Estimating Settlement carbon stock and density using an inventory approach and quantifying their variation by land use and parcel size

A lack of very-high resolution land-cover data and in-situ carbon sampling in Settlement areas has limited the quantification of terrestrial carbon in Canadian Settlements and elsewhere. Without those data, it is difficult to quantify Settlement area terrestrial carbon for United Nations Framework Convention on Climate Change reporting within the Land Use, Land Use Change, and Forestry sector. The presented research takes a step toward filling this gap by first classifying Settlement land cover at a very-high resolution (<=80 cm, 93% overall accuracy). Then, with those data, an inventory approach is used to estimate carbon stocks based on local data, standard rulesets from the IPCC, and practice-based assumptions. Guided by FAIR principles, our approach is operationalized and available in a Jupyter Notebook for distribution, use, and extension by others. Results found that the study Settlement comprised 30% tree cover and 18% turfgrass. When analyzed by parcel size, carbon densities varied little for parcels less than 1.6 ha (3.9–4.2 kg C m-2), but then increased with larger parcels up to 6.6 Gg C m-2 for parcels > 8.1 ha. Among different land uses, industrial, commercial, and transportation had the lowest carbon densities (2.4–2.8 kg C m-2), followed by high, medium, and low-density residential (3.6, 4.0, 8.9 kg C m-2, respectively) with low-density residential almost achieving carbon densities in protected and recreation areas (10.2 kg C m-2). Our results suggest that land use is a stronger driver of carbon-storage relative to parcel size, but their combination best represents the variation in carbon-storage in low-density residential land use. Currently, only carbon change in urban trees and deforestation around urban centres is reported in Canada’s National GHG Inventory Report. The present research quantifying carbon densities provides an analysis that could inform carbon change resulting from other land-use conversions and improve deforestation estimates by better defining the final state of a land-use change.