Approaches for Improving Estimation of Belowground Biomass of Mangrove Forests

Mangroves store and sequester large quantities of belowground biomass (BGB) in their root system. Given the important ecosystem services played by mangroves, particularly biomass storage and mitigation of climate change, interventions aimed to ensure that mangroves are safeguarded from deforestation and degradation through REDD+ and effective forest management initiatives are imperative. However, there are several challenges and uncertainties in the estimation of BGB of mangroves. Therefore, this study aimed to determine uncertainties in the estimation of BGB. The study hypothesised that root sampling techniques that do not involve complete root excavation are relatively inaccurate and hence tend to underestimate the BGB of mangroves. Besides, the study intended to provide step-by-step approaches for quantifying tree BGB. BGB data were generated using complete root excavation for Avicenia marina, Sonneratia alba, and Rhizophora mucronata in Mainland Tanzania to be used as reference data. The findings showed that all the local species-specific BGB models were statistically accurate in the prediction of tree BGB for the three species. Furthermore, results showed that all BGB models were constructed using data generated by means of the trench, incomplete root excavation, and pull-up methods under-estimated tree BGB by producing negative prediction errors (PEs) which were significantly different from zero at a 5% level of significance. The PEs ranged from -33% with relative precision (RP) of 68% (R. mucronata) to -82% (RP of 83%) for S. alba. Based on various sources of uncertainties in the estimation of BGB, the study also documented step-by-step approaches for quantifying BGB. This study concludes that there are large uncertainties in the estimation of BGB of mangroves. Models constructed using data generated by means of root sampling methods (i.e. root excavation, trench and pull-up) that do not entail complete root excavation are biased and hence tend to under-estimate tree BGB of mangroves. Similar results are observed at stand-level. It is recognised that, apart from methods for root sampling, uncertainties in the estimation of BGB may also be attributed to other sources of uncertainties e.g. BGB models, and the use of models beyond the data range. However, based on observed systematic under-estimation of BGB both at tree and stand-levels for estimates generated using data that do not entail complete root excavation, the study concludes that uncertainties in the estimation of BGB are largely attributable to root sampling methods applied rather than other sources of uncertainties. Consequently, the study rejected the null hypothesis in favour of the alternative hypothesis that root sampling that does not involve complete root excavation is relatively inaccurate and tends to underestimate the BGB of mangroves. The study also notes that the consequences of under-estimation of BGB of mangroves observed in this study are large and subsequently may lead to under-prioritization and valuation of mangroves in the overall forest management, planning, and decision-making. The step-by-step approaches for estimation of BGB established in this study aim to contribute to existing efforts with the objective of minimising associated uncertainties and improving existing knowledge on ecosystem services played by mangroves through biomass stored in their roots. The study further contributes to the accurate determination of belowground- and overall carbon stock of mangroves for accurate and informed decisions and policies both at national and international (e.g. REDD+) levels. This study recommends the step-by-step approaches for quantifying tree BGB of mangroves proposed in this study since they ensure complete root sampling and minimise uncertainties in the estimation of BGB of mangroves.