Biology subjects

Vincent, G.

Publications and source records attributed to Vincent, G..

13 recordsLinked to original sources

bioRxiv2026

Genomic forecasting links climate-associated variation to spring phenology in the European wild apple, a crop wild relative of the cultivated apple

Climate-associated genomic variation is often used as a proxy for adaptive potential, yet whether it reflects adaptation with phenotypic consequences, rather than demography alone, is rarely tested in long-lived species. Genomic offsets forecast climate maladaptation from genotype-environment models but are seldom validated against phenotypes. We combined whole-genome resequencing, SNP and transposable element insertion polymorphism (TIP) variation, landscape genomics, genome-wide association, genomic offset, and common-garden phenotyping in the European wild apple (Malus sylvestris Mill.), a crop wild relative with a complex demographic history, to test this. Results: Of 148 resequenced accessions (135 wild, 13 cultivated), 89 wild individuals formed the core dataset. Population structure was associated with Last Glacial Maximum precipitation rather than contemporary climate, and genetic diversity increased from west to east. Genome-environment associations across 9.75 million SNPs identified climate-associated loci enriched in photoperiod-related flowering-time genes, with limited overlap with phenotype-associated loci. SNP-based genomic offset was associated with bud-break timing in the Saclay common garden (marginal R2 = 0.31), with higher offsets corresponding to earlier bud break, but not with growth or ecophysiological traits. This association depended on Austrian and Romanian populations and was clearest in non-admixed wild individuals. Offsets based on TIPs or glacial climate did not predict measured phenotypes. Conclusions: Climate association, phenotypic relevance, and predictive power of genomic variation are separable. In a long-lived tree, genomic offset predicted spring phenology, a key adaptive trait, but this signal varied across provenances and ancestries and was absent for transposable elements. Genomic offsets can therefore forecast potential phenological maladaptation, provided they are validated against phenotypes such as bud break.

ecology↗

bioRxiv2026

Phenological regularity, not functional traits, determines whether tropical tree species can be mapped from imaging spectroscopy

AO_SCPLOWBSTRACTC_SCPLOWO_LIAirborne imaging spectroscopy enables species-level classification in hyperdiverse tropical forests, but accuracy varies enormously among species. We asked which ecological and evolutionary attributes make a tropical tree species spectrally separable. C_LIO_LIUsing 3,256 field-verified crowns spanning 169 species in a hyperdiverse moist forest in French Guiana, we tested seven hypothesised determinants of classification accuracy at species, pairwise, and individual-crown scales using random forest, beta regression, elastic net, and binomial GLMM analyses. C_LIO_LIPhenological regularity - the strength and consistency of seasonal leaf-cycling - was the single strongest predictor of separability, emerging as the top-ranked variable across all analyses. The presence of congeneric species in the classification pool also reduced accuracy, while broader phylogenetic isolation contributed in multivariate models. At the crown level, crown area was the strongest predictor of correct classification, while liana infestation reduced odds of correct identification by 38%. Leaf chemical traits did not predict separability. C_LIO_LIIt is the consistency of a species ecological signal - its phenological rhythm, spatial sampling, and freedom from canopy contamination - rather than any single functional trait, that determines whether it can be reliably mapped from imaging spectroscopy. C_LI

ecology↗

bioRxiv2026

Trait diversity enhances biomass gains via canopy packing in old-growth but not in disturbed Amazon forests.

The existence of a causal link between biodiversity and forest productivity remains largely unexplored in natural systems, especially in hyper-diverse tropical forests. Canopy packing-- greater crown complementarity, resulting in more densely packed canopies--has recently emerged as a key structural pathway through which diversity influences forest functioning, though evidence remains limited and sometimes contradictory.In this study, we used repeated airborne LiDAR acquisitions and long-term field monitoring from a tropical logging experiment in French Guiana to quantify canopy packing using the Shannon evenness of plant area density (PAD) and assess its role in mediating the relationship between trait diversity and biomass gains in old-growth and disturbed Amazonian forest stands.Our results show that, in undisturbed forests, functionally diverse communities promote greater canopy packing, which in turn enhances biomass gains. However, this effect was absent in previously logged stands, where forest structural diversity did not fully recover even after 40 years. Our findings indicate that logging reduces canopy structural complexity and disrupts the link between species composition, canopy packing, and productivity in these hyper-diverse, hyper-productive ecosystems. Significance StatementIn this study, measurements from repeated airborne LiDAR acquisitions and long-term field monitoring from a tropical logging experiment in the Amazon forest are used to understand the causal link between biodiversity and forest productivity. The study shows that greater crown complementarity mediates diversity-productivity relationships, with functionally diverse communities promoting greater canopy packing, which in turn enhances biomass gains. However, this effect is lost in disturbed forests. These findings are relevant for understand the ecological mechanisms driving forest productivity and tropical forests response to disturbance and for forest carbon management strategies.

ecology↗

bioRxiv2025

Monitoring tropical forests with light drones: ensuring spatial and temporal consistency in stereophotogrammetric products

O_LILight drones provide a cheap and effective tool to monitor forest canopy, especially in tropical and equatorial contexts, where infrastructure and resources are limiting. In these regions, good quality optical satellite images are rare, yet the stakes are maximal to characterize forest function, dynamics, diversity, and phenology, and more generally the vegetation-climate interplay. C_LIO_LIWe describe a complete processing chain based on photogrammetric tools that seeks to optimize the spatial and spectral coherence between repeat image mosaics at centimetric resolution. Our target is to allow individual tree-level monitoring over tens to hundreds of hectare scales with consumer grade equipment (i.e., quadcopter with stabilized RGB camera, standard GNSS positioning). C_LIO_LIWe demonstrate the increase in spatial precision achieved using Time-SIFT and Arosics algorithms, which allow (individually and synergistically) to reduce global and local spatial misalignment between mosaics from several meters to a few centimeters. Time-SIFT provides the advantage of increased robustness in initial image alignment and 3D reconstruction, and hence reduces occasional distortions or data gaps. Using Agisofts color and white balance corrections combined with the use of vegetation indices provides meaningful quantitative signal despite considerable changes in acquisition conditions. C_LIO_LIIn particular, indices that are less sensitive to illumination changes, like the green chromatic coordinate (GCC), allowed evidencing a seasonal signal over four years of monitoring in the evergreen moist forest at Paracou in French Guiana. The signal was decorrelated from obvious geometrical effect (sun height), and provided information on the vegetative stage at tree, species, and stand levels. C_LI Data/Code for peer reviewThe complete processing chain, as well as the data and scripts used for producing the analyses presented here, are available for review on Zenodo: https://zenodo.org/records/15449377?token=eyJhbGciOiJIUzUxMiJ9.eyJpZCI6IjdjNWExMzIzLThkNDMtNDllNy1iYWY4LTY2MGZlZjkyZmQ3OCIsImRhdGEiOnt9LCJyYW5kb20iOiI5MGQ4NTE2YTE4OGViNDQ3YTFiZmMyYTFkZDlhZTZmMiJ9.CsJ0VRuQ90A1qzO1VJC1Q9eXFSp1N5UpeJlyr6otgXRPlf-I-jcwBJ6ytiBbbu8enCNJ2Ke6-oxNV8aeJ_AWIw

ecology↗

bioRxiv2024

Harnessing temporal and spectral dimensionality to map and identify species of individual trees in diverse tropical forests

Species-level canopy maps underpin tropical forest biodiversity monitoring, conservation planning, and carbon accounting, yet high species richness and structural complexity make remote classification challenging. Here we evaluate how far a two-step mapping approach can be taken, and where it breaks down, in hyperdiverse moist forest at the Paracou Field Station, French Guiana. First, we delineate individual tree crowns from ten repeat uncrewed aerial vehicle (UAV) RGB surveys with Mask R-CNN, fusing predictions across dates by temporal consensus: mean segmentation F1 rose from 0.68 (single date) to 0.78 (ten dates), covering approximately 86% of test-region canopy area. Second, we classify each crown from a single airborne hyperspectral acquisition (416-2500 nm, 1 m) using machine learning classifiers trained and tested on 3,186 field-verified crowns spanning 169 species. Linear Discriminant Analysis performed best (weighted F1 = 0.75), outperforming more flexible models, but unevenly: across repeated cross-validation (20 x 5-fold), on average 50 species (95% CI: 41-63) attained F1 [≥] 0.7 in a given fold and only 15 did so reliably, with many rare species unclassifiable (macro-average F1 = 0.48). Combining both steps, we estimate approximately 70% of the landscapes canopy area was correctly mapped to species. Band-importance and ablation analyses identified the far-red edge (748-775 nm) as the most informative spectral region. These results advance on studies limited to 20 or fewer species and set out spectral-resolution and training-data requirements for airborne and forthcoming spaceborne imaging spectrometers, while showing that accuracy remains strongly conditioned by training-data availability and the single-site, single-acquisition design.

ecology↗

bioRxiv2024

Robust characterization of forest structure from airborne laser scanning - a systematic assessment and sample workflow for ecologists

O_LIForests display tremendous structural diversity, shaping carbon cycling, microclimates, and terrestrial habitats. One of the most common tools for forest structure assessments are canopy height models (CHMs): maps of canopy height obtained at high resolution and large scale from airborne laser scanning (ALS). CHMs can be computed in many ways, but little is known about the robustness of different CHM algorithms and how they affect ecological analyses. C_LIO_LIHere, we used high-quality ALS data from nine sites in Australia, ranging from semi-arid shrublands to 90-m tall Mountain Ash canopies, to comprehensively assess CHM algorithms. This included testing their sensitivity to point cloud degradation and quantifying the propagation of errors to derived metrics of canopy structure. C_LIO_LIWe found that CHM algorithms varied widely both in their height predictions (differences up to 10 m, or 60% of canopy height) and in their sensitivity to point cloud characteristics (biases of up to [~]5 m or 40% of canopy height). Impacts of point cloud properties on CHM-derived metrics varied, from robust inference for height percentiles, to considerable errors in aboveground biomass estimates ([~]50 Mg ha-1, or 10% of total), and high volatility in metrics that quantify spatial associations in canopies (e.g., gaps or spatial autocorrelation). In some cases, biases exceeded ecological variation across sites by a factor of 2. However, we also found that two CHM algorithms - a variation on a "spikefree" algorithm that adapts to local pulse densities and a simple Delaunay triangulation of first returns - allowed for robust canopy characterization and should thus create a secure foundation for ecological comparisons in space and time. C_LIO_LICanopy height models are a widely used tool in ecology, but their derivation is not trivial. Our study provides a best-practice guideline and a sample workflow to create robust CHMs and minimize biases and uncertainty in downstream analyses. In doing so we pave the way for global-scale comparisons of forest structural complexity from airborne laser scanning. C_LI

ecology↗

bioRxiv2023

Plant mutations: slaying beautiful hypotheses by surprising evidence

Somatic mutations potentially play a role in plant evolution, but common expectations pertaining to plant somatic mutations remain insufficiently tested. Unlike in most animals, the plant germline is assumed to be set aside late in development, leading to the expectation that plants accumulate somatic mutations along growth. Therefore, several predictions were made on the fate of somatic mutations: mutations have generally low frequency in plant tissues; mutations at high frequency have a higher chance of intergenerational transmission; branching topology of the tree dictates mutation distribution; and, exposure to UV radiation increases mutagenesis. To provide new insights into mutation accumulation and transmission in plants, we produced two high-quality reference genomes and a unique dataset of 60 high-coverage whole-genome sequences of two tropical tree species, Dicorynia guianensis (Fabaceae) and Sextonia rubra (Lauraceae). We identified 15,066 de novo somatic mutations in D. guianensis and 3,208 in S. rubra, surprisingly almost all found at low frequency. We demonstrate that: 1) low-frequency mutations can be transmitted to the next generation; 2) mutation phylogenies deviate from the branching topology of the tree; and 3) mutation rates and mutation spectra are not demonstrably affected by differences in UV exposure. Altogether, our results suggest far more complex links between plant growth, ageing, UV exposure, and mutation rates than commonly thought. Significance StatementThe origin and fate of new mutations have received less attention in plants than in animals. Similarly to animals, plant mutations are expected to accumulate with growth and time, and under exposure to UV light. However, contrary to animals, plant reproductive organs form late in an individuals development, allowing the transmission to the progeny of mutations accumulated along growth. Here, we resequenced DNA from different branches differentially exposed to sunlight of two tropical tree species. We showed that new mutations are generally rare in plant tissues and do not mimic branching patterns but can nevertheless be transmitted to the progeny. Our findings provide a new perspective on heritable plant mutation and its pivotal role as the engine of evolution.

genomics↗

bioRxiv2022

Accurate tropical forest individual tree crown delineation from RGB imagery using Mask R-CNN

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 recognise the irregular edges of individual tree crowns from airborne RGB imagery. We trained and evaluated this model with 3,797 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

ecology↗

bioRxiv2022

Contribution of deep soil layers to the transpiration of a temperate deciduous forest: quantification and implications for the modelling of productivity

Climate change is imposing drier atmospheric and edaphic conditions on temperate forests. Here, we investigated how deep soil (down to 300 cm) water extraction contributed to the provision of water in the Fontainebleau-Barbeau temperate oak forest over two years, including the 2018 record drought. Deep water provision was key to sustain canopy transpiration during drought, with layers below 150 cm contributing up to 60% of the transpired water in August 2018, despite their very low density of fine roots. We further showed that soil databases used to parameterize ecosystem models largely underestimated the amount of water extractable from the soil by trees, due to a considerable underestimation of the tree rooting depth. The consensus database established for France gave an estimate of 207 mm for the soil water holding capacity (SWHC) at Fontainebleau-Barbeau, when our estimate based on the analysis of soil water content measurements was 1.9 times as high, reaching 390{+/-}17 mm. Running the CASTANEA forest model with the database-derived SWHC yielded a 350 gC m-2 y-1 average underestimation of annual gross primary productivity under current climate, reaching up to 700 gC m-2 y-1 under climate change scenario RCP8.5. It is likely that the strong underestimation of SWHC that we show at our site is not a special case, and concerns a large number of forest sites. Thus, we argue for a generalisation of deep soil water content measurements in forests, in order to improve the estimation of SWHC and the simulation of the forest carbon cycle in the current context of climate change. HighlightsO_LIForest-atmosphere carbon exchanges remained insensitive to record drought. C_LIO_LIDeep soil (150-300 cm) provisioned up to 60% of the water transpired by the forest during drought. C_LIO_LISoil databases were underestimating soil water holding capacity by a factor of two. C_LIO_LISimulated forest productivity is strongly sensitive to soil water holding capacity parameter. C_LIO_LIDeep soil water content measurements are urgently needed to correctly estimate the soil water holding capacity. C_LI

ecology↗

bioRxiv2021

Leaf area index estimation of even-aged oak (Quercus petraea) forests using in situ stand dendrometric parameters

The leaf area index (LAI) is a key characteristic of forest stand aboveground net productivity (ANP), and many methods have been developed to estimate the LAI. However, every method has flaws, e.g., methods may be destructive, require means or time and/or show intrinsic bias and estimation errors. A relationship using basal area (G) and stand age to estimate LAI was proposed by Sonohat et al. (2004). We used literature data in addition to data form measurements campaign made in the northern half of France to build a data set with large ranges of pedoclimatic conditions, stand age and measured LAI. We validated the Sonohat et al. (2004) relationship and attempted to improve or modify it using other stand/dendrometric characteristics that could be predictors of the LAI. The result is a series of three models using the G, age and/or quadratic mean diameter (Dg), and the models were able to estimate the LAI of an oak only even-aged forest stand with good confidence (root mean square error, RMSE < 0.75) While G is the main predictor here, age and Dg could be used conjointly or exclusively given the available data, with variable precision in the estimations. Although these models could not, by construction, relate to the interannual variability of the LAI, they may provide the theoretical LAI of an untouched forest (no meteorological, biotic or anthropogenic perturbation) in recent years. additionally, the use of this model may be more interesting than an LAI measurement campaign, depending on the means to be invested in such a campaign.

ecology↗

bioRxiv2021

Potential of Synthetic Aperture Radar Sentinel-1 time series for the monitoring of phenological cycles in a deciduous forest

Annual time-series of the two satellites C-band SAR (Synthetic Aperture Radar) Sentinel-1 A and B data over five years were used to characterize the phenological cycle of a temperate deciduous forest. Six phenological markers of the start, middle and end of budburst and leaf expansion stage in spring and the leaf senescence in autumn were extracted from time-series of the ratio (VV/VH) of backscattering at co-polarization VV (vertical-vertical) and at cross polarization VH (vertical-horizontal). These markers were compared to field phenological observations, and to phenological dates derived from various proxies (Normalized Difference Vegetation Index NDVI time-series from Sentinel-2 A and B images, in situ NDVI measurements, Leaf Area Index LAI and litterfall temporal dynamics). We observe a decrease in the backscattering coefficient ({sigma}0) at VH cross polarization during the leaf development and expansion phase in spring and an increase during the senescence phase, contrary to what is usually observed on various types of crops. In vertical polarization, {sigma}0VV shows very little variation throughout the year. S-1 time series of VV/VH ratio provides a good description of the seasonal vegetation cycle allowing the estimation of spring and autumn phenological markers. Estimates provided by VV/VH of budburst dates differ by approximately 8 days on average from phenological observations. During senescence phase, estimates are positively shifted (later) and deviate by about 20 days from phenological observations of leaf senescence while the differences are of the order of 2 to 4 days between the phenological observations and estimates based on in situ NDVI and LAI time-series, respectively. A deviation of about 7 days, comparable to that observed during budburst, is obtained between the estimates of senescence from S-1 and those determined from the in situ monitoring of litterfall. While in spring, leaf emergence and expansion described by LAI or NDVI explains the increase of VV/VH (or the decrease of {sigma}0VH), during senescence, S-1 VV/VH is decorrelated from LAI or NDVI and is better explained by litterfall temporal dynamics. This behavior resulted in a hysteresis phenomenon observed on the relationships between VV/VH and NDVI or LAI. For the same LAI or NDVI, the response of VV/VH is different depending on the phenological phase considered. This study shows the high potential offered by Sentinel-1 SAR C-band time series for the detection of forest phenology for the first time, thus overcoming the limitations caused by cloud cover in optical remote sensing of vegetation phenology. HighlightsO_LIWe study S-1 C-band dual polarized data potential to predict forest phenology C_LIO_LISeasonal phenological transitions were accurately described by S-1 time-series C_LIO_LIBudburst and senescence dates from S-1 differ from direct observations by one week C_LIO_LITime-series of S-1 VV/VH, NDVI, LAI and litterfall were also compared C_LIO_LIRelationships VV/VH vs NDVI and LAI show a hysteresis according to the season C_LI

ecology↗

bioRxiv2020

A survey of proximal methods for monitoring leaf phenology in temperate deciduous forests

Tree phenology is a major driver of forest-atmosphere mass and energy exchanges. Yet tree phenology has historically not been recorded at flux measurement sites. Here, we used seasonal time-series of ground-based NDVI (Normalized Difference Vegetation Index), RGB camera GCC (Greenness Chromatic Coordinate), broad-band NDVI, LAI (Leaf Area Index), fAPAR (fraction of Absorbed Photosynthetic Active Radiation), CC (Canopy Closure), fRvis (fraction of Reflected Radiation) and GPP (Gross Primary Productivity) to predict six phenological markers detecting the start, middle and end of budburst and of leaf senescence in a temperate deciduous forest. We compared them to observations of budburst and leaf senescence achieved by field phenologists over a 13-year period. GCC, NDVI and CC captured very well the interannual variability of spring phenology (R2 > 0.80) and provided the best estimates of the observed budburst dates, with a mean absolute deviation (MAD) less than 4 days. For the CC and GCC methods, mid-amplitude (50%) threshold dates during spring phenological transition agreed well with the observed phenological dates. For the NDVI-based method, on average, the mean observed date coincides with the date when NDVI reaches 25% of its amplitude of annual variation. For the other methods, MAD ranges from 6 to 17 days. GPP provides the most biased estimates. During the leaf senescence stage, NDVI- and CC-derived dates correlated significantly with observed dates (R2 =0.63 and 0.80 for NDVI and CC, respectively), with MAD less than 7 days. Our results show that proximal sensing methods can be used to derive robust phenological indexes. They can be used to retrieve long-term phenological series at flux measurement sites and help interpret the interannual variability and decadal trends of mass and energy exchanges. HighlightsO_LIWe used 8 indirect methods to predict the timing of phenological events. C_LIO_LIGCC, NDVI and CC captured very well the interannual variation of spring phenology. C_LIO_LIGCC, NDVI and CC provided the best estimates of observed budburst dates. C_LIO_LINDVI and CC derived-dates correlated with observed leaf senescence dates. C_LI

ecology↗

bioRxiv2019

\"Green pointillism\": Detecting the within-population variability of budburst in temperate deciduous trees with phenological cameras

O_LIPhenological cameras have been used over a decade for identifying plant phenological markers (budburst, leaf senescence) and more generally the greenness dynamics of forest canopies. The analysis is usually carried out over the full camera field of view, with no particular analysis of the variability of phenological markers among trees.\nC_LIO_LIHere we show that images produced by phenological cameras can be used to quantify the within-population variability of budburst (WPVbb) in temperate deciduous forests. Using 7 site-years of image analyses, we report a strong correlation (r{superscript 2}=0.97) between the WPVbb determined with a phenological camera and its quantification through ground observation.\nC_LIO_LIWe show that WPVbb varies strongly (by a factor of 4) from year to year in a given population, and that those variations are linked with temperature conditions during the budburst period, with colder springs associated to a higher differentiation of budburst (higher WPVbb) among trees.\nC_LIO_LIDeploying our approach at the continental scale, i.e. throughout phenological cameras networks, would improve the understanding of the spatial (across populations) and temporal (across years) variations of WPVbb, which have strong implications on forest functioning, tree fitness and phenological modelling.\nC_LI

ecology↗