The collapse of benthic macroalgal blooms in response to self‐shading

1. Our goal was to use physiological indicators photosynthesis–irradiance (P–I) response, nutrient status], population level feedbacks (self‐shading) and ambient environmental conditions (dissolved nutrients, light, temperature) to improve our understanding of the seasonal and spatial population dynamics of Cladophora. 2. Cladophora grew in three distinct phases, rapid growth early in the season (May–July), a mid‐season population collapse (July–August) and autumn re‐growth. Across all sites and dates, mean net maximal photosynthesis P_M (NET)] was 6.9 ± 3.9 mg O₂ g DM^?1 h^?1, and α was 0.055 ± 0.025 mg O₂ g DM^?1 μm photons^?1 m^?2. Mean values for critical irradiance (I_CR) and the half‐saturation light intensity (I_K), were 42.9 ± 32.1 and 189.3 ± 123.8 μm photons^?1 m^?2 s^?1 respectively. 3. At most sites growth was phosphorus‐limited. Values of α were significantly higher at a site influenced by a nutrient enriched river plume, where algal growth was phosphorus‐sufficient. 4. Photoinhibition was not apparent in any of our P–I experiments. Even if photoinhibition had been apparent during in vitro P–I experiments, population level photosynthetic rates in the field would be little affected because intense self‐shading restricts inhibiting irradiances to the upper few mm–cm of the algal canopy. 5. Our physiological (P–I response) experiments contradicted previous assertions that high ambient temperatures, or nutrient deficiency, were primary causes of mid‐summer sloughing. In our study, sloughing occurred simultaneously at nutrient enriched and nutrient deficient sites, at temperatures well below critical values found during in vitro experiments, and our indicator of physiological condition (P–I response) remained unchanged leading up to, or immediately after, the sloughing event. 6. Self‐shading can reduce the convexity of the P–I response within in vitro incubations, even when the amount of algal material is low. Our experiments used 0.08 g DM of algal material that formed clumps c. 1 cm thick. Under these conditions, we estimated negligible (<1%) effects on P_M, a 12% reduction in apparent values of α, and 14% and 17% increases in values of the α‐dependent terms I_CR and I_K, respectively. 7. Our results are consistent with the hypothesis that a population‐level negative feedback (self‐shading) is responsible for sloughing in dense macroalgal beds. Sloughing was probably inevitable once macroalgal bed density and thickness surpassed a critical threshold. Cells towards the base of the bed received insufficient light to maintain metabolic balance, began to decay and weaken, and became increasingly susceptible to physical detachment from shear stress.