Ecometrics: A Trait-Based Approach to Paleoclimate and Paleoenvironmental Reconstruction

Leaf mass per area: An investigation into the application of the ubiquitous functional trait from a paleobotanical perspective

PREMISE

Leaf mass per area (LMA) is a widely used functional trait in both neobotanical and paleobotanical research that provides a window into how plants interact with their environment. Paleobotanists have used site-level measures of LMA as a proxy for climate, biome, deciduousness, and community-scale plant strategy, yet many of these relationships have not been grounded in modern data. In this study, we evaluated LMA from the paleobotanical perspective, seeking to add modern context to paleobotanical interpretations and discover what a combined modern and fossil data set can tell us about how LMA can be best applied toward interpreting plant communities.

METHODS

We built a modern data set by pulling plant trait data from the TRY database, and a fossil data set by compiling data from studies that have used the petiole-width proxy for LMA. We then investigated the relationships of species-mean, site-mean, and site-distribution LMA with different climatic, phylogenetic, and physiognomic variables.

RESULTS

We found that LMA distributions are correlated with climate, site taxonomic composition, and deciduousness. However, the relative contributions of these factors are not distinctive, and ultimately, LMA distributions cannot accurately reconstruct the biome or climate of an individual site.

CONCLUSIONS

The correlations that make up the leaf economics spectrum are stronger than the correlations between LMA and climate, phylogeny, morphospace, or depositional environment. Fossil LMA should be understood as the culmination of the influences of these variables rather than as a predictor.

Opportunities and Challenges in Applying AI to Evolutionary Morphology

Artificial intelligence (AI) is poised to revolutionize many aspects of science, including the study of evolutionary morphology. While classical AI methods such as principal component analysis and cluster analysis have been commonplace in the study of evolutionary morphology for decades, recent years have seen increasing application of deep learning to ecology and evolutionary biology. As digitized specimen databases become increasingly prevalent and openly available, AI is offering vast new potential to circumvent long-standing barriers to rapid, big data analysis of phenotypes. Here, we review the current state of AI methods available for the study of evolutionary morphology, which are most developed in the area of data acquisition and processing. We introduce the main available AI techniques, categorizing them into 3 stages based on their order of appearance: (1) machine learning, (2) deep learning, and (3) the most recent advancements in large-scale models and multimodal learning. Next, we present case studies of existing approaches using AI for evolutionary morphology, including image capture and segmentation, feature recognition, morphometrics, and phylogenetics. We then discuss the prospectus for near-term advances in specific areas of inquiry within this field, including the potential of new AI methods that have not yet been applied to the study of morphological evolution. In particular, we note key areas where AI remains underutilized and could be used to enhance studies of evolutionary morphology. This combination of current methods and potential developments has the capacity to transform the evolutionary analysis of the organismal phenotype into evolutionary phenomics, leading to an era of "big data" that aligns the study of phenotypes with genomics and other areas of bioinformatics.

Disruption of trait-environment relationships in African megafauna occurred in the middle Pleistocene

Mammalian megafauna have been critical to the functioning of Earth's biosphere for millions of years. However, since the Plio-Pleistocene, their biodiversity has declined concurrently with dramatic environmental change and hominin evolution. While these biodiversity declines are well-documented, their implications for the ecological function of megafaunal communities remain uncertain. Here, we adapt ecometric methods to evaluate whether the functional link between communities of herbivorous, eastern African megafauna and their environments (i.e., functional trait-environment relationships) was disrupted as biodiversity losses occurred over the past 7.4 Ma. Herbivore taxonomic and functional diversity began to decline during the Pliocene as open grassland habitats emerged, persisted, and expanded. In the mid-Pleistocene, grassland expansion intensified, and climates became more variable and arid. It was then that phylogenetic diversity declined, and the trait-environment relationships of herbivore communities shifted significantly. Our results divulge the varying implications of different losses in megafaunal biodiversity. Only the losses that occurred since the mid-Pleistocene were coincident with a disturbance to community ecological function. Prior diversity losses, conversely, occurred as the megafaunal species and trait pool narrowed towards those adapted to grassland environments.

Eco-ISEA3H, a machine learning ready spatial database for ecometric and species distribution modeling

We present the Eco-ISEA3H database, a compilation of global spatial data characterizing climate, geology, land cover, physical and human geography, and the geographic ranges of nearly 900 large mammalian species. The data are tailored for machine learning (ML)-based ecological modeling, and are intended primarily for continental- to global-scale ecometric and species distribution modeling. Such models are trained on present-day data and applied to the geologic past, or to future scenarios of climatic and environmental change. Model training requires integrated global datasets, describing species' occurrence and environment via consistent observational units. The Eco-ISEA3H database incorporates data from 17 sources, and includes 3,033 variables. The database is built on the Icosahedral Snyder Equal Area (ISEA) aperture 3 hexagonal (3H) discrete global grid system (DGGS), which partitions the Earth's surface into equal-area hexagonal cells. Source data were incorporated at six nested ISEA3H resolutions, using scripts developed and made available here. We demonstrate the utility of the database in a case study analyzing the bioclimatic envelopes of ten large, widely distributed mammalian species.

Do species factories exist? Detecting exceptional patterns of evolution in the mammalian fossil record

A species factory refers to the source that gives rise to an exceptionally large number of species. However, what is it exactly: a place, a time or a combination of places, times and environmental conditions, remains unclear. Here we search for species factories computationally, for which we develop statistical approaches to detect origination, extinction and sorting hotspots in space and time in the fossil record. Using data on European Late Cenozoic mammals, we analyse where, how and how often species factories occur, and how they potentially relate to the dynamics of environmental conditions. We find that in the Early Miocene origination hotspots tend to be located in areas with relatively low estimated net primary productivity. Our pilot study shows that species first occurring in origination hotspots tend to have a longer average longevity and a larger geographical range than other species, thus emphasizing the evolutionary importance of the species factories.

Comparison of environmental inference approaches for ecometric analyses: Using hypsodonty to estimate precipitation

Ecometrics is the study of community-level functional trait-environment relationships. We use ecometric analyses to estimate paleoenvironment and to investigate community-level functional changes through time.We evaluate four methods that have been used or have the potential to be used in ecometric analyses for estimating paleoenvironment to determine whether there have been systematic differences in paleoenvironmental estimation due to choice of the estimation method. Specifically, we evaluated linear regression, polynomial regression, nearest neighbor, and maximum-likelihood methods to explore the predictive ability of the relationship for a well-known ecometric dataset of mammalian herbivore hypsodonty metrics (molar tooth crown to root height ratio) and annual precipitation. Each method was applied to 43 Pleistocene fossil sites and compared to annual precipitation from global climate models. Sites were categorized as glacial or interglacial, and paleoprecipitation estimates were compared to the appropriate model.Estimation methods produce results that are highly correlated with log precipitation and estimates from the other methods (p < 0.001). Differences between estimated precipitation and observed precipitation are not significantly different across the four methods, but maximum likelihood produces the most accurate estimates of precipitation. When applied to paleontological sites, paleoprecipitation estimates align more closely with glacial global climate models than with interglacial models regardless of the age of the site.Each method has constraints that are important to consider when designing ecometric analyses to avoid misinterpretations when ecometric relationships are applied to the paleontological record. We show interglacial fauna estimates of paleoprecipitation more closely match glacial global climate models. This is likely because of the anthropogenic effects on community reassembly in the Holocene.

Feeding ecology has a stronger evolutionary influence on functional morphology than on body mass in mammals

Ecological specialization is a central driver of adaptive evolution. However, selective pressures may uniquely affect different ecomorphological traits (e.g., size and shape), complicating efforts to investigate the role of ecology in generating phenotypic diversity. Comparative studies can help remedy this issue by identifying specific relationships between ecologies and morphologies, thus elucidating functionally relevant traits. Jaw shape is a dietary correlate that offers considerable insight on mammalian evolution, but few studies have examined the influence of diet on jaw morphology across mammals. To this end, I apply phylogenetic comparative methods to mandibular measurements and dietary data for a diverse sample of mammals. Especially powerful predictors of diet are metrics that capture either the size of the angular process, which increases with greater herbivory, or the length of the posterior portion of the jaw, which decreases with greater herbivory. The size of the angular process likely reflects sizes of attached muscles that produce jaw movements needed to grind plant material. Further, I examine the impact of feeding ecology on body mass, an oft-used ecological surrogate in macroevolutionary studies. Although body mass commonly increases with evolutionary shifts to herbivory, it is outperformed by functional jaw morphology as a predictor of diet. Body mass is influenced by numerous factors beyond diet, and it may be evolutionarily labile relative to functional morphologies. This suggests that ecological diversification events may initially facilitate body mass diversification at smaller taxonomic and temporal scales, but sustained selective pressures will subsequently drive greater trait partitioning in functional morphologies.

Ecometric modelling of tooth shape and precipitation gradients among lemurs on Madagascar

Ecometric modelling relates spatial environmental variables to phenotypic characters to better understand morphological adaptation and help reconstruct past environments. Here, the community means of the dental topography metrics Dirichlet normal energy (DNE) and orientation patch count (OPC) are tested against annual precipitation and precipitation seasonality among lemurs across Madagascar. Dry, seasonal environments are expected to be associated with high DNE and OPC, as lemurs living in these environments are more likely to rely on tougher foods. Ecometric models are also used to calculate ecometric loads for lemur taxa hypothesized to be experiencing evolutionary disequilibria and to reconstruct annual precipitation and precipitation seasonality at the ~500 years BP subfossil cave site of Ankilitelo. DNE was highest in highly seasonal but wet environments. Seasonal exploitation of fallback foods and the availability of new leaves during wet periods may be most important in driving community DNE. OPC was weakly predicted by annual precipitation and seasonality but its distribution appeared to be driven by a stepwise increase in its community values in rainforest environments. The lemur fauna from Ankilitelo appears to resemble communities from moister environments than occur in the spiny desert zone in which the site is situated today.