Jonathan W. Chipman

Alpine zones above treeline in the Appalachian Mountains and adjacent ranges of northeastern North America support distinctive ecological communities that thrive where conditions are too harsh for trees to grow. These communities differ from alpine zones elsewhere on the continent and are a regionally important source of floristic biodiversity, but are subject to overlapping forces of environmental change including rising temperatures, CO ₂ fertilization, varying air quality and atmospheric nutrient deposition, declining snow cover, invasive species, and intense recreational use. Using 41 years of Landsat imagery, greening trends (represented by the normalized difference vegetation index, NDVI) and phenological cycles were modeled at 35 alpine sites in the Adirondacks, New England, and Quebec's Gaspé Peninsula. Most sites (69%, representing 88% of the region's alpine area) have experienced significant increases in alpine zone NDVI _max . Most sites also appear to have shifted towards earlier phenological peak greenness, but rates of greening are spatially and temporally heterogeneous and phenological shifts have not yet reached significance due to the sparse Landsat observational record. Notably, significant greening occurs disproportionately in the interior of alpine zones, not only where treeline is advancing. Greening trends are moderated by landscape position factors related to wind exposure and insolation and vary among seven alpine community types studied, with cliff, sedge meadow, and cushion‐tussock communities experiencing some of the largest changes. Methodological improvements such as an adaptive phenological model used for seasonality adjustment of observations reduced the noise level in estimated trends and effectively minimized the impact of observation frequency bias. Development of remote sensing monitoring programs for alpine zones will provide methodologically consistent regional‐scale information on how alpine ecosystems respond to environmental change, complementing the local‐scale work of field ecologists and better informing alpine zone stewardship and management activities.

Alpine zones in northeastern North America (NNA) are rare and support diverse ecological communities distinct from lower elevation forests. Global change drivers can alter the composition and functioning of these plant communities. The forest-alpine ecotone (treeline) is also influenced by a myriad of environmental drivers and is anticipated to encroach into alpine systems. We sought to determine the extent and drivers of treeline advance in NNA using multiple methodologies, including aerial and satellite-based imagery and long-term in-situ observation. For our approach, we (1) compared current and historical high-resolution aerial imagery of two ranges to quantify the advance of treeline over the last four decades. Vegetation delineation of aerial images were coupled with in-situ surveys to ground-truth treeline classifications. We used mixed effects models to examine the importance of both topographic and climatic variables on treeline advance. (2) Greening trends (NDVI) were modeled at 35 alpine and treeline ecotone sites in the Adirondacks (New York), New England, and Quebec’s Gaspé Peninsula using Landsat 5-8 imagery (1984-2024). (3) Permanent alpine vegetation point-intercept transects (with associated environmental data) in the Adirondacks and White Mountains (New Hampshire) were periodically sampled to quantify changes in alpine species composition and monitored for increased tree abundance. NNA treelines have shifted upslope on average by 3 m/decade since the 1970’s. Diffuse treelines (low tree density) displayed significantly greater upslope shifts (5 m/decade) compared to denser treelines, suggesting that both warming and demography are important correlates of treeline shifts. Topographical features (slope, aspect) as well as climate (accumulated growing degree days, AGDD) explained significant variation in the magnitude of treeline advance (R2 = 0.32). Most sites (88% total alpine area) have experienced significant greening via annual NDVImax. Greening occurs in the alpine zone interior in addition to some ecotone positions, potentially revealing both increased alpine vegetation productivity and tree establishment. Greening trends are moderated by landscape position and vary among plant community type (herbaceous vs. shrub-dominated). Non-metric multidimensional scaling (NMDS) illustrates an increase in shrub and tree species abundance over the past four decades, with a decrease in arctic specialist species abundance. These changes were highly positively correlated with site-level temperature and negatively correlated with anthropogenic atmospheric nitrogen (N) loading. Tree encroachment (and shrubbification) of NNA alpine challenges the future character and functioning of these rare systems. Development of remote sensing-based monitoring programs for NNA alpine will provide methodologically-consistent regional-scale information on how such ecosystems respond to environmental change, better informing stewardship and management activities.

Using street view data, in replace of remotely sensed (RS) data, to study the health impact of greenspace has become popular. However, direct comparisons of these two methods of measuring greenspace are still limited, and their findings are inconsistent. On the other hand, almost all studies of greenspace focus on urban areas. The effectiveness of greenspace in rural areas remains to be investigated. In this study, we compared measures of greenspace based on the Google Street View data with those based on RS data by calculating the correlation between the two and evaluating their associations with birth outcomes. Besides the direct measures of greenness, we also compared the measures of environmental diversity, calculated with the two types of data. Our study area consists of the States of New Hampshire and Vermont, USA, which are largely rural. Our results show that the correlations between the two types of greenness measures were weak to moderate, and the greenness at an eye‐level view largely reflects the immediate surroundings. Neither the street view data‐ nor the RS data‐based measures identify the influence of greenspace on birth outcomes in our rural study area. Interestingly, the environmental diversity was largely negatively associated with birth outcomes, particularly gestational age. Our study revealed that in rural areas, the effectiveness of greenspace and environmental diversity may be considerably different from that in urban areas.

Beneficial effects on health outcomes have been observed from exposure to spaces with substantial green vegetation (“greenspace”). This includes studies of greenspace exposure on birth outcomes; however, these have been conducted largely in urban regions. We characterized residential exposure to greenspace and land cover diversity during pregnancy in rural northern New England, USA, investigating whether variation in greenspace or diversity related to newborn outcomes. Five landscape variables (greenspace land cover, land cover diversity, impervious surface area, tree canopy cover, and the Normalized Difference Vegetation Index) were aggregated within six circular zones of radii from 100 to 3,000 m around residential addresses, and distance to conservation land was measured, providing a total of 31 greenspace and diversity metrics. Four birth outcomes along with potentially confounding variables were obtained from 1,440 participants in the New Hampshire Birth Cohort Study. Higher greenspace land cover up to 3,000 m was associated with larger newborn head circumference, while impervious surface area (non‐greenspace) had the opposite association. Further, birth length was positively associated with land cover diversity. These findings support beneficial health impacts of greenspace exposure observed in urban regions for certain health outcomes, such as newborn head circumference and length but not others such as birthweight and gestational age. Further our results indicate that larger radius buffer zones may be needed to characterize the rural landscape. Vegetation indices may not be interchangeable with other greenspace metrics such as land cover and impervious surface area in rural landscapes.

The magnitude of tropical cooling during the Last Glacial Maximum (LGM; ∼19–26.5 ka) remains controversial, with sea‐surface temperatures cooling by several degrees less than most temperatures reconstructed at high elevations. To explain this discrepancy, past studies proposed a steeper (increased) lapse rate—the temperature decrease with elevation—during the LGM relative to today. For instance, LGM temperatures in East Africa reconstructed from branched GDGTs from multiple elevations support an ∼0.9°C/km increase in the lapse rate during the LGM relative to present day. Lapse rates are a critical part of the Earth's climate sensitivity and atmospheric energy transfer, and it is vital to know whether and by how much the tropical lapse rate steepened during the LGM. Here, we simulate LGM glacier extents in the Rwenzori Mountains of Uganda with and without a change in lapse rate using a range of temperature and precipitation estimates. We find that the lapse rate must have been steeper than present for glaciers to reach their LGM positions using available sea‐level temperature and precipitation estimates for East Africa.

The Indo-Malaysian region is a hot spot of rapid land-cover and land-use change (LCLUC) with little consensus about the rates and magnitudes of such change. Here we use temporal convolutional neural networks (TempCNNs) to generate a spatiotemporally consistent LCLUC data set for nearly thirty-five years (1982–2015), validated against two reference data sets with over 80 percent accuracy, better than other LCLUC products for the region. Our results both confirm and complicate estimates from earlier work that relied on decadal, rather than interannual, changes in regional land cover. We find forests decrease in mainland and maritime Southeast Asia and increase in South China and South Asia. Consistent with geographic theory about the drivers of land-use change, we find cropland expansion is a driving force for deforestation in mainland Southeast Asia with savannas playing a superior role, suggesting widespread forest degradation in this region. In contrast to earlier work and theory, we find that South China’s increasing forest cover comes principally from savanna (rather than cropland) conversion. The explicit interannual LCLUC patterns, rates, and transitions identified in this study provide a valuable data source for studies of land-use theory, environmental and climate changes, and regional land-use policy evaluations.

There has been little rigorous investigation of the transferability of existing empirical water clarity models developed at one location or time to other lakes and dates of imagery with differing conditions. Machine learning methods have not been widely adopted for analysis of lake optical properties such as water clarity, despite their successful use in many other applications of environmental remote sensing. This study compares model performance for a random forest (RF) machine learning algorithm and a simple 4-band linear model with 13 previously published empirical non-machine learning algorithms. We use Landsat surface reflectance product data aligned with spatially and temporally co-located in situ Secchi depth observations from northeastern USA lakes over a 34-year period in this analysis. To evaluate the transferability of models across space and time, we compare model fit using the complete dataset (all images and samples) to a single-date approach, in which separate models are developed for each date of Landsat imagery with more than 75 field samples. On average, the single-date models for all algorithms had lower mean absolute errors (MAE) and root mean squared errors (RMSE) than the models fit to the complete dataset. The RF model had the highest pseudo-R² for the single-date approach as well as the complete dataset, suggesting that an RF approach outperforms traditional linear regression-based algorithms when modeling lake water clarity using satellite imagery.

Ephemeral rivers in arid regions act as linear oases, where corridors of vegetation supported by accessible groundwater and intermittent surface flows provide biological refugia in water-limited landscapes. The ecological and hydrological dynamics of these systems are poorly understood compared to perennial systems and subject to wide variation over space and time. This study used imagery obtained from an unmanned aerial vehicle (UAV) to enhance satellite data, which were then used to quantify change in woody vegetation cover along the ephemeral Kuiseb River in the Namib Desert over a 35-year period. Ultra-high resolution UAV imagery collected in 2016 was used to derive a model of fractional vegetation cover from five spectral vegetation indices, calculated from a contemporaneous Landsat 8 Operational Land Imager (OLI) image. The Normalized Difference Vegetation Index (NDVI) provided the linear best-fit relationship for calculating fractional cover; the model derived from the two 2016 datasets was subsequently applied to 24 intercalibrated Landsat images to calculate fractional vegetation cover for the Kuiseb extending back to 1984. Overall vegetation cover increased by 33% between 1984 and 2019, with the most highly vegetated reach of the river exhibiting the greatest positive change. This reach corresponds with the terminal alluvial zone, where most flood deposition occurs. The spatial and temporal trends discovered highlight the need for long-term monitoring of ephemeral ecosystems and demonstrate the efficacy of a multi-sensor approach to time series analysis using a UAV platform.