POST-DISASTER RECOLONIZATION of MANGROVE FORESTS with A STOCHASTIC AGENT-BASED MODEL

Abstract

Mangrove forests in the Philippine coastline are susceptible to severe damage due to tropical storms. These mangrove forests provide a home for other plants and animals as well as providing resources for people living in coastal areas. Thus, it is important to promote proper conservation and judicious replanting in areas affected by storms. Since different species vary on their tolerance to physical conditions such as water salinity and soil composition, the appropriate genus must be used in reforestation efforts. This study aims to model the change in soil composition due to the introduction of a non-native species, Rhizophora mucronata, and restoring soil condition to aid recolonization of the existing native species, Avicennia and Sonneratia. The study uses an agent-based model for the prediction of the regenerative behaviour of mangrove stands consisting of the native species and the planted or non-native species in a fragmented habitat, with the use of spatio-temporal coloured noise to simulate stochastic seedling dispersal and subject to storm damage. The model uses Salmo and Juanico's model for mangrove growth. Stochastic experiments were carried out in a shoreline habitat with an existing native population of varying ages and a larger population of planted, non-native seedlings. The GIS data of Bangrin Marine Protected Area was used to simulate the recovery trajectory of the stand after typhoon Chan-hom of 2009.