Megan Marie Seeley

Forest ecosystems worldwide are increasingly threatened by biological invasions and non-native pathogens, with consequences for forest composition, ecosystem function, and management. Compared to continental ecosystems, the Hawaiian Islands have been disproportionately impacted by invasive species, with many island native species now threatened or endangered. In Hawaiʻi, both the invasive tree Psidium cattleianum and non-native pathogenic Ceratocystis fungi threaten native Metrosideros polymorpha forests. To assess the spatial and temporal relationships between these stressors, we used high-resolution airborne imaging spectroscopy and Sentinel-2 multispectral imagery to map M. polymorpha forest and track P. cattleianum expansion from 2019 to 2025 across the North Hilo and Hāmākua districts of Hawai’i Island. We then compared patterns of P. cattleianum spread with the distribution of Rapid ʻŌhiʻa Death (ROD)-affected canopies, a disease complex caused by two Ceratocystis species. We found that invasion risk was ~18% higher in regions near existing P. cattleianum populations, indicating strong dispersal limitation. The presence of ROD further elevated invasion risk 11.9% in these areas. Generalized additive models revealed a significant interaction between proximity to existing P. cattleianum populations and ROD occurrence, with ROD increasing invasion probability. These results suggest that P. cattleianum expansion is constrained by dispersal, and ROD mortality events create canopy gaps that elevate light and nutrient availability, thereby facilitating invasion near existing P. cattleianum populations. More broadly, our findings demonstrate how multiple biological stressors can accelerate transitions from native-dominated to non-native-dominated forest states, highlighting the importance of considering invasive species and disease dynamics simultaneously in forest conservation and management.

Globally, vegetation biodiversity is expected to decline as the rate of plant adaptation struggles to keep pace with rising temperatures. To support conservation efforts through remote sensing, we disentangled the nested effects of genetic and environmental influences on reflectance spectra, leveraging spectroscopy to assess plant adaptations to temperature. Specifically, we quantified the relative effect of plasticity and heritability on Populus fremontii (Fremont cottonwood) leaf reflectance using clonal replicates propagated from 16 populations and grown across three common gardens spanning a mean annual temperature gradient representing the thermal range of P. fremontii. We used variance partitioning to decompose phenotypic variation expressed in the leaf spectra into genotypic and environmental components to estimate broad-sense heritability. Heritability was strongly expressed in the spectral red edge (~ 680–750 nm) and shortwave infrared (~ 1400–3000 nm), though the heritability peak in the red edge was sensitive to extreme temperatures. By comparing distances of group centroids in principal component space, we determined that P. fremontii intraspecific spectral variation was shaped by the interaction between common garden site conditions and source population. Support vector machine models indicated pronounced environmental influence on spectral variation, as P. fremontii source population and garden location were classified at 71.8% and 92.6% accuracy, respectively. These findings emphasize the utility of reflectance data in separating genetic and environmental influences on plant phenotypes, offering a pathway to scale these insights across broader landscapes and aid in the conservation and management of vulnerable ecosystems in a warming climate.

Understanding how vegetation responds to drought is fundamental for understanding the broader implications of climate change on foundation tree species that support high biodiversity. Leveraging remote sensing technology provides a unique vantage point to explore these responses across and within species. We investigated interspecific drought responses of two Populus species (P. fremontii, P. angustifolia) and their naturally occurring hybrids using leaf‐level visible through shortwave infrared (VSWIR; 400–2500 nm) reflectance. As F₁ hybrids backcross with either species, resulting in a range of backcross genotypes, we heretofore refer to the two species and their hybrids collectively as ‘cross types’. We additionally explored intraspecific variation in P. fremontii drought response at the leaf and canopy levels using reflectance data and thermal unmanned aerial vehicle (UAV) imagery. We employed several analyses to assess genotype‐by‐environment (G × E) interactions concerning drought, including principal component analysis, support vector machine and spectral similarity index. Five key findings emerged: (1) Spectra of all three cross types shifted significantly in response to drought. The magnitude of these reaction norms can be ranked from hybrids>P. fremontii>P. angustifolia, suggesting differential variation in response to drought; (2) Spectral space among cross types constricted under drought, indicating spectral—and phenotypic—convergence; (3) Experimentally, populations of P. fremontii from cool regions had different responses to drought than populations from warm regions, with source population mean annual temperature driving the magnitude and direction of change in VSWIR reflectance. (4) UAV thermal imagery revealed that watered, warm‐adapted populations maintained lower leaf temperatures and retained more leaves than cool‐adapted populations, but differences in leaf retention decreased when droughted. (5) These findings are consistent with patterns of local adaptation to drought and temperature stress, demonstrating the ability of leaf spectra to detect ecological and evolutionary responses to drought as a function of adaptation to different environments. Synthesis. Leaf‐level spectroscopy and canopy‐level UAV thermal data captured inter‐ and intraspecific responses to water stress in cottonwoods, which are widely distributed in arid environments. This study demonstrates the potential of remote sensing to monitor and predict the impacts of drought on scales varying from leaves to landscapes.

Understanding how vegetation responds to drought is fundamental for understanding the broader implications of climate change on foundation tree species that support high biodiversity. Leveraging remote sensing technology provides a unique vantage point to explore these responses across and within species. We investigated interspecific drought responses of two Populus species ( P . fremontii , P . angustifolia ) and their naturally occurring hybrids using leaf-level visible through shortwave infrared (VSWIR; 400-2500 nm) reflectance. As F ₁ hybrids backcross with either species, resulting in a range of backcross genotypes, we heretofore refer to the two species and their hybrids collectively as “cross types.” We additionally explored intraspecific variation in P. fremontii drought response at the leaf and canopy levels using reflectance data and thermal unmanned aerial vehicle (UAV) imagery. We employed several analyses to assess genotype-by-environment (GxE) interactions concerning drought, including principal component analysis, support vector machine, and spectral similarity index. Five key findings emerged: (1) Spectra of all three cross types shifted significantly in response to drought. The magnitude of these reaction norms can be ranked from hybrids> P. fremontii > P. angustifolia, suggesting differential variation in response to drought; (2) Spectral space among cross types constricted under drought, indicating spectral—and phenotypic—convergence; (3) Experimentally, populations of P. fremontii from cool regions had different responses to drought than populations from warm regions, with source population mean annual temperature driving the magnitude and direction of change in VSWIR reflectance. (4) UAV thermal imagery revealed that watered, warm-adapted populations maintained lower leaf temperatures and retained more leaves than cool-adapted populations, but differences in leaf retention decreased when droughted. (5) These findings are consistent with patterns of local adaptation to drought and temperature stress, demonstrating the ability of leaf spectra to detect ecological and evolutionary responses to drought as a function of adaptation to different environments. Synthesis. Leaf-level spectroscopy and canopy-level UAV thermal data captured inter- and intraspecific responses to water stress in cottonwoods, which are widely distributed in arid environments. This study demonstrates the potential of remote sensing to monitor and predict the impacts of drought on scales varying from leaves to landscapes.

Globally, vegetation biodiversity is expected to decline as the rate of plant adaptation struggles to keep pace with rising temperatures. To support conservation efforts through remote sensing, we disentangled the nested effects of genetic and environmental influences on reflectance spectra, leveraging spectroscopy to assess plant adaptations to temperature. Specifically, we quantified the relative effect of plasticity and heritability on Populus fremontii (Fremont cottonwood) leaf reflectance using clonal replicates propagated from 16 populations and grown across three common gardens spanning a mean annual temperature gradient representing the thermal range of P. fremontii . We used variance partitioning to decompose phenotypic variation expressed in the leaf spectra into genotypic and environmental components to estimate broad-sense heritability. Heritability was strongly expressed in the spectral red edge (∼680-750nm) and shortwave infrared (∼1400-3000nm), though the heritability peak in the red edge was sensitive to extreme temperatures. By comparing distances of group centroids in principal component space, we determined that P. fremontii intraspecific spectral variation was shaped by the interaction between common garden site conditions and source population. Support vector machine models indicated pronounced environmental influence on spectral variation, as P. fremontii source population and garden location were classified at 71.8% and 92.6% accuracy, respectively. These findings emphasize the utility of reflectance data in separating genetic and environmental influences on plant phenotypes, offering a pathway to scale these insights across broader landscapes and aid in the conservation and management of vulnerable ecosystems in a warming climate.

Vegetation species mapping using airborne imaging spectroscopy yields accurate results and is important for advancing conservation objectives and biogeographic studies. As these data become more readily available owing to the upcoming launch of spaceborne imaging spectrometers, it is necessary to understand how these data can be used to consistently classify species across large geographic scales. However, few studies have attempted to map species across multiple ecosystems; therefore, little is known regarding the effect of intra-specific variation on the mapping of a single species across a wide range of environments and among varying backgrounds of other non-target species. To explore this effect, we developed and tested species classification models for Metrosideros polymorpha, a highly polymorphic canopy species endemic to Hawai’i, which is found in a diverse array of ecosystems. We compared the accuracies of support vector machine (SVM) and random forest models trained on canopy reflectance data from each of eight distinct ecosystems (ecosystem-specific) and a universal model trained on data from all ecosystems. When applied to ecosystem-specific test datasets, the ecosystem-specific models outperformed the universal model; however, the universal model retained high (>81%) accuracies across all ecosystems. Additionally, we found that models from ecosystems with broad variation in M. polymorpha canopy traits, as estimated using chemometric equations applied to canopy spectra, accurately predicted M. polymorpha in other ecosystems. While species classifications across ecosystems can yield accurate results, these results will require sampling procedures that capture the intra-specific variation of the target species.

The role of intraspecific trait variation in functional ecology has gained traction in recent years as many papers have observed its importance in driving community diversity and ecology. Yet much of the work in this field relies on field-based trait surveys. Here, we used continuous canopy trait information derived from remote sensing data of a highly polymorphic tree species, Metrosideros polymorpha, to quantify environmental controls on intraspecific trait variation. M. polymorpha, an endemic, keystone tree species in Hawai’i, varies morphologically, chemically, and genetically across broad elevation and soil substrate age gradients, making it an ideal model organism to explore large-scale environmental drivers of intraspecific trait variation. M. polymorpha canopy reflectance (visible to shortwave infrared; 380–2510 nm) and light detection and ranging (LiDAR) data collected by the Global Airborne Observatory were modeled to canopy trait estimates of leaf mass per area, chlorophyll a and b, carotenoids, total carbon, nitrogen, phosphorus, phenols, cellulose, and top of canopy height using previously developed leaf chemometric equations. We explored how these derived traits varied across environmental gradients by extracting elevation, slope, aspect, precipitation, and soil substrate age data at canopy locations. We then obtained the feature importance values of the environmental factors in predicting each leaf trait by training random forest models to predict leaf traits individually. Of these environmental factors, elevation was the most important predictor for all canopy traits. Elevation not only affected canopy traits directly but also indirectly by influencing the relationships between soil substrate age and canopy traits as well as between nitrogen and other traits, as indicated by the change in slope between the variables at different elevation ranges. In conclusion, intraspecific variation in M. polymorpha traits derived from remote sensing adheres to known leaf economic spectrum (LES) patterns as well as interspecific LES traits previously mapped using imaging spectroscopy.

Vegetation classifications on large geographic scales are necessary to inform conservation decisions and monitor keystone, invasive, and endangered species. These classifications are often effectively achieved by applying models to imaging spectroscopy, a type of remote sensing data, but such undertakings are often limited in spatial extent. Here we provide accurate, high-resolution spatial data on the keystone species Metrosideros polymorpha, a highly polymorphic tree species distributed across bioclimatic zones and environmental gradients on Hawai’i Island using airborne imaging spectroscopy and LiDAR. We compare two tree species classification techniques, the support vector machine (SVM) and spectral mixture analysis (SMA), to assess their ability to map M. polymorpha over 28,000 square kilometers where differences in topography, background vegetation, sun angle relative to the aircraft, and day of data collection, among others, challenge accurate classification. To capture spatial variability in model performance, we applied Gaussian process classification (GPC) to estimate the spatial probability density of M. polymorpha occurrence using only training sample locations. We found that while SVM and SMA models exhibit similar raw score accuracy over the test set (96.0% and 93.4%, respectively), SVM better reproduces the spatial distribution of M. polymorpha than SMA. We developed a final 2 m × 2 m M. polymorpha presence dataset and a 30 m × 30 m M. polymorpha density dataset using SVM classifications that have been made publicly available for use in conservation applications. Accurate, large-scale species classifications are achievable, but metrics for model performance assessments must account for spatial variation of model accuracy.

Despite recent calls to make geographical analyses more reproducible, formal attempts to reproduce or replicate published work remain largely absent from the geographic literature. The reproductions of geographic research that do exist typically focus on computational reproducibility—whether results can be recreated using data and code provided by the authors—rather than on evaluating the conclusion and internal validity and evidential value of the original analysis. However, knowing if a study is computationally reproducible is insufficient if the goal of a reproduction is to identify and correct errors in our knowledge. We argue that reproductions of geographic work should focus on assessing whether the findings and claims made in existing empirical studies are well supported by the evidence presented. We aim to facilitate this transition by introducing a model framework for conducting reproduction studies, demonstrating its use, and reporting the findings of three exemplar studies. We present three model reproductions of geographical analyses of COVID‐19 based on a common, open access template. Each reproduction attempt is published as an open access repository, complete with pre‐analysis plan, data, code, and final report. We find each study to be partially reproducible, but moving past computational reproducibility, our assessments reveal conceptual and methodological concerns that raise questions about the predictive value and the magnitude of the associations presented in each study. Collectively, these reproductions and our template materials offer a practical framework others can use to reproduce and replicate empirical spatial analyses and ultimately facilitate the identification and correction of errors in the geographic literature.

Vegetation classifications on large geographic scales are necessary to inform conservation decisions and monitor keystone, invasive, and endangered species. These classifications are often effectively achieved by applying models to imaging spectroscopy, a type of remote sensing, data, but such undertakings are often limited in spatial extent. Here we provide accurate, high-resolution spatial data on the keystone species Metrosideros polymorpha, a highly polymorphic tree species distributed across bioclimatic zones and environmental gradients on Hawai'i island, using airborne imaging spectroscopy and LiDAR. We compare two tree species classification techniques, support vector machine (SVM) and spectral mixture analysis (SMA), to assess their ability to map M. polymorpha over 28,000 square kilometers where differences in topography, background vegetation, sun angle relative to the aircraft, and day of data collection, among others, challenge accurate classification. To capture spatial variability in model performance, we applied gaussian process classification (GPC) to estimate the spatial probability density of M. polymorpha occurrence using only training sample locations. We found that, while SVM and SMA models exhibit similar raw score accuracy over the test set (96.0% and 93.4%, respectively), SVM better reproduces the spatial distribution of M. polymorpha than SMA. We developed a final 2 m x 2 m M. polymorpha presence dataset and a 30 m x 30 m M. polymorpha density dataset using SVM classifications that have been made publicly available for use in conservation applications. Accurate, large-scale species classifications are achievable, but metrics for model performance assessments must account for spatial variation of model accuracy.