Many species are responding to global warming by shifting their distributions upslope to higher elevations, but the observed rates of shifts vary considerably among studies. Here, we test the hypothesis that this variation is in part explained by latitude, with tropical species being particularly responsive to warming temperatures. We analyze two independent empirical datasets—shifts in species’ elevational ranges, and changes in composition of forest inventory tree plots. Tropical species are tracking rising temperatures 2.1–2.4 times (range shift dataset) and 10 times (tree plot dataset) better than their temperate counterparts. Models predict that for a 100 m upslope shift in temperature isotherm, species at the equator have shifted their elevational ranges 93–96 m upslope, while species at 45° latitude have shifted only 37–42 m upslope. For tree plots, models predict that a 1°C increase in temperature leads to an increase in community temperature index (CTI), a metric of the average temperature optima of tree species within a plot, of 0.56°C at the equator but no change in CTI at 45° latitude (–0.033°C). This latitudinal gradient in temperature tracking suggests that tropical montane communities may be on an “escalator to extinction” as global temperatures continue to rise.

One-fifth of anthropogenic greenhouse gas emissions are sequestered by the terrestrial biosphere, where forests serve as an important "natural solution" to climate change (1). Forests are expected to persist as a substantial carbon sink, dampening future rises in atmospheric CO 2 levels (2-4). However, a significant part of this carbon uptake occurs in forests regrowing from past land-use changes or natural disturbances. A clear example is the landscape history of central New England in the United States, where significant transformations have occurred: from pristine old-growth forests to clear-cut areas giving way to agriculture, to old-field succession, and to expansion of regrowth forests ( Fig. 1). As more forests approach old-growth conditions, their rate of carbon uptake may begin to decline. To anticipate the future of the global carbon sink, Pugh et al. (5) show that it is necessary to account for forest regrowth and demography and to consider the broader issue of the terrestrial biosphere's ultimate capacity to sequester carbon.In PNAS, Pugh et al. (5) utilize a new global database of forest age to inform a vegetation model and find that regrowth forests constitute a carbon sink that is even greater than that of old-growth forests. Nearly half of the carbon uptake in regrowth forests, however, can be attributed to changes in forest demography instead of environmental change. Pugh et al. (5) demonstrate that this demographic approach estimates a greater global regrowth sink but a smaller tropical regrowth sink in comparison with the traditional land-use-change approach. Combining the unique and shared strengths of forest age and landuse datasets significantly improves our understanding of carbon sink estimates, especially for regrowth forests. Projecting into the future, Pugh et al. (5) calculate the total amount of carbon in live biomass that is missing in forests relative to a world in which forests were allowed to readjust to a business-as-usual disturbance rate. Their assessment suggests that the current forest carbon sink is largely transient in nature.This study has several implications for our understanding and predictions of the global carbon sink.First, Pugh et al. (5) highlight the importance of present-day forest regrowth in driving carbon sequestration. Many studies so far have attributed the recent increase in the forest carbon sink to environmental changes such as CO 2 fertilization, nitrogen deposition, and climate change (6, 7). Those findings suggest that forest growth may continue to buffer the negative impacts of anthropogenic carbon emissions. However, more researchers are starting to find that the forest carbon sink might be dominated by regrowth and driven by postdisturbance recovery (8,9). Using a global forest age dataset that informs disturbance history, Pugh et al. (5) show that the regrowth process alone drives about one-fourth of the carbon uptake, while the remainder is driven by environmental change. This is a great step forward in recognizing the key factors causing the chan...

Macrophytes are widely recognized for improving water quality and stabilizing the desirable clear‐water state in lakes. The positive effects of macrophytes on water quality have been noted to be weaker in the (sub)tropics compared to those of temperate regions. We conducted a global meta‐analysis using 47 studies that met our set criteria to assess the overall effects of macrophytes on water quality (measured by phytoplankton chlorophyll a concentration, total nitrogen concentration, total phosphorus concentration, Secchi depth and the trophic state index) and to investigate how these effects correlate with latitude using meta‐regressions. We also examined if the effects of macrophytes on lake‐water quality differ with growth form and study design in (sub)tropical and temperate areas by grouping the data and then comparing the effect sizes. We found that macrophytes significantly reduced phytoplankton chlorophyll a concentration, total nitrogen concentration, total phosphorus concentration, as well as the trophic state index, but they did not have a significant overall effect on Secchi depth. The effects of macrophytes on reducing phytoplankton chlorophyll a concentration, total nitrogen concentration and the trophic state index did not differ with latitude. However, the reduction of total phosphorus concentration was greater at lower latitudes. We showed that at lower latitudes, the positive effects of macrophytes on water quality are similar to or greater than those at higher latitudes, thus challenging the prevailing paradigm of macrophytes being less effective at enhancing lake‐water quality in the (sub)tropics. Furthermore, our data showed that the macrophyte effects vary by growth forms, and the growth forms that positively affect water quality differ between the (sub)tropical and temperate areas. We showed a lack of significant macrophyte effects in surveys within and outside macrophyte stands, suggesting difference in the sensitivities of study designs or possibly weaker effects of macrophytes in lakes compared to experimental settings.

No abstract