The importance of definition and scale: Soil and vegetation carbon across an urban to rural gradient

Background/Question/Methods

Urban areas are growing in population, land area, and ecological significance; however, we are only beginning to understand how the process of urbanization influences ecosystem dynamics. The most obvious changes that accompany urbanization are increased impervious surface area and replacement of natural vegetation (or agricultural fields) with lawns and ornamental plants. These changes in land use and land cover will have major implications for regional and global biogeochemical cycling. In this study we have combined field-based measurements with remote sensing and demography to assess how changes in human and environmental factors influence carbon and nitrogen stocks across an urban to rural gradient extending westward from Boston, MA. We used a transect-based, stratified random sampling design that allowed us to evaluate our results using a diversity of urban definitions and spatial scales.  

Results/Conclusions

Our findings indicate that, on a per-area basis, residential land use contains considerable aboveground tree biomass (35% as much as forested areas). Commercial, industrial, and other developed areas contain significantly less biomass (14 % as much as forested areas). Mean aboveground biomass was 83.5 ± 8.2 Mg C/ha for the state of Massachusetts and 71.5 ± 7.7 Mg C/ha for the Boston Metropolitan Statistical Area (MSA). These biomass estimates reflect a high proportion of forest cover in the State and MSA. Developed land uses had similar soil C, N, and C:N regardless of whether the site was urban or rural. Forests, by contrast, had greater soil N concentrations and lower soil C:N in urban areas compared to rural areas. We reclassified our 140 research sites five times, using a different urban-rural classification scheme each time, and found that state- and MSA-wide estimates of biomass and soil C did not vary significantly based on the urban definition used to scale the data; however, the proportion of C attributed to areas classified as urban varied dramatically with definition. There is no question that future patterns of development will have significant implications for regional C and N stocks, however, our conclusions about the importance of urban areas in these trends are very sensitive to the definition of urban used by the investigator. Our findings highlight the importance of developing a more objective and consistent method for classifying urban areas that does not rely as heavily on political or census defined boundaries.