Weak links between microclimate, plant traits, and decomposition at the landscape scale could be explained by absence of other known controls not accounted for in our analysis
Despite considerable variation in both environmental conditions (Figure 2c) and decomposition (Figure 3) across our study plots, the observed relationships between environmental conditions and decomposition were weak (Figure 4). This suggests that the topographical, vegetation, soil, and microclimate predictors we measured did not capture all ecologically relevant conditions and processes that control decomposition dynamics within this tundra landscape. For instance, we did not explicitly measure differences in decomposer communities and/or temporal activity. Comparable SEM studies from both tundra (Robinson et al. 2022) and other environments (e.g., Tresch et al. 2019, Xiao et al. 2019) identified decomposer community composition and activity as a key control on decomposition (but see Joly et al. 2023), and vegetation patterns along elevational gradients might not always reflect differences in microbial communities (Bryant et al. 2008, Fierer et al. 2011). Furthermore, soil organic matter content has been shown to drive abundances of soil microbial groups as well as decomposition rates in subarctic Iceland (Robinson et al. 2022). Also, longer-term soil moisture measurements might more accurately capture the variability in soil moisture conditions experienced by decomposers than our short (one growing season) time-series and hence could hold more power for predicting decomposition rates (Blok et al. 2010, Christiansen et al. 2017). Similarly, as environmental conditions in the Arctic tundra vary considerably between growing season and winter, complementing burial periods (e.g., growing season vs. winter vs. year-round) would help to disentangle potential seasonal differences in the controlling factors of Arctic decomposition dynamics, as well as differences in early vs. late-stage decomposition (Keuskamp et al. 2013, GarcĂ­a-Palacios et al. 2016, Canessa et al. 2021). We suggest that future studies account for these additional factors to model tundra plant litter decomposition more accurately.