Search

Forest disturbances can reduce the potential of ecosystems to provide resources and services. Despite the urgent need to understand the effects of logging on tropical ecosystems, the quantification of disturbances arising from selective logging remains a challenge. Here, we used canopy-three-dimensional information retrieved from Terrestrial Laser Scanner (TLS) measurements to investigate the impacts of logging on key structural traits relevant to forest functioning. We addressed the following questions: 1) Which canopy structural traits were mostly affected by logging? 2) Can remotely-sensed canopy structural traits be used to quantify forest distur-bances? Fourteen canopy structural traits were applied as input to machine learning models, which were trained to quantify the intensity of logging disturbance. The plots were located in Malaysian Borneo, over a gradient of logging intensity, ranging from forest not recently disturbed by logging, to forest at the early stage of recovery following logging. Our results showed that using the Random Forest regression approach, the Plant Area Index (PAI) between 0 m -5 m aboveground, Relative Height at 50 %, and metrics describing plant allocation in the middle-higher canopy layer were the strongest predictors of disturbance. In particular, PAI between 35 m and 40 m explained 12 % to 19 % of the structural variability between plots, followed by the relative height at 50 %, (10.5 % -18.6 %), and the foliage height diversity (7.5 % -16.9 %). The approach presented in this study allowed a spatially explicitly characterization of disturbances, providing a novel approach for quantifying and monitoring the integrity of tropical forests. Our results indicate that canopy structural traits can provide a robust indication of disturbances, with strong potential to be applied at regional or global scales. The data used in this study are openly available and we encourage other researchers to use them as a benchmark data set to test larger scale approaches based on satellite and airborne platforms.

International audience; Due to the more frequent use of crop models at regional and national scale, the effects of spatial data input resolution have gained increased attention. However, little is known about the influence of variability in crop management on model outputs. A constant and uniform crop management is often considered over the simulated area and period. This study determines the influence of crop management adapted to climatic conditions and input data resolution on regional-scale outputs of crop models. For this purpose, winter wheat and maize were simulated over 30 years with spatially and temporally uniform management or adaptive management for North Rhine-Westphalia ((similar to)34 083 km(2)), Germany. Adaptive management to local climatic conditions was used for 1) sowing date, 2) N fertilization dates, 3) N amounts, and 4) crop cycle length. Therefore, the models were applied with four different management sets for each crop. Input data for climate, soil and management were selected at five resolutions, from 1 x 1 km to 100 x 100 km grid size. Overall, 11 crop models were used to predict regional mean crop yield, actual evapotranspiration, and drainage. Adaptive management had little effect (< 10% difference) on the 30-year mean of the three output variables for most models and did not depend on soil, climate, and management resolution. Nevertheless, the effect was substantial for certain models, up to 31% on yield, 27% on evapotranspiration, and 12% on drainage compared to the uniform management reference. In general, effects were stronger on yield than on evapotranspiration and drainage, which had little sensitivity to changes in management. Scaling effects were generally lower than management effects on yield and evapotranspiration as opposed to drainage. Despite this trend, sensitivity to management and scaling varied greatly among the models. At the annual scale, effects were stronger in certain years, particularly the management effect on yield. These results imply that depending on the model, the representation of management should be carefully chosen, particularly when simulating yields and for predictions on annual scale.

International audience; One strategy to reduce usage of crop protection agrochemicals in viticulture is to adjust the spray rate according to local vegetation characteristics. Acquisitions of mobile 2D LiDAR data and deposition rate measurements with a face-to-face sprayer were made on plant blocks in 5 fields and at 4 dates in a Mediterranean vineyard. Two indicators were calculated: the leaf wall area by points and the tree area index. Univariate deposition prediction models using either indicator were estimated from data acquired in 2 fields, and evaluated with regards to data from 3 other fields. In this experiment with a 3 m spatial resolution, the model based on LWApts was found a little more accurate.

International audience; The effects of leaf inclination on plant light capture, growth, and water balance of monospecific canopies are well documented, but we still lack information on such effects in the case of multispecific canopies. We investigated the effects of leaf inclination on the absorption of photosynthetically active radiation (PAR) of a mixed forest. We ran a 3D mechanistic radiation transfer model for a Mediterranean forest where Pinus halepensis makes the upper strata while Quercus ilex occupies the lower strata. As factors, we included (1) the distributions of leaf inclinations that ranged from vertical to horizontal (including the actual inclinations), (2) the fraction of diffuse light, sun position, and leaf area index (LAI), and (3) the Pinus/Quercus LAI ratio. Simulated PAR absorption was more than twice as sensitive to leaf inclination in oaks than in pines because oaks depended on PAR transmitted below the pine layer. The extent of the effect depended on season, fraction of diffuse light, LAI, and vegetation spatial structure. None of the observed inclinations maximized PAR absorption, suggesting a trade-off with water economy. Erroneous assumptions about leaf inclination lead to larger errors when modelling heterogeneous, mixed canopies. This also highlights potential caveats when using models that do not account for the spatial structure of canopies.

Estimating forest structural attributes using multispectral remote sensing is challenging because of the saturation of multispectral indices at high canopy cover. The objective of this study was to assess the utility of hyperspectral data in estimating and mapping forest structural parameters including mean diameter-at-breast height (DBH), mean tree height and tree density of a closed canopy beech forest ( Fagus sylvatica L.). Airborne HyMap images and data on forest structural attributes were collected from the Majella National Park, Italy in July 2004. The predictive performances of vegetation indices (VI) derived from all possible two-band combinations (VI ( i , j ) = ( R i − R j )/( R i + R j ), where R i and R j = reflectance in any two bands) were evaluated using calibration ( n = 33) and test ( n = 20) data sets. The potential of partial least squares (PLS) regression, a multivariate technique involving several bands was also assessed. New VIs based on the contrast between reflectance in the red-edge shoulder (756–820 nm) and the water absorption feature centred at 1200 nm (1172–1320 nm) were found to show higher correlations with the forest structural parameters than standard VIs derived from NIR and visible reflectance (i.e. the normalised difference vegetation index, NDVI). PLS regression showed a slight improvement in estimating the beech forest structural attributes (prediction errors of 27.6%, 32.6% and 46.4% for mean DBH, height and tree density, respectively) compared to VIs using linear regression models (prediction errors of 27.8%, 35.8% and 48.3% for mean DBH, height and tree density, respectively). Mean DBH was the best predicted variable among the stand parameters (calibration R 2 = 0.62 for an exponential model fit and standard error of prediction = 5.12 cm, i.e. 25% of the mean). The predicted map of mean DBH revealed high heterogeneity in the beech forest structure in the study area. The spatial variability of mean DBH occurs at less than 450 m. The DBH map could be useful to forest management in many ways, e.g. thinning of coppice to promote diameter growth, to assess the effects of management on forest structure or to detect changes in the forest structure caused by anthropogenic and natural factors.

Leaf angle distribution (LAD) is one of the most important parameters used to describe the structure of horizontally homogeneous vegetation canopies such as field crops. LAD affects how incident photosynthetically active radiation is distributed on plant leaves, thus directly affecting plant productivity. Knowledge of LAD is also required for retrieval of other important biophysical variables from measurements of canopy radiation transmittance or spectral reflectance. Unfortunately, its determination is laborious and measured data is rarely available. In this study, we applied a recently developed method for determination of the inclination angles using leveled digital photography to the leaves of six cool-temperate crops: faba bean, narrow-leafed lupin, turnip rape, wheat, barley, and oat. The method, previously applied only to small and flat leaves of broadleaved trees and bushes, was extended to be applicable to the narrow and curved leaves of cereals. A reasonable match was found between the leaf angles determined by photographic measurements and the mean leaf tilt angles (MTA) measured using a LAI-2000 plant canopy analyzer for five out of the six species (R2=0.92). The error caused by assuming a spherical LAD, when calculating LAI from canopy transmittance measurements, varied between 0 and 1.5 LAI units, depending on species. Finally, we analyzed the correlation between photographically determined species-specific LADs and airborne imaging spectroscopy data acquired for the same species in a similar growth stage. The highest correlation between spectral reflectance factor and leaf mean tilt angle was found at a wavelength of 748nm (R=0.80). The high correlation between MTA and this red edge waveband can be useful for MTA determination from imaging spectroscopy.

Made available in DSpace on 2019-09-12T16:53:49Z (GMT). No. of bitstreams: 0 Previous issue date: 2009 Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) Universidade de São Paulo (USP) Comité Français d´Evaluation de la Coopération Universitaire avec le Brésil (COFECUB) European Integrated Project 'Ultra Low CO2 Steelmaking' [515960] French Ministry of Foreign Affairs Eucalyptus grandis (W. Hill ex Maiden) leaf traits and tree growth Were Studied over 3 years after the establishment of two adjacent complete randomized block designs in Southern Brazil. In a nitrogen (N) input experiment, a treatment with the application of 120 kg N ha(-1) was compared to a control treatment without N addition, and in I potassium (K) input experiment a control treatment without K addition was compared to a treatment with the application of 116 kg K ha(-1). Young leaves Were tagged 9 months after planting to estimate the effect of N and K fertilizations oil leaf lifespan. Leaf mass, specific leaf area and nutrient concentrations were measured oil a composite sample per plot every 28 days Until the last tagged leaf fell. Successive inventories, destructive sampling of trees and leaf litter fall collection made it possible to assess the effect of N and K fertilization oil the dynamics of biomass accumulation in above-ground tree components. Whilst the effects of N fertilization oil tree growth Only Occurred in the first 24 months after planting, K fertilization increased the above-ground net primary production from 4478 to 8737 g m(-2) over the first 36 months after planting. The average lifespan of tagged leaves was not modified by N addition but it increased from 111 to 149 days with K fertilization. The peak of leaf production Occurred in the second year after planting (about 800 g m(-2) year(-1)) and wits not significantly modified (P < 0.05) by N and K fertilizations. By contrast, K addition significantly increased the maximum leaf standing biomass from 292 to 528 g m(-2) mainly as a consequence of the increase in leaf lifespan. Potassium fertilization increased the stand biomass mainly through the enhancement in leaf area index (LAI) since growth efficiency (defined as the ratio between woody biomass production and LAI) was not significantly modified. A better understanding of the physiological processes governing the leaf lifespan is necessary to improve process-based models Currently used in Eucalyptus plantations. [Laclau, Jean-Paul; Saint-Andre, Laurent; Nouvellon, Yann] CIRAD, Persyst UPR80, TA10 D, F-34398 Montpellier 5, France [Laclau, Jean-Paul; Almeida, Julio C. R.; Goncalves, Jose Leonardo M.; Ventura, Marcelo; Moreira, Rildo M.; Nouvellon, Yann] Univ Sao Paulo, Dept Ciencias Florestais, BR-13418900 Piracicaba, SP, Brazil [Ranger, Jacques] INRA, Biogeochim Ecosyst Forestiers, F-54280 Seichamps, France [Almeida, Julio C. R.] Universidade de Taubaté (Unitau), BR-1202070 Taubate, SP, Brazil

Accurately classifying 3-D point clouds into woody and leafy components has been an interest for applications in forestry and ecology including the better understanding of radiation transfer between canopy and atmosphere. The past decade has seen an increase in the methods attempting to classify leaves and wood in point clouds based on radiometric or geometric features. However, classification purely based on radiometric features is sensor-specific, and the method by which the local neighborhood of a point is defined affects the accuracy of classification based on geometric features. Here, we present a leaf-wood classification method combining geometrical features defined by radially bounded nearest neighbors at multiple spatial scales in a machine learning model. We compared the performance of three different machine learning models generated by the random forest (RF), XGBoost, and lightGBM algorithms. Using multiple spatial scales eliminates the need for an optimal neighborhood size selection and defining the local neighborhood by radially bounded nearest neighbors makes the method broadly applicable for point clouds of varying quality. We assessed the model performance at the individual tree- and plot-level on field data from tropical and deciduous forests, as well as on simulated point clouds. The method has an overall average accuracy of 94.2% on our data sets. For other data sets, the presented method outperformed the methods in literature in most cases without the need for additional postprocessing steps that are needed in most of the existing methods. We provide the entire framework as an open-source python package.