Combination of ground and remote sensing data to assess carbon stock changes in the main urban park of Florence

Abstract

Optical and LiDAR datasets taken by airborne platforms are informative on important forest attributes, and particularly on growing stock volume (GSV), which is related to woody biomass and carbon storage. This paper presents the integration of such datasets with information collected on the ground for assessing the GSV variation that occurred in the Cascine urban park (Florence, Italy) mostly as a consequence of an extreme wind storm in March 2015. Two LiDAR acquisitions taken before and after the disastrous event (2007 and 2017) were combined with conventional forest observations derived from a complete inventory of the park area and from restricted ground surveys conducted around the first and second study dates. The dataset of the first period was used to evaluate the applicability of an area-based estimation approach to the second dataset and consequently assess the magnitude of the GSV change. In particular, the change was statistically characterized based only on the recently collected ground samples and on the combination of these with LiDAR data through a ratio estimator. The combination of the LiDAR and ground data increased the precision of the estimates obtained, highlighting a significant GSV decrease during the study decade (about 20%), that was concentrated in two of four park sectors. This GSV decrease, which can be mostly attributed to the effect of the 2015 wind storm, corresponds to a total loss of carbon stored in the park of around 2700 tons.