Modelling Impacts Of Climate Change On Tree Biomass And Distribution In Arabuko Sokoke Forest Reserve, Kenya

Abstract

The Arabuko Sokoke forest ecosystem is characterized by degradation from natural and anthropogenic drivers. Despite these challenges the forest has no study on how climate change will impact on forest biomass and species distribution. The main aim of this study was to project how climate change would impact on the tree biomass in the Arabuko Sokoke forest ecosystem. Experimental research design was used to determine the biomass accumulation rates of vegetation types as well as how climate change would impacts its distribution based on Representative Concentration Pathways (RCP) 4.5 and RCP 8.5 using MaxEnt model. Tree data was collected through direct measurement of Diameter at Breast Height (DBH), while species were identified and recorded together with the plot centre coordinates. The total tree biomass was calculated using allometric equation and shoot: root ratios. The historical rainfall, mean maximum and minimum temperature were collected from meteorological station in Malindi Airport, while the future climate data were downloaded from Worldclim data and downscaled to Arabuko Sokoke forest using geographical information system. The analyzed results for the study indicated that tree biomass accumulated in brachystegia (335MgC-ha), Mixed forest (164.5Mg C-ha) and Cynometra (92.1Mg C-ha) in the last 25 years. Statistical analysis confirms that the accumulation was significant (F1 68. =43.5, p=0.00). While the trend analysis for mean minimum temperature had significant trend (S = 338, p = 0.00). The results further show that the tree biomass related significantly with rainfall, maximum temperature and minimum temperatures (F1 67. =9.78, p=0.00, R2=0.55), (F1 67. =32.00, p=0.00, R2=0.55) and (F2 68. =40.27, p=0.00, R2=0.54) respectively. The MaxEnt model prediction based on RCP 4.5 and 8.5 indicated well the geographical distribution of brachystegia and mixed forest at 2050 and 2070 ( AUC= 0.80-0.90), while cynometra forest had a poor model fit (AUC=0.60-0.70). Based on jackknife test and analysis of variable contribution the mean annual, anomalies and extremes of precipitations and temperature had an impact on the predictive power of three models for Arabuko Sokoke. In conclusion the study findings indicated that tree biomass in Arabuko Sokoke has significantly accumulated over time. Secondly, the evidence provided by this study indicated that there was significant temperature and rainfall variability between 1990 and 2014 in Arabuko Sokoke forest. Thirdly, the finding indicates that rainfall and temperature significantly related with vi biomass across the forest landscape. Fourthly, the results shows that the site suitability for mixed and brachystegia forests can be predicted using Maxent Model based on general climate model scenarios of RCP 4.5 and RCP 8. Lastly, species distribution predictive model for Arabuko Sokoke was strongly influenced by annual trends, seasonality and extremities of temperature and rainfall parameters. Based on the findings, the study recommended that the forest managers consider development of strategies to deal with possible shift species and fundamental niche reduction for key species in Arabuko Sokoke forest. Secondly, communities are advised to diversify their sources of livelihoods and reduce their dependency on forest. Thirdly, carbon accounting systems and greenhouse gases systems should take into consideration carbon accumulation and possible impacts of climate change on tree biomass in Arabuko Sokoke and finally further research is recommended on species distribution modeling with inclusion of non-climatic parameters such as forest use pressure and natural forest disturbances.