Abstract: Plant phenology characterises the seasonal cyclicity of biological events such as budburst, flowering, fructification, leaf senescence and leaf fall. These biological events are genetically pre-determined but also strongly modulated by climatic conditions, particularly temperature, daylength and water availability. Therefore, the timing of these events is considered as a good indicator of climate change impacts and as a key parameter for understanding and modelling vegetation-climate interactions. In situ observations, empirical or bioclimatic models and remotely sensed time-series data constitute the three possible ways for monitoring the timing of plant phenological events. Remote sensing has the advantage of being the only way of surface sampling at high temporal frequency and, in the case of satellite-based remote sensing, over large regions. Nevertheless, exogenous factors, particularly atmospheric conditions, lead to some uncertainties on the seasonal course of surface reflectance and cause bias in the identification of vegetation phenological events. Since 2005, a network of forest and herbaceous sites has been equipped with laboratory made NDVI sensors to monitor the temporal dynamics of canopy structure and phenology at an intra-daily time step. In this study, we present recent results obtained in several contrasting biomes in France, French Guiana, Belgium and Congo. These sites represent a gradient of vegetation ecosystems: the main evergreen and deciduous forest ecosystems in temperate climate region, an evergreen tropical rain forest in French Guiana, an herbaceous savanna ecosystem in Congo, and a succession of three annual crops in Belgium. In this paper, (1) we provide an accurate description of the seasonal dynamics of vegetation cover in these different ecosystems (2) we identify the most relevant remotely sensed markers from NDVI time-series for determining the dates of the main phenological events that characterize these ecosystems and (3) we discuss the relationships between temporal canopy dynamics and climate factors. In addition to its importance for phenological studies, this ground-based Network of NDVI measurement provides data needed for the calibration and direct validation of satellite observations and products. © 2012 Elsevier Inc.

Abstract: Simple sequence repeat (SSR) markers are widely used tools for inferences about genetic diversity, phylogeography and spatial genetic structure. Their applications assume that variation among alleles is essentially caused by an expansion or contraction of the number of repeats and that, accessorily, mutations in the target sequences follow the stepwise mutation model (SMM). Generally speaking, PCR amplicon sizes are used as direct indicators of the number of SSR repeats composing an allele with the data analysis either ignoring the extent of allele size differences or assuming that there is a direct correlation between differences in amplicon size and evolutionary distance. However, without precisely knowing the kind and distribution of polymorphism within an allele (SSR and the associated flanking region (FR) sequences), it is hard to say what kind of evolutionary message is conveyed by such a synthetic descriptor of polymorphism as DNA amplicon size. In this study, we sequenced several SSR alleles in multiple populations of three divergent tree genera and disentangled the types of polymorphisms contained in each portion of the DNA amplicon containing an SSR. The patterns of diversity provided by amplicon size variation, SSR variation itself, insertions/deletions (indels), and single nucleotide polymorphisms (SNPs) observed in the FRs were compared. Amplicon size variation largely reflected SSR repeat number. The amount of variation was as large in FRs as in the SSR itself. The former contributed significantly to the phylogenetic information and sometimes was the main source of differentiation among individuals and populations contained by FR and SSR regions of SSR markers. The presence of mutations occurring at different rates within a marker's sequence offers the opportunity to analyse evolutionary events occurring on various timescales, but at the same time calls for caution in the interpretation of SSR marker data when the distribution of within-locus polymorphism is not known.

Abstract: We predict stand basal area (BA) from small footprint LiDAR data in 129 one-ha tropical forest plots across four sites in French Guiana and encompassing a great diversity of forest structures resulting from natural (soil and geological substrate) and anthropogenic effects (unlogged and logged forests). We use predictors extracted from the Canopy Height Model to compare models of varying complexity: single or multiple regressions and nested models that predict BA by independent estimates of stem density and quadratic mean diameter. Direct multiple regression was the most accurate, giving a 9.6% Root Mean Squared Error of Prediction (RMSEP). The magnitude of the various errors introduced during the data collection stage is evaluated and their contribution to MSEP is analyzed. It was found that these errors accounted for less than 10% of model MSEP, suggesting that there is considerable scope for model improvement. Although site-specific models showed lower MSEP than global models, stratification by site may not be the optimal solution. The key to future improvement would appear to lie in a stratification that captures variations in relations between LiDAR and forest structure. © 2012 Elsevier Inc.

Abstract: Cocoa black pod rot, a disease caused by Stramenopiles of the genus Phytophthora, and particularly by the pan-tropical species P. palmivora, causes serious production losses worldwide. In order to reduce the impact of these pests and diseases, preference is given to genetic control using resistant varieties and, to that end, breeders seek sources of resistance in wild cocoa trees. For instance, surveys of spontaneous cocoa trees in French Guiana between 1985 and 1995 led to the collection of abundant plant material forming a particular genetic group (the “Guiana” group). Following numerous one-off studies demonstrating the merits of this group as a source of resistance to Phytophthora, this article presents the results of a comprehensive study assessing the resistance of 186 “Guiana” clones in relation to the Guianan strain (GY 27) of P. palmivora. This study, undertaken in French Guiana, using an efficient methodology (ten series of tests and a statistical test adapted to the ordinal nature of the data) confirmed that the “Guiana” genetic group does indeed constitute an important source of resistance to P. palmivora, though with some variations depending on the demes of origin. Numerous clones (59) proved to be as resistant as the SCAVINA 6 resistance control, whilst nine were statistically more resistant. The “Resistant” and “Moderately Resistant” Guianan clones totalled 108 (58% of the total tested). Some of the clones more resistant than SCAVINA 6 could be incorporated into numerous cocoa breeding programmes, particularly those that also display other notable qualities. The same applies for numerous other clones equivalent to SCAVINA 6, especially the “elite”' clones GU 134-B, GU 139-A and GU 285-A. © 2012 Thevenin et al.

Abstract: Volatile terpenes are among the most diverse class of defensive compounds in plants, and they are implicated in both direct and indirect defense against herbivores. In terpenes, both the quantity and the diversity of compounds appear to increase the efficiency of defense as a diverse blend of compounds provides a more efficient protection against a broader range of herbivores and limits the chances that an enemy evolves resistance. Theory predicts that plant defensive compounds should be allocated differentially among tissues according to the value of the tissue, its cost of construction and the herbivore pressure on it. We collected volatile terpenes from bark and leaves of 178 individual tree belonging to 55 angiosperm species in French Guiana and compare the kind, amount, and diversity of compounds in these tissues. We hypothesized that in woody plants, the outermost part of the trunk should hold a more diverse blend of volatile terpenes. Additionally, as herbivore communities associated with the leaves is different to the one associated with the bark, we also hypothesized that terpene blends should be distinct in the bark vs. the leaves of a given species. We found that the mixture of volatile terpenes released by bark is different and more diverse than that released by leaves, both in monoterpenes and sesquiterpenes. This supports our hypothesis and further suggests that the emission of terpenes by the bark should be more important for trunk defense than previously thought.