Carbon-dioxide-driven increase in foliage projective cover is not the same as increased woody plant density: lessons from an Australian tropical savanna

Carbon accounting in tropical savannas relies on a good understanding of the effects of atmospheric carbon dioxide (CO2) and land management on foliage projective cover (FPC) and vegetation structure. We used generalised additive modelling to track changes in Autumn Persistent Green (APG, a satellite-image-derived measure of FPC) in six vegetation types on Cape York Peninsula, Australia, over an 18-year period, and examined the influence of fire and grazing land tenure. We then used field monitoring and variography (analysis of spatial autocorrelation) in a smaller study area to determine whether changes in APG reflected vegetation structural change. APG increased through the 18-year study period and was significantly influenced by vegetation type, recent fire history and grazing land tenure. Residual year-on-year increases suggest CO2 fertilisation was the main driver of APG increase. APG was reduced by fires in the previous year, with early dry season fires having greater impact than late dry season fires, particularly in grassland and rainforest. This is consistent with leaves being most fire sensitive early in the year, when they are actively growing, than in the late dry season, when they are dormant. As seedlings and suckers would be particularly fire-sensitive, early fires may therefore be more effective than late fires at preventing woody encroachment. We demonstrated that variography provides a good indication of whether APG increases are caused by increases in FPC alone, or by an increase in tree density. We found support for increased woody plant density in grasslands, and that this increase was most pronounced on grazing lands. Conversely, we found no support for stem density increases in the dominant eucalypt woodland, despite APG increases being highest in this vegetation type. Hence, increases in FPC cannot always be equated to increases in woody biomass, and may occur in their absence. This conclusion has serious implications for global carbon accounting.

Automated summary

This study examined the changes in Autumn Persistent Green (APG) in six vegetation types on Cape York Peninsula, Australia, over an 18-year period. The results showed that APG increases were mainly driven by increased carbon dioxide concentration, while fire and grazing also had an impact. The study also demonstrated the importance of considering vegetation structure when interpreting APG increases, as increases in foliage projective cover do not always equate to increases in woody biomass.