Continuous and discontinuous variation in ecosystem carbon stocks with elevation across a treeline ecotone

Abstract

Abstract. Treelines differentiate vastly contrasting ecosystems: open tundra from closed forest. Treeline advance has implications for the climate system due to the impact of the transition from tundra to forest ecosystem on carbon (C) storage and albedo. Treeline advance has been seen to increase above-ground C stocks as low vegetation is replaced with trees, but decrease organic soil C stocks as old carbon is decomposed. However, studies comparing across the treeline typically do not account for elevational variation within the ecotone. Here we sample ecosystem C stocks along an elevational gradient (970 to 1300 m), incorporating a large-scale and long-term livestock grazing experiment, in the Southern Norwegian mountains. We investigate whether there are continuous or discontinuous changes in C storage across the treeline ecotone, and whether these are modulated by grazing. We find that vegetation C stock decreases with elevation, with a clear breakpoint between the forest line and treeline above which the vegetation C stock is constant. In contrast, C stocks in organic surface horizons of the soil increase linearly with elevation within the study's elevational range, whereas C stocks in mineral soil horizons are unrelated to elevation. Total ecosystem C stocks also showed a discontinuous elevational pattern, increasing with elevation above the treeline (8 g m−2 m−1 increase in elevation), but decreasing with elevation below the forest line (−15 g m−2 m−1 increase in elevation), such that ecosystem C storage reaches a minimum between the forest line and treeline. We did not find any effect of short-term (12 years) grazing on the elevational patterns. Our findings demonstrate that patterns of C storage across the treeline are complex, and should be taken account of when estimating ecosystem C storage with shifting treelines.

Soil carbon stocks in forest-tundra ecotones along a 500 km latitudinal gradient in northern Norway

“…Our large-scale dataset demonstrates that SOC stocks do not always decrease from the forest to the tundra as has previously been reported for surface soils in mountain birch forest-tundra ecotones 6,[25][26][27] . The contrasting results between different study sites strongly suggest that changes in SOC stock sizes across the treeline depend on the extent to which local environmental conditions alter the balance between litter input and soil respiration.…”

“…The contrasting results between different study sites strongly suggest that changes in SOC stock sizes across the treeline depend on the extent to which local environmental conditions alter the balance between litter input and soil respiration. It should be noted that we reported SOC stocks for the entire OA complex, while earlier work gave estimates for individual soil layers 12,[25][26][27] . Thus, direct comparisons of SOC stock sizes with previous studies should be done with caution because of methodological differences.…”

“…First, carbon cycle models predict that the decrease in aboveground biomass and litter fall along the transition from forest into generally low-productive tundra results in smaller tundra SOC pools 44 . Second, previous field studies in the Scandes mountains suggest that SOC stocks in the organic horizon are smaller www.nature.com/scientificreports/ beneath mountain birch forest soils than beneath tundra soils due to higher soil respiration rates 12,[25][26][27] . One commonly proposed explanation is that shrubs and trees trap more snow, which insulates the soil and thereby increases microbial activity 45,46 .…”

“…For instance, SOC stocks will be impacted by direct effects of warming on decomposition 12 . Also, the response of SOC dynamics to treeline shifts is likely to change as the forest matures 26 . However, since we sampled 14 study sites, of which half had similar SOC stock sizes in forest and tundra soils, our study provides strong evidence that a shifting treeline does not necessarily lead to a change in SOC storage at all sites.…”

Long‐term P weathering and recent N deposition control contemporary plant‐soil C, N, and P

“…Site data locations and distributions of mean annual temperature (MAT), soil pH, nitrogen deposition in year 2000, and observed organic topsoil carbon, nitrogen, and phosphorus contents. Data are detailed in the SI, Tables S3–S5 [ Andersson et al , ; Beier et al , ; Berggren et al , ; Blaser et al , ; Bloemerts and Vries , ; Corre and Lamersdorf , ; de Vries et al , ; UKREATE , 2010; DEFRA , 2006; B A Emmett et al , ; Evans et al , ; Gundersen , ; Jackson‐Blake et al , ; Kleja et al , ; Kreutzer et al , ; Lamersdorf and Meyer , ; MacDonald et al , ; Mol‐Dijkstra and Kros , ; Parr et al , ; Pilkington et al , ; I K Schmidt et al , ; Sjøeng et al , ; J D Speed et al , ; J D M Speed et al , ; Tietema et al , ; Tipping et al , ; van Meeteren et al , ; von Oheimb et al , ; Walthert et al , ; Wright et al , ; Zimmermann et al , ].…”