Tropical forests are a major component of the global carbon cycle and home to two‐thirds of terrestrial species. Upper‐canopy trees store the majority of forest carbon and can be vulnerable to drought events and storms. Monitoring their growth and mortality is essential to understanding forest resilience to climate change, but in the context of forest carbon storage, large trees are underrepresented in traditional field surveys, so estimates are poorly constrained. Aerial photographs provide spectral and textural information to discriminate between tree crowns in diverse, complex tropical canopies, potentially opening the door to landscape monitoring of large trees. Here we describe a new deep convolutional neural network method, Detectree2, which builds on the Mask R‐CNN computer vision framework to recognize the irregular edges of individual tree crowns from airborne RGB imagery. We trained and evaluated this model with 3797 manually delineated tree crowns at three sites in Malaysian Borneo and one site in French Guiana. As an example application, we combined the delineations with repeat lidar surveys (taken between 3 and 6 years apart) of the four sites to estimate the growth and mortality of upper‐canopy trees. Detectree2 delineated 65 000 upper‐canopy trees across 14 km² of aerial images. The skill of the automatic method in delineating unseen test trees was good (F₁ score = 0.64) and for the tallest category of trees was excellent (F₁ score = 0.74). As predicted from previous field studies, we found that growth rate declined with tree height and tall trees had higher mortality rates than intermediate‐size trees. Our approach demonstrates that deep learning methods can automatically segment trees in widely accessible RGB imagery. This tool (provided as an open‐source Python package) has many potential applications in forest ecology and conservation, from estimating carbon stocks to monitoring forest phenology and restoration.Python package available to install at https://github.com/PatBall1/Detectree2.

The detection and quantification of volcanic ash is extremely important to the aviation industry, civil defense organizations, and those in peril from volcanic ashfall. To exploit the remote sensing techniques that are used to monitor a volcanic cloud and return information on its properties, the effective complex refractive index of the volcanic ash is required. This paper presents the complex refractive index determined in the laboratory at 450.0 nm, 546.7 nm, and 650.0 nm for volcanic ash samples from eruptions of Aso (Japan), Grímsvötn (Iceland), Chaitén (Chile), Etna (Italy), Eyjafjallajökull (Iceland), Tongariro (New Zealand), Askja (Iceland), Nisyros (Greece), Okmok (Alaska), Augustine (Alaska), and Spurr (Alaska). The Becke line method was used to measure the real part of the refractive index with an accuracy of 0.01. The values measured differed between eruptions and were in the range 1.51–1.63 at 450.0 nm, 1.50–1.61 at 546.7 nm, and 1.50–1.59 at 650.0 nm. A novel method is introduced to derive the imaginary part of the refractive index from the attenuation of light by ash. The method has a precision in the range 10⁻³–10⁻⁴. The values for the ash imaginary refractive index ranged 0.22–1.70 × 10⁻³at 450.0 nm, 0.16–1.93 × 10⁻³at 546.7 nm, and 0.15–2.08 × 10⁻³at 650.0 nm. The accuracy of Becke and attenuation methods was assessed by measuring the complex refractive index of Hoya neutral density glass and found to have an accuracy of <0.01 and <2 × 10⁻⁵for the real and imaginary parts of the refractive index, respectively.

Tropical forests are subject to diverse deforestation pressures while their conservation is essential to achieve global climate goals. Predicting the location of deforestation is challenging due to the complexity of the natural and human systems involved but accurate and timely forecasts could enable effective planning and on‐the‐ground enforcement practices to curb deforestation rates. New computer vision technologies based on deep learning can be applied to the increasing volume of Earth observation data to generate novel insights and make predictions with unprecedented accuracy. Here, we demonstrate the ability of deep convolutional neural networks (CNNs) to learn spatiotemporal patterns of deforestation from a limited set of freely available global data layers, including multispectral satellite imagery, the Hansen maps of annual forest change (2001–2020) and the ALOS PALSAR digital surface model, to forecast deforestation (2021). We designed four model architectures, based on 2D CNNs, 3D CNNs, and Convolutional Long Short‐Term Memory (ConvLSTM) Recurrent Neural Networks (RNNs), to produce spatial maps that indicate the risk to each forested pixel (~30 m) in the landscape of becoming deforested within the next year. They were trained and tested on data from two ~80,000 km² tropical forest regions in the Southern Peruvian Amazon. The networks could predict the location of future forest loss to a high degree of accuracy (F₁ = 0.58–0.71). Our best performing model (3D CNN) had the highest pixel‐wise accuracy (F₁ = 0.71) when validated on 2020 forest loss (2014–2019 training). Visual interpretation of the mapped forecasts indicated that the network could automatically discern the drivers of forest loss from the input data. For example, pixels around new access routes (e.g. roads) were assigned high risk, whereas this was not the case for recent, concentrated natural loss events (e.g. remote landslides). Convolutional neural networks can harness limited time‐series data to predict near‐future deforestation patterns, an important step in harnessing the growing volume of satellite remote sensing data to curb global deforestation. The modelling framework can be readily applied to any tropical forest location and used by governments and conservation organisations to prevent deforestation and plan protected areas.

Climate velocity—the speed and direction species must move to track climate change—is often estimated without accounting for vegetation-driven microclimatic variation. Using mechanistic microclimate models parameterized with three-dimensional maps of topography and vegetation structure, here we show that microclimate heterogeneity reduces the magnitude and alters the direction of climate velocity for maximum and minimum temperatures. For understory-dwelling organisms, the magnitude of maximum temperature velocity was halved and generally oriented towards areas with dense vegetation. For canopy-dwelling organisms, the magnitude of maximum temperature velocity was nearly zero, with vectors oriented vertically downward. These results demonstrate that vegetation complexity produces localized microrefugia, enabling short-term persistence of species under warming conditions. Our findings emphasize the need to integrate fine-scale habitat heterogeneity into predictions of climate resilience and highlight the value of structurally complex forests in providing microclimatic refugia.