A Space-Time Stochastics Model for Analysing Hydrological Feedback of Tropical Peatland
Utriweni Mukhaiyar, Tarasinta Prastoro, Adilan W. Mahdiyasa*

Faculty of Mathematics and Natural Sciences, Bandung Institute of Technology
Jalan Ganesha 10, Bandung 40132, Indonesia
*adilan[at]math.itb.ac.id

Abstract

Indonesian peatlands serve as one of the most efficient carbon sinks and are one of the largest near-surface reserves for terrestrial organic carbon. The total area of tropical peatlands in Indonesia is around 13.43 million ha or equivalent to 7% of Indonesia land area, and they store about 57 gigatonnes of carbon Gt C. Tropical peatlands are highly significant to global efforts to combat climate change, as well as broader sustainable development goals. The protection and restoration of tropical peatlands are vital in the transition toward a low-carbon and circular economy for Indonesia. However, tropical peatlands are complex systems with the potential to shift dramatically between equilibrium states in response to hydrological changes. One approach to understanding this complex behaviour is through the space-time stochastics model that provides a prediction on how the water table position evolves on a wide range of timeframes. We estimate the water table position of Indonesian peatlands using Generalized Space-Time Autoregressive (GSTAR) and analyse the implication of water table fluctuation on the peatland carbon stock and resilience. This approach would provide an early warning system that can be used as a reliable indicator of catastrophic events, for instance, peat fire that commonly occurs in Indonesia.

Keywords: Generalized Space-Time Autoregressive, stochastics model, tropical peatland hydrology, carbon balance

Topic: MATHEMATICS AND STATISTICS