Leibniz Center for Tropical Marine Ecology (ZMT)

Model description and standard settings

    1. General

    We applied and further developed a spatially explicit, individual-based model of a coral reef to represent and analyze particular reef sites by incorporating an arbitrary number of coral species with differing live-history traits, algae, and site specific environmental conditions, such as temperature regimes and perturbations [52]. A basic model description and the first version of the model can be downloaded at sourceforge (http://sourceforge.net/projects/siccom/). Here we exemplify a benthic community of a typical Western Indian Ocean reef site with four different coral species, representing common functional types, algae, and characteristic environmental conditions. The simulation area (40 x 40 m) allows settling for all defined organism types and dynamics of the benthic reef community are simulated with monthly iterations (Fig. 2). Coral colonies and macroalgae are represented as individuals reacting situational to neighbors (e.g. competition) and environmental conditions (e.g. temperature-dependent bleaching) in accordance with their respective life history. Turf algae are represented by a grid with a cell size of 1 x 1 m and their density is measured as percentage cover.

    2. Corals

    We use four different coral species with contrasting life histories and differing responses to external influences (Table 1) – different growth rates, growth patterns, modes of reproduction, and recruitment, interactions with neighbors and sensitivities to extreme temperature events –, each of which represents trait combinations and functions common for many coral species. Porites lobata and P. lutea are massive corals which grow relatively slowly, feature a low susceptibility to bleaching [53] and high spatial competitiveness against other organisms [54]. Acropora muricata and Pocillopora damicornis are branching corals with fast growth rates, higher susceptibilities to bleaching and lower spatial competitiveness compared to the two massive coral species, above (see citations above). Coral bleaching and bleaching-related mortality indices were parameterized and calibrated with temperature data from Chumbe Island (Zanzibar, Tanzania) and the findings of [53]. A maximum temperature (100 % bleaching) and the specific bleaching and mortality indices of [53] were used to calculate specific, temperature-dependent probabilities for bleaching and bleaching-related mortality (Table 1). Coral reproduction is elaborated in Section 3.2.2 Recruitment loop.

    3. Algae

    Analogous to coral colonies, macroalgal thalli are represented as individual entities that grow, interact with neighbors and their environment and reproduce. For the simulations macroalgae are parameterized with data from Sargassum spp. known as a representative taxon in the region [55,56]. They are specified to grow 30 cm month-1, can reach a maximum height of 60 cm and a maximum diameter of 45 cm. The threshold height for fragmenting is set to 30 cm and for each produced fragment the height is reduced by 5 cm. Macroalgae recruit with a constant rate of 0.5 recruits m-2 year-1. Grazing is implemented as an average rate per month and is modeled to affect macroalgal individuals, which are then removed from the ongoing simulation. In contrast to macroalgae, turf algae are represented with 1 x 1 m patches whose percent cover increases by 20 % month-1 [57,58] until it reaches 100 % and is reduced by 50 % if it is grazed.

    4. Interactions

    In benthic reef systems spatial interactions are important processes determining the community composition and survival of individual organisms. In the model, the extension of a coral colony into a particular direction depends on the species, its growth form, its grade of being bleached and on the properties (species, growth form, bleach grade) of competing neighboring organisms (Table 2). Bleaching reduces the interaction strength of an affected colony with a linear factor from the onset of bleaching (100 % reduction) to complete recovery (0 % reduction) after 6 months. A coral colony dies if a substantial fraction of its covered space is overgrown by a larger colony or, if small, by a macroalga; i.e., if 50 % or 75 % are overgrown for branching and massive colonies, respectively. A recruit dies if it settles on another larger organism. If it settles on algal turf, its mortality is proportional to the percentage cover of the turf cell. Macroalgae can be overgrown by corals or die if they share more than 50 % of the same space with a larger macroalgal individual. While algal turf is displaced by coral colonies it can coexist with macroalgae.

    5. Environmental Conditions

    The virtual coral reef is subject to external influences, such as temperature and physical damage. Temperature data are taken from long-term observations (1997 2010) from Chumbe Island near Zanzibar [52]. In order to simulate longer time spans we concatenate yearly data sets in random order. Within the data set 1998 is treated as an extreme year because during that period an El Niño event struck the East African coast and observations of large scale bleaching were numerous [59–61]. Thus, the data set of 1998 is separated from the other yearly data and can be set to reoccur in designated intervals (15 years under standard conditions). A physical damage event can happen everywhere on the patch, causes death of all organisms within the affected area, except for turf algae and can occur with two different intensities. In the standard settings smaller disturbances represent anchor damage or destructive fishing, affect an area of 2 4 m in diameter and occur every year while larger disturbances represent damage by boats or destructive fishing as well, affect areas of 5 10 m in diameter and occur every 5 years. For a more detailed description of model processes and parameters see [52].