Centre de documentation
scientifique

Pyrenean silver fir forests retain legacies of past disturbances and climate change in their growth, structure and composition / Antonio Gazol in Forests, vol 14 n° 4 (April 2023)  

Public [article]  inForests > vol 14 n° 4 (April 2023) . - n° 713 Titre : Pyrenean silver fir forests retain legacies of past disturbances and climate change in their growth, structure and composition Type de document : Article/Communication Auteurs : Antonio Gazol, Auteur ; Ester González-de-Andrés, Auteur ; Michele Colangelo, Auteur ; et al., Auteur Année de publication : 2023 Article en page(s) : n° 713 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Termes IGN] Abies alba [Termes IGN] croissance des arbres [Termes IGN] dendrochronologie [Termes IGN] dépérissement [Termes IGN] échantillonnage [Termes IGN] Espagne [Termes IGN] historique [Termes IGN] Pyrénées (montagne) [Termes IGN] sécheresse [Termes IGN] sous-étage [Vedettes matières IGN] Végétation et changement climatique Résumé : (auteur) Recent drought-induced dieback alters forest dynamics, which are also shaped by past management. In western Pyrenean silver fir (Abies alba) stands, dieback concurs in space and time with the legacies of past management, but the impacts on forest growth, structure and composition are unknown. We aim to disentangle how dieback interacts with the legacies of past human use and modulates the recent dynamics of silver fir forests. To this end, we sampled eleven silver fir forests across wide climatic gradients and included declining and non-declining sites. We measured radial growth, structure, composition, understory cover and type and amount of deadwood. Silver fir growth declines in response to late-summer drought. In declining sites, most defoliated stands showed the lowest silver fir density and were those where growth depended more on water availability. Tree death enhanced the cover of dominant understory plants such as Buxus sempervirens. Past management activities leave an imprint in the growth of silver fir, such as releases due to past logging, but also affect the number of stumps and snags and the current tree density. A more extensive monitoring will be required to fully disentangle the multiple influences of past management legacies and current climate change on forest dynamics. Numéro de notice : A2023-202 Affiliation des auteurs : non IGN Thématique : FORET Nature : Article DOI : 10.3390/f14040713 Date de publication en ligne : 30/03/2023 En ligne : https://doi.org/10.3390/f14040713 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=103104 [article]

A novel entropy-based method to quantify forest canopy structural complexity from multiplatform lidar point clouds / Xiaoqiang Liu in Remote sensing of environment, vol 282 (December 2022)  

Public [article]  inRemote sensing of environment > vol 282 (December 2022) . - n° 113280 Titre : A novel entropy-based method to quantify forest canopy structural complexity from multiplatform lidar point clouds Type de document : Article/Communication Auteurs : Xiaoqiang Liu, Auteur ; Qin Ma, Auteur ; Xiaoyong wu, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 113280 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] canopée [Termes IGN] Chine [Termes IGN] couvert forestier [Termes IGN] densité des points [Termes IGN] données lidar [Termes IGN] données localisées 3D [Termes IGN] échantillonnage [Termes IGN] écosystème forestier [Termes IGN] entropie [Termes IGN] estimation par noyau [Termes IGN] image captée par drone [Termes IGN] modèle numérique de surface de la canopée [Termes IGN] semis de points Résumé : (auteur) Forest canopy structural complexity (CSC) describes the three-dimensional (3D) arrangement of canopy elements, and has become an emergent forest attribute mediating forest ecosystem functioning along with species diversity. Light detection and ranging (lidar), especially the emerging near-surface lidar platforms (e.g., terrestrial laser scanning/TLS, backpack laser scanning/BLS, unmanned aerial vehicle laser scanning/ULS), can depict 3D canopy information with high efficiency and accuracy, providing an ideal data source for forest CSC quantification. However, current existing lidar-based CSC quantification indices may share common limitations of getting saturated in structurally complex forest stands and not fully capturing within-canopy structural variations. In this study, we introduced the concept of entropy into forest CSC quantification, and proposed a new forest CSC index, namely canopy entropy (CE). Two major bottlenecks were addressed in the CE calculation procedure, including (1) using a Mann-Kendall (MK) test-based resampling strategy to address the issue of incongruent sampling chances of canopy elements at different locations from different lidar systems, and (2) using a kernel density estimation (KDE)-based method to reduce its dependence on point density. The effectiveness and generality of CE were evaluated by simulating TLS and ULS point clouds from nine forest stands and collecting TLS, BLS, and ULS point clouds from 110 field plots distributed in five forest sites, covering a large variety of forest types and forest CSC conditions. The results showed that CE was an effective forest CSC quantification index that successfully captured CSC variations caused by both tree density and the number of vertical canopy layers. It had significant positive correlations with four widely used CSC indices (i.e., canopy cover, foliage height diversity, canopy top rugosity, and fractal dimension; R2: 0.32 to 0.67), but outperformed them by overcoming their common limitations. CE estimates from multiplatform lidar point clouds agreed well with each other (R2 ≥ 0.70, RMSE ≤0.10), indicating it has generality in cross-platform forest CSC quantification practices. We believe the proposed CE index has great potential to help us unravel the correlations among forest CSC, species diversity, and forest ecosystem functions, and therefore improve our understanding on forest ecosystem processes. Numéro de notice : A2022-795 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article DOI : 10.1016/j.rse.2022.113280 Date de publication en ligne : 26/09/2022 En ligne : https://doi.org/10.1016/j.rse.2022.113280 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101930 [article]

Age-independent diameter increment models for mixed mountain forests / Albert Ciceu in European Journal of Forest Research, vol 141 n° 5 (October 2022)  

Public [article]  inEuropean Journal of Forest Research > vol 141 n° 5 (October 2022) . - pp 781 - 800 Titre : Age-independent diameter increment models for mixed mountain forests Type de document : Article/Communication Auteurs : Albert Ciceu, Auteur ; Karol Bronisz, Auteur ; Juan Garcia-Duro, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : pp 781 - 800 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Termes IGN] Abies alba [Termes IGN] croissance des arbres [Termes IGN] diamètre à hauteur de poitrine [Termes IGN] échantillonnage [Termes IGN] Fagus sylvatica [Termes IGN] forêt alpestre [Termes IGN] forêt inéquienne [Termes IGN] modèle de croissance végétale [Termes IGN] modélisation de la forêt [Termes IGN] peuplement mélangé [Termes IGN] Picea abies [Termes IGN] Roumanie [Vedettes matières IGN] Foresterie Résumé : (auteur) Mixed mountain forests with an uneven-aged structure are characterized by a high tree-growth variability making traditional age-dependent growth models inapplicable. Estimating site productivity is yet another impediment for modelling tree growth in such forests. Uneven-aged mixed-stand forests are known for their high resilience, resistance and productivity, and are being promoted as a suitable alternative to even-aged, pure plantations for climate change adaptation and mitigation. However, their growth must be accurately measured and predicted, but diameter at the breast height (dbh) increment models specifically designed for uneven-aged mixed mountain forests are still rare. Using permanent sampling network data and 465 increment cores, we built two age-independent dbh increment (id) models for the main species of the study area, namely Norway spruce (Picea abies (L.) Karst.), silver fir (Abies alba Mill.) and European beech (Fagus sylvatica L.). Mixed effects models and the algebraic difference approach were employed to develop id models based on empirical and commonly used theoretical growth functions. A past growth index was further developed and introduced in the model in order to explain the id variability. Several mixed effects calibration strategies were assessed in order to obtain the most accurate localized curve for new plots. Tree size, competition and biogeoclimatic variables were found to explain the id through the empirical growth function, while the growth index significantly improved the theoretical growth function for Norway spruce. The optimization of the calibration strategy for the mixed effects modelling framework enables the growth index implementation in forest practice as an accurate method for estimating site productivity. The accuracy of the two id models was similar: the root mean squared error of the empirical growth function varied between 0.940 and 1.042 cm for spruce, beech and fir, while the root mean squared error obtained through the theoretical growth function for spruce only was 1.105 cm. The basal area increment prediction at the plot level based on the theoretical growth function reached a root mean squared error of 0.043 m2 while using the empirical growth function the root mean squared error is 0.047 m2. The high accuracy obtained using age-independent models underlines their suitability for predicting growth in mixed uneven-aged forests. The developed models can be easily integrated into forest practice to accurately obtain id estimates. Numéro de notice : A2022-758 Affiliation des auteurs : non IGN Thématique : FORET Nature : Article DOI : 10.1007/s10342-022-01473-5 Date de publication en ligne : 13/08/2022 En ligne : https://doi.org/10.1007/s10342-022-01473-5 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101767 [article]

Estimating urban functional distributions with semantics preserved POI embedding / Weiming Huang in International journal of geographical information science IJGIS, vol 36 n° 10 (October 2022)  

Public [article]  inInternational journal of geographical information science IJGIS > vol 36 n° 10 (October 2022) . - pp 1905 - 1930 Titre : Estimating urban functional distributions with semantics preserved POI embedding Type de document : Article/Communication Auteurs : Weiming Huang, Auteur ; Lizhen Cui, Auteur ; Meng Chen, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : pp 1905 - 1930 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Analyse spatiale [Termes IGN] attention (apprentissage automatique) [Termes IGN] Chine [Termes IGN] classe sémantique [Termes IGN] classification par réseau neuronal récurrent [Termes IGN] distribution spatiale [Termes IGN] échantillonnage [Termes IGN] lissage de données [Termes IGN] matrice de co-occurrence [Termes IGN] Perceptron multicouche [Termes IGN] point d'intérêt [Termes IGN] triangulation de Delaunay [Termes IGN] zone urbaine Résumé : (auteur) We present a novel approach for estimating the proportional distributions of function types (i.e. functional distributions) in an urban area through learning semantics preserved embeddings of points-of-interest (POIs). Specifically, we represent POIs as low-dimensional vectors to capture (1) the spatial co-occurrence patterns of POIs and (2) the semantics conveyed by the POI hierarchical categories (i.e. categorical semantics). The proposed approach utilizes spatially explicit random walks in a POI network to learn spatial co-occurrence patterns, and a manifold learning algorithm to capture categorical semantics. The learned POI vector embeddings are then aggregated to generate regional embeddings with long short-term memory (LSTM) and attention mechanisms, to take account of the different levels of importance among the POIs in a region. Finally, a multilayer perceptron (MLP) maps regional embeddings to functional distributions. A case study in Xiamen Island, China implements and evaluates the proposed approach. The results indicate that our approach outperforms several competitive baseline models in all evaluation measures, and yields a relatively high consistency between the estimation and ground truth. In addition, a comprehensive error analysis unveils several intrinsic limitations of POI data for this task, e.g. ambiguous linkage between POIs and functions. Numéro de notice : A2022-738 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article DOI : 10.1080/13658816.2022.2040510 Date de publication en ligne : 08/03/2022 En ligne : https://doi.org/10.1080/13658816.2022.2040510 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101714 [article]

Spatial regression graph convolutional neural networks: A deep learning paradigm for spatial multivariate distributions / Di Zhu in Geoinformatica, vol 26 n° 4 (October 2022)  

Public [article]  inGeoinformatica > vol 26 n° 4 (October 2022) . - pp 645 - 676 Titre : Spatial regression graph convolutional neural networks: A deep learning paradigm for spatial multivariate distributions Type de document : Article/Communication Auteurs : Di Zhu, Auteur ; Yu Liu, Auteur ; Xin Yao, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : pp 645 - 676 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géomatique [Termes IGN] analyse multivariée [Termes IGN] analyse spatiale [Termes IGN] apprentissage profond [Termes IGN] distribution spatiale [Termes IGN] échantillonnage [Termes IGN] intelligence artificielle [Termes IGN] régression [Termes IGN] régression géographiquement pondérée [Termes IGN] réseau neuronal convolutif [Termes IGN] réseau neuronal de graphes Mots-clés libres : Geospatial artificial intelligence (GeoAI) Résumé : (auteur) Geospatial artificial intelligence (GeoAI) has emerged as a subfield of GIScience that uses artificial intelligence approaches and machine learning techniques for geographic knowledge discovery. The non-regularity of data structures has recently led to different variants of graph neural networks in the field of computer science, with graph convolutional neural networks being one of the most prominent that operate on non-euclidean structured data where the numbers of nodes connections vary and the nodes are unordered. These networks use graph convolution – commonly known as filters or kernels – in place of general matrix multiplication in at least one of their layers. This paper suggests spatial regression graph convolutional neural networks (SRGCNNs) as a deep learning paradigm that is capable of handling a wide range of geographical tasks where multivariate spatial data needs modeling and prediction. The feasibility of SRGCNNs lies in the feature propagation mechanisms, the spatial locality nature, and a semi-supervised training strategy. In the experiments, this paper demonstrates the operation of SRGCNNs with social media check-in data in Beijing and house price data in San Diego. The results indicate that a well-trained SRGCNN model is capable of learning from samples and performing reasonable predictions for unobserved locations. The paper also presents the effectiveness of incorporating the idea of geographically weighted regression for handling heterogeneity between locations in the model approach. Compared to conventional spatial regression approaches, SRGCNN-based models tend to generate much more accurate and stable results, especially when the sampling ratio is low. This study offers to bridge the methodological gap between graph deep learning and spatial regression analytics. The proposed idea serves as an example to illustrate how spatial analytics can be combined with state-of-the-art deep learning models, and to enlighten future research at the front of GeoAI. Numéro de notice : A2022-865 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE/INFORMATIQUE Nature : Article DOI : 10.1007/s10707-021-00454-x Date de publication en ligne : 02/11/2021 En ligne : https://doi.org/10.1007/s10707-021-00454-x Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=102158 [article]

Regional climate moderately influences species-mixing effect on tree growth-climate relationships and drought resistance for beech and pine across Europe / Géraud de Streel in Forest ecology and management, vol 520 (September-15 2022)  

Permalink

Characterizing the calibration domain of remote sensing models using convex hulls / Jean-Pierre Renaud in International journal of applied Earth observation and geoinformation, vol 112 (August 2022)  

Permalink

Comment déterminer l'exposition aux changements climatiques des zones de production forestière française ? Méthodologie utilisée dans le projet ESPERENSE pour cibler les zones d’intérêt pour l’installation d’essais de comparaison d’essences et de provenances / Hedi Kebli in Revue forestière française, vol 73 n° 5 (2021)  

Permalink

Research on automatic identification method of terraces on the Loess plateau based on deep transfer learning / Mingge Yu in Remote sensing, vol 14 n° 10 (May-2 2022)  

Permalink

Effect of climate change on the growth of tree species: Dendroclimatological analysis / Archana Gauli in Forests, vol 13 n° 4 (April 2022)  

Permalink

Two-phase forest inventory using very-high-resolution laser scanning / Henrik J. Persson in Remote sensing of environment, vol 271 (March- 2 2022)  

Permalink

Survival time and mortality rate of regeneration in the deep shade of a primeval beech forest / R. Petrovska in European Journal of Forest Research, vol 141 n° 1 (February 2022)  

Permalink

Une généralisation de la méthode de partage des poids dans le cas où la base de sondage est continue / Philippe Brion (2022)    

Permalink

Metalearning / Pavel Brazdil (2022)  

Permalink

A square-grid sampling support to reconcile systematicity and adaptivity in the periodic spatial survey of natural resources / Olivier Bouriaud (2022)  

Permalink