Search

Biomes are large‐scale ecosystems occupying large spaces. The biome concept should theoretically facilitate scientific synthesis of global‐scale studies of the past, present, and future biosphere. However, there is neither a consensus biome map nor universally accepted definition of terrestrial biomes, making joint interpretation and comparison of biome‐related studies difficult. "Desert," "rainforest," "tundra," "grassland," or "savanna," while widely used terms in common language, have multiple definitions and no universally accepted spatial distribution. Fit‐for‐purpose classification schemes are necessary, so multiple biome‐mapping methods should for now co‐exist. In this review, we compare biome‐mapping methods, first conceptually, then quantitatively. To facilitate the description of the diversity of approaches, we group the extant diversity of past, present, and future global‐scale biome‐mapping methods into three main families that differ by the feature captured, the mapping technique, and the nature of observation used: (1) compilation biome maps from expert elicitation, (2) functional biome maps from vegetation physiognomy, and (3) simulated biome maps from vegetation modeling. We design a protocol to measure and quantify spatially the pairwise agreement between biome maps. We then illustrate the use of such a protocol with a real‐world application by investigating the potential ecological drivers of disagreement between four broadly used, modern global biome maps. In this example, we quantify that the strongest disagreement among biome maps generally occurs in landscapes altered by human activities and moderately covered by vegetation. Such disagreements are sources of bias when combining several biome classifications. When aiming to produce realistic biome maps, biases could be minimized by promoting schemes using observations rather than predictions, while simultaneously considering the effect of humans and other ecosystem engineers in the definition. Throughout this review, we provide comparison and decision tools to navigate the diversity of approaches to encourage a more effective use of the biome concept. [ABSTRACT FROM AUTHOR]

The antiquity of human dispersal into Mediterranean islands and ensuing coastal adaptation have remained largely unexplored due to the prevailing assumption that the sea was a barrier to movement and that islands were hostile environments to early hunter-gatherers [J. F. Cherry, T. P. Leppard, J. Isl. Coast. Archaeol. 13, 191-205 (2018), 10.1080/15564894.2016.1276489]. Using the latest archaeological data, hindcasted climate projections, and age-structured demographic models, we demonstrate evidence for early arrival (14,257 to 13,182 calendar years ago) to Cyprus and predicted that large groups of people (~1,000 to 1,375) arrived in 2 to 3 main events occurring within <100 y to ensure low extinction risk. These results indicate that the postglacial settlement of Cyprus involved only a few large-scale, organized events requiring advanced watercraft technology. Our spatially debiased and Signor-Lipps-corrected estimates indicate rapid settlement of the island within <200 y, and expansion to a median of 4,000 to 5,000 people (0.36 to 0.46 km-2) in <11 human generations (<300 y). Our results do not support the hypothesis of inaccessible and inhospitable islands in the Mediterranean for pre-agropastoralists, agreeing with analogous conclusions for other parts of the world [M. I. Bird et al., Sci. Rep. 9, 8220 (2019), 10.1038/s41598-019-42946-9]. Our results also highlight the need to revisit these questions in the Mediterranean and test their validity with new technologies, field methods, and data. By applying stochastic models to the Mediterranean region, we can place Cyprus and large islands in general as attractive and favorable destinations for paleolithic peoples. [ABSTRACT FROM AUTHOR]

The dominant paradigm is that large tracts of Southeast Asia's lowland rainforests were replaced with a "savanna corridor" during the cooler, more seasonal climates of the Last Glacial Maximum (LGM) (23,000 to 19,000 y ago). This interpretation has implications for understanding the resilience of Asia's tropical forests to projected climate change, implying a vulnerability to "savannization". A savanna corridor is also an important foundation for archaeological interpretations of how humans moved through and settled insular Southeast Asia and Australia. Yet an up-to-date, multiproxy, and empirical examination of the palaeoecological evidence for this corridor is lacking. We conducted qualitative and statistical analyses of 59 palaeoecological records across Southeast Asia to test the evidence for LGM savannization and clarify the relationships between methods, biogeography, and ecological change in the region from the start of Late Glacial Period (119,000 y ago) to the present. The pollen records typically show montane forest persistence during the LGM, while δ13C biomarker proxies indicate the expansion of C4-rich grasslands. We reconcile this discrepancy by hypothesizing the expansion of montane forest in the uplands and replacement of rainforest with seasonally dry tropical forest in the lowlands. We also find that smooth forest transitions between 34,000 and 2,000 y ago point to the capacity of Southeast Asia's ecosystems both to resist and recover from climate stressors, suggesting resilience to savannization. Finally, the timing of ecological change observed in our combined datasets indicates an 'early' onset of the LGM in Southeast Asia from ~30,000 y ago. [ABSTRACT FROM AUTHOR]

The biosphere is changing rapidly due to human endeavour. Because ecological communities underlie networks of interacting species, changes that directly affect some species can have indirect effects on others. Accurate tools to predict these direct and indirect effects are therefore required to guide conservation strategies. However, most extinction‐risk studies only consider the direct effects of global change—such as predicting which species will breach their thermal limits under different warming scenarios—with predictions of trophic cascades and co‐extinction risks remaining mostly speculative. To predict the potential indirect effects of primary extinctions, data describing community interactions and network modelling can estimate how extinctions cascade through communities. While theoretical studies have demonstrated the usefulness of models in predicting how communities react to threats like climate change, few have applied such methods to real‐world communities. This gap partly reflects challenges in constructing trophic network models of real‐world food webs, highlighting the need to develop approaches for quantifying co‐extinction risk more accurately. We propose a framework for constructing ecological network models representing real‐world food webs in terrestrial ecosystems and subjecting these models to co‐extinction scenarios triggered by probable future environmental perturbations. Adopting our framework will improve estimates of how environmental perturbations affect whole ecological communities. Identifying species at risk of co‐extinction (or those that might trigger co‐extinctions) will also guide conservation interventions aiming to reduce the probability of co‐extinction cascades and additional species losses. [ABSTRACT FROM AUTHOR]

Species interactions play a fundamental role in ecosystems. However, few ecological communities have complete data describing such interactions, which is an obstacle to understanding how ecosystems function and respond to perturbations. Because it is often impractical to collect empirical data for all interactions in a community, various methods have been developed to infer interactions. Machine learning is increasingly being used for making interaction predictions, with random forest being one of the most frequently used of these methods. However, performance of random forest in inferring predator‐prey interactions in terrestrial vertebrates and its sensitivity to training data quality remain untested. We examined predator–prey interactions in two diverse, primarily terrestrial vertebrate classes: birds and mammals. Combining data from a global interaction dataset and a specific community (Simpson Desert, Australia), we tested how well random forest predicted predator–prey interactions for mammals and birds using species' ecomorphological and phylogenetic traits. We also tested how variation in training data quality – manipulated by removing records and switching interaction records to non‐interactions – affected model performance. We found that random forest could predict predator–prey interactions for birds and mammals using ecomorphological or phylogenetic traits, correctly predicting up to 88 and 67% of interactions and non‐interactions in the global and community‐specific datasets, respectively. These predictions were accurate even when there were no records in the training data for focal species. In contrast, false non‐interactions for focal predators in training data strongly degraded model performance. Our results demonstrate that random forest can identify predator–prey interactions for birds and mammals that have few or no interaction records. Furthermore, our study provides guidance on how to prepare training data to optimise machine learning classifiers for predicting species interactions, which could help ecologists 1) address knowledge gaps and explore network‐related questions in data‐poor situations, and 2) predict interactions for range‐expanding species. [ABSTRACT FROM AUTHOR]