratio between Al and Pmobile. The Al treatment significantly decreased lake water P, and P precipitated from the lake water was recovered as Al-bound P in the sediment after the treatment. Internal P loading was reduced by 93% in the two posttreatment years, relative to 2001. Accordingly, average summer concentrations of total P in lake water declined from 1.28 (SE = 0.17) and 1.3 (SE = 0.14) mg L(-1) in the two pretreatment years to 0.09 (SE = 0.01) and 0.13 (SE = 0.01) mg L(-1) in the posttreatment years. pH levels remained unchanged relative to pretreatment levels, while the total alkalinity was reduced from 3.2 (SE = 0.04) to 2.7 (SE = 0.03) mequiv L(-1).

No abstract

Carbon and nitrogen mineralization were determined along a transect from a mangrove forest to a seagrass meadow in the Bangrong area, Phuket Island, Thailand. Vertical sediment profiles of carbon oxidation were measured as sulfate reduction rates (SRR) using the 35 S technique and by monitoring net TCO 2 and DOC production and Fe(III) reduction using anaerobic sediment incubations ('jar' technique). Nitrogen transformations were measured simultaneously as net NH 4 + and DON production. In addition, total benthic metabolism and net nitrogen exchange were determined as fluxes of O 2 , TCO 2 , DOC, and DIN (NO 3 -and NH 4 + ) across the sediment-water interface. Rates of carbon and nitrogen transformations in this vascular-plant (high C:N)-dominated area were low compared with areas fuelled by detritus of marine origin (low C:N). It appears that the high content of structural biopolymers (e.g. lignocelluloses) hampers microbial activity. Suboxic respiration with Fe(III) as electron acceptor accounted for 70 to 80% of the total carbon oxidation in the rooted mangrove forest sediment, whereas SRR and aerobic respiration were responsible for about 20 and < 6%, respectively. The role of SRR decreased to about 10% and aerobic respiration increased to 45-65% in an adjacent bioturbated mudflat, while Fe(III) respiration decreased to 30-40%. At the sand flat and seagrass meadow outside the mangrove forest, Fe(III) respiration only accounted for 15 and ~0%, respectively, whereas SRR was responsible for 20 to 45% of the total carbon oxidation. However, the most important electron acceptor in the area outside was oxygen (55 to 75%). The shift in dominance of electron acceptors along the transect is primarily related to the presence of roots and infauna, but the sediment composition (e.g. grain size, organic content and iron content) is believed to be an important co-factor. The net production of ammonium in the sediment was not balanced by fluxes of DIN across the sediment-water interface. The missing nitrogen was assigned to a rapid and efficient bacterial ammonium assimilation at the sediment surface as indicated by ammonium turnover times of about 1 d. KEY WORDS: Mangrove forest · Seagrass · Carbon · Nitrogen · Mineralization · Sulfate reduction · Fe(III) reduction · Benthic metabolismResale or republication not permitted without written consent of the publisher Aquat Microb Ecol 22: 199-213, 2000 grove and seagrass environments have shown that densities are generally low compared with other marine habitats (Alongi & Sasekumar 1992). The infaunal density and diversity is particularly low within the mangrove forest, where burrowing crabs are the dominating faunal feature. These may, however, handle and consume a considerable fraction of the litter fall (Robertson 1986). Highest faunal densities are generally found on adjacent mudflats, where large populations of epibenthic mollusks graze on microphytobenthos. Tropical seagrass meadows, on the other hand, are characterized by highly diverse, but not very abundan...