Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / This study aims to quantify and analyze the distribution of the stock of above-ground biomass in Deciduous Forest fragments (FED) in the Region of the Northwest RS, generating the necessary information to support the development of forestry projects. Thus, we installed seven sampling unit of 12 x 12 m with subunits of 5 x 5 m and 1 x 1 m to quantify the natural regeneration classified as Stratum 1 (E1), 2 (E2) and 3 (E3), the E1 is composed of plants with less than 1.3 m from the ground, the E2 all vegetation with over 1.3 m in height and less than 5 cm diameter at breast height (DBH) and E3 all vegetation with DBH between 5 and 10 cm. In the main units all vegetation with more than 10 cm DBH were classified as stratum 4 (E4). For quantification of biomass above ground, after the overthrow of the trees, all tree vegetation was weighed directly in the field, with split trunk (wood with bark), thick branches (diameter greater than 5 cm), thin branches (diameter less than 5 cm), and miscellaneous sheets, these being identified at the species level. To estimate the biomass of the allometric models adjustments were collected the following information dendrometric E4 diameters at 0 (baseline), DAP, 25, 50, 75 and 100% of morphological inversion point height (HPIM) of height HPIM, total height. To estimate the above-ground biomass of different specific gravity methods (MEB) was obtained from wood and other literature were calculated from the disks taken at the time of the DAP, these disk, shelled, had approximately 2 cm thick. The results indicated that the average stock of biomass for the FED was 371.1 Mg.ha-1. The trees with the breast height diameter (DBH) greater 10 cm accounted for over 89% of the biomass (4759 kg). To build the allometric models, the equations adjusted without stratification, have adjusted coefficient of determination (R2aj.) Between 0.726 to 0.972 and standard error of estimate in percentage (Sxy%) ranging from 33.5 to 119.6, the best adjusted model, for not stratified data set was obtained by the stepwise procedure, represented by the following equation: PST = β0 + β1.(DAP3) + β2.H + β3.(DAP3.H), with R2aj.. 0.972 and Sxy 33.5%. For the stratified, only diameter class above 15 cm of acceptable parameters, R2aj. 0.968 and Sxy 26.8%. For the estimation of biomass by different methods, the direct method has accumulated 11,450 kg bole biomass (PSF) for the seven installments. The indirect method that is closer to the PSF was the biomass obtained from the volume of the equation and the MEB arithmetic average of 13,141 kg, and the one with the biggest difference was obtained by rigorous and volume weighted average MEB, which totaled 20,060 kg, and arithmetic mean MEB differed from the PSF by Dunnett's test at 5% probability. It is recommended reliable quantification of above-ground biomass in stratum E4, especially the larger ones, as they may provide greater error in the quantification and estimation of forest biomass. / Este estudo teve por objetivo quantificar e analisar a distribuição do estoque de biomassa acima do solo em fragmentos de Floresta Estacional Decidual (FED) na Região Noroeste do Estado do RS, gerando as informações necessárias para subsidiar a elaboração de projetos florestais. Para tanto, foram instaladas sete unidade amostrais de 12 x 12 m com subunidades de 5 x 5 m e 1 x 1 m para quantificação da regeneração natural classificada em Estrato 1 (E1), 2 (E2) e 3 (E3), o E1 é composto das plantas com menos de 1,3 m de altura do solo, o E2 toda vegetação com mais de 1,3 m de altura e menos que 5 cm de diâmetro a altura do peito (DAP) e o E3 toda vegetação com DAP entre 5 e 10 cm. Nas unidades principais toda vegetação com mais de 10 cm de DAP ficaram classificadas com estrato 4 (E4). Para a quantificação da biomassa acima do solo, após a derrubada das árvores, toda a vegetação arbórea foi pesada diretamente em campo, sendo fraccionadas em tronco (madeira com casca), galhos grossos (diâmetro maior que 5 cm), galhos finos (diâmetro menor que 5 cm), folhas e miscelâneas, estas sendo identificadas a nível de espécies. Para as estimativas de biomassa por ajustes de modelos alométricos foram coletadas as seguintes informação dendrométricas do E4, os diâmetros à 0 (base), DAP, 25, 50, 75 e 100% da altura do ponto de inversão morfológico (HPIM), altura do HPIM, altura total. Para a estimativa da biomassa acima do solo por diferentes métodos a massa específica básica (MEB) da madeira foi obtida na literatura e outras foram calculadas a partir dos discos tirados na altura do DAP, estes disco, sem casca, apresentavam aproximadamente 2 cm de espessura. Os resultados indicaram que o estoque médio de biomassa para a FED foi de 371,1 Mg.ha-1. As árvores com o diâmetro altura do peito (DAP) maior 10 cm representaram mais de 89% da biomassa (4759 Kg). Para o ajuste de modelos alométricos, as equações ajustadas, sem estratificação, apresentam coeficiente de determinação ajustado (R2aj.) entre 0,726 à 0,972 e erro padrão da estimativa em porcentagem (Sxy%) variando de 33,5 a 119,6, o melhor modelo ajustado, para o conjunto de dados não estratificado, foi obtido através do procedimento de Stepwise, sendo representado pela seguinte equação: PST = β0 + β1.(DAP3) + β2.H + β3.(DAP3.H), com R2aj. de 0,972 e Sxy% de 33,5. Para a forma estratificada, apenas a classe de diâmetro acima de 15 cm apresentou parâmetros aceitáveis, com R2aj. de 0,968 e Sxy% de 26,8. Para a estimativa da biomassa por diferentes métodos, o método direto acumulou 11.450 Kg de biomassa do fuste (PSF) para as sete parcelas. O método indireto que mais se aproximou do PSF foi a biomassa obtida através do volume da equação e da MEB média aritmética, 13.141 Kg, e o que apresentou a maior diferença foi obtida pelo volume rigoroso e MEB média ponderada, que totalizou 20.060 Kg, e MEB média aritmética diferiram estatisticamente do PSF pelo teste de Dunnett em nível de 5% de probabilidade de erro. Recomenda-se quantificação fidedigna da biomassa acima do solo no estrato E4, principalmente os maiores indivíduos, pois estes podem propiciar maior erro na quantificação e estimativa da biomassa florestal.

Experten dreier deutscher Hochschulen entwickelten gemeinsam mit der Ostdeutschen Gesellschaft für Forstplanung mbH ein neues, forstliches Monitoringsystem. Das Verbundprojekt „Entwicklung eines forstlichen Monitoringsystems unter Berücksichtigung von Kohlenstoffspeicherung und Klimaanpassung“ (FOMOSY-KK) wird vorgestellt.

Biomass and carbon stock of the natural forests in Vietnam are still not clear due to limitation of knowledge and financial. In this paper, the results of estimating biomass and carbon stocks of the natural forests at Me Linh Biodiversity Station are shown. There are two forest types in this study: the forest vegetation restored after shifting cultivation (vegetation type I) and the forest vegetation restored after clear cutting exploitation (vegetation type II). As the results, the estimated biomass of the forest vegetation restored after shifting cultivation is 86.80 ton.ha-1 and the estimated biomass of the forest vegetation restored after clear cutting exploitation is higher, about 131.59 ton.ha-1. The carbon stock in plants was about 43.40 ton.ha-1 of vegetation type I and 65.79 ton.ha-1 of vegetation type II. The carbon storage in soil of vegetation type I is 79.01 ton.ha-1 and vegetation type II is 99.65 ton.ha-1. Hence, the total of carbon stock in forest vegetation I and II are accounted by 122.41ton.ha-1 and 165.44 ton.ha-1, respectively. In general, it can be pointed out that the naturally recovering secondary forest at Me Linh Station is the secondary young forest with the low economic value due to shortly restored process (about 10-20 years), the flora is not rich and abundant, and there are only commonly pioneer and light demanding tree species. / Sinh khối và trữ lượng các bon của rừng tự nhiên ở Việt Nam vẫn ít được quan tâm của do hạn chế về kiến thức và tài chính. Trong bài báo này, chúng tôi đưa ra kết quả của việc ước lượng sinh khối và tổng hợp các bon của các thảm thực vật rừng thứ sinh phục hồi tự nhiên tại Trạm Đa dạng Sinh học Mê Linh, tỉnh Vĩnh Phúc- Việt Nam, nơi có loại hình thảm thực vật chính, đó là thảm thực vật phục hồi sau nương rẫy (kiểu thảm thục vật I) và thảm thực vật phục hồi sau khai thác kiệt (kiểu thảm thực vật II) nhằm mục đích đánh giá tiềm năng của rừng thứ sinh tại khu vực nghiên cứu. Sinh khối của thảm thực vật phục hồi sau nương rẫy là 86,80 tấn/ha. Sinh khối của thảm thực vật phục hồi sau khai thác cao hơn, đạt 131.59 tấn/ha. Lượng các bon hấp thu trong đất của thảm thực vật I là 79,01 tấn/ha và thảm thực vật II là 99,65 tấn/ha. Như vậy, tổng lượng các bon được hấp thu trong mỗi loại hình thảm thực vật trên là: 122,41 tấn/ha (thảm thực vật I) và 165,14 tấn/ha. Nhìn chung, rừng thứ sinh phục hồi tự nhiên tại Trạm Đa dạng Mê Linh chủ yếu là rừng non thứ sinh, ít có giá trị kinh tế do quá trình phục hồi diễn ra ngắn (khoảng 10-20 năm) nên thành phần thực vật nghèo nàn, không phong phú, thành phần chính chủ yếu là các cây gỗ tiên phong, ưa sáng.

Stem diameter at breast height (D1.3m) and tree height (H) are commonly used measures of tree growth. Based on correlation analysis between biomass of stem, branches and leaves and stem diameter and height of tree we can identify allometric equation for predicting biomass and carbon sequestration of the vegetation. This study was carried out in the natural forests of Me Linh Station for biodiversity to develop allometric equation between biomass and diameter at breast height and height of tree. The study results indicated that twenty tree species dominate in natural forests in Me Linh Station for Biodiversity and they were selected for sampling. Through the 80 established linear equation models for above and below –ground biomass (AGB and BGB), we found that the biomass of tree species in Me Linh Station for Biodiversity were closely correlated with the diameter factor (R>0.902) and not clearly correlated with the height (correlation coefficient = 0.5498, R2< 0.549). Four regression equations were established, including: Pstem = 25.3051*(D1.3m)0.4627 (R2 : 9.661); Pbranch = 12.1043*( D1.3m)0.5416 (R2 : 9.8); Pleaves = 9.446*(D1.3m)0.5976 (R2 : 0.9363); P total biomass of forest = 25.882*D1.725 with R2: 0.8561) for estimating biomass and carbon sequestration of natural forest at the research site. / Đường kính ngang ngực (D1.3m) và chiều cao (H) cây là hai nhân tố thường được dùng để đánh giá sự phát triển của cây gỗ. Việc xây dựng các phương trình tương quan giữa sinh khối (SK) thân, cành, lá, sinh khối tầng cây gỗ, sinh khối của quần xã thực vật với đường kính và chiều cao cây góp phần rất lớn trong dự báo sinh khối và khả năng hấp thụ khí carbon của thảm thực vật. Kết quả nghiên cứu cho thấy 20 loài cây gỗ chiếm ưu thế trong rừng tự nhiên và chúng được chọn để thu mẫu. Mối tương quan giữa sinh khối với 2 nhân tố điều tra rừng là đường kính ngang ngực và chiều cao cây đã đươc kiểm tra thông qua 80 phương trình tương quan. Nhìn chung, sinh khối có tương quan chặt chẽ với nhân tố đường kính (hệ số tương quan R > 0,902), và không tương quan rõ với nhân tố chiều cao (R < 0,5498). Bốn phương trình tính sinh khối cho thảm rừng tại khu vực nghiên cứu đã được thiết lập: SKthân = 25,3051*(D1,3m)0,4627 (R2: 9,661); SKcành: 12,1043*(D1,3m)0,5416 (R2: 9,8); SKlá: 9,446*(D1,3m)0,5976 (R2: 0,9363) và SKtổng = 25,882*D1,725 with R2: 0,8561).

Biomass and carbon stock of the natural forests in Vietnam are still not clear due to limitation of knowledge and financial. In this paper, the results of estimating biomass and carbon stocks of the natural forests at Me Linh Biodiversity Station are shown. There are two forest types in this study: the forest vegetation restored after shifting cultivation (vegetation type I) and the forest vegetation restored after clear cutting exploitation (vegetation type II). As the results, the estimated biomass of the forest vegetation restored after shifting cultivation is 86.80 ton.ha-1 and the estimated biomass of the forest vegetation restored after clear cutting exploitation is higher, about 131.59 ton.ha-1. The carbon stock in plants was about 43.40 ton.ha-1 of vegetation type I and 65.79 ton.ha-1 of vegetation type II. The carbon storage in soil of vegetation type I is 79.01 ton.ha-1 and vegetation type II is 99.65 ton.ha-1. Hence, the total of carbon stock in forest vegetation I and II are accounted by 122.41ton.ha-1 and 165.44 ton.ha-1, respectively. In general, it can be pointed out that the naturally recovering secondary forest at Me Linh Station is the secondary young forest with the low economic value due to shortly restored process (about 10-20 years), the flora is not rich and abundant, and there are only commonly pioneer and light demanding tree species. / Sinh khối và trữ lượng các bon của rừng tự nhiên ở Việt Nam vẫn ít được quan tâm của do hạn chế về kiến thức và tài chính. Trong bài báo này, chúng tôi đưa ra kết quả của việc ước lượng sinh khối và tổng hợp các bon của các thảm thực vật rừng thứ sinh phục hồi tự nhiên tại Trạm Đa dạng Sinh học Mê Linh, tỉnh Vĩnh Phúc- Việt Nam, nơi có loại hình thảm thực vật chính, đó là thảm thực vật phục hồi sau nương rẫy (kiểu thảm thục vật I) và thảm thực vật phục hồi sau khai thác kiệt (kiểu thảm thực vật II) nhằm mục đích đánh giá tiềm năng của rừng thứ sinh tại khu vực nghiên cứu. Sinh khối của thảm thực vật phục hồi sau nương rẫy là 86,80 tấn/ha. Sinh khối của thảm thực vật phục hồi sau khai thác cao hơn, đạt 131.59 tấn/ha. Lượng các bon hấp thu trong đất của thảm thực vật I là 79,01 tấn/ha và thảm thực vật II là 99,65 tấn/ha. Như vậy, tổng lượng các bon được hấp thu trong mỗi loại hình thảm thực vật trên là: 122,41 tấn/ha (thảm thực vật I) và 165,14 tấn/ha. Nhìn chung, rừng thứ sinh phục hồi tự nhiên tại Trạm Đa dạng Mê Linh chủ yếu là rừng non thứ sinh, ít có giá trị kinh tế do quá trình phục hồi diễn ra ngắn (khoảng 10-20 năm) nên thành phần thực vật nghèo nàn, không phong phú, thành phần chính chủ yếu là các cây gỗ tiên phong, ưa sáng.

En écologie, dans le cadre par exemple d’études des services fournis par les écosystèmes, les modélisations descriptive, explicative et prédictive ont toutes trois leur place distincte. Certaines situations bien précises requièrent soit l’un soit l’autre de ces types de modélisation ; le bon choix s’impose afin de pouvoir faire du modèle un usage conforme aux objectifs de l’étude. Dans le cadre de ce travail, nous explorons dans un premier temps le pouvoir explicatif de l’arbre de régression multivariable (ARM). Cette méthode de modélisation est basée sur un algorithme récursif de bipartition et une méthode de rééchantillonage permettant l’élagage du modèle final, qui est un arbre, afin d’obtenir le modèle produisant les meilleures prédictions. Cette analyse asymétrique à deux tableaux permet l’obtention de groupes homogènes d’objets du tableau réponse, les divisions entre les groupes correspondant à des points de coupure des variables du tableau explicatif marquant les changements les plus abrupts de la réponse. Nous démontrons qu’afin de calculer le pouvoir explicatif de l’ARM, on doit définir un coefficient de détermination ajusté dans lequel les degrés de liberté du modèle sont estimés à l’aide d’un algorithme. Cette estimation du coefficient de détermination de la population est pratiquement non biaisée. Puisque l’ARM sous-tend des prémisses de discontinuité alors que l’analyse canonique de redondance (ACR) modélise des gradients linéaires continus, la comparaison de leur pouvoir explicatif respectif permet entre autres de distinguer quel type de patron la réponse suit en fonction des variables explicatives. La comparaison du pouvoir explicatif entre l’ACR et l’ARM a été motivée par l’utilisation extensive de l’ACR afin d’étudier la diversité bêta. Toujours dans une optique explicative, nous définissons une nouvelle procédure appelée l’arbre de régression multivariable en cascade (ARMC) qui permet de construire un modèle tout en imposant un ordre hiérarchique aux hypothèses à l’étude. Cette nouvelle procédure permet d’entreprendre l’étude de l’effet hiérarchisé de deux jeux de variables explicatives, principal et subordonné, puis de calculer leur pouvoir explicatif. L’interprétation du modèle final se fait comme dans une MANOVA hiérarchique. On peut trouver dans les résultats de cette analyse des informations supplémentaires quant aux liens qui existent entre la réponse et les variables explicatives, par exemple des interactions entres les deux jeux explicatifs qui n’étaient pas mises en évidence par l’analyse ARM usuelle. D’autre part, on étudie le pouvoir prédictif des modèles linéaires généralisés en modélisant la biomasse de différentes espèces d’arbre tropicaux en fonction de certaines de leurs mesures allométriques. Plus particulièrement, nous examinons la capacité des structures d’erreur gaussienne et gamma à fournir les prédictions les plus précises. Nous montrons que pour une espèce en particulier, le pouvoir prédictif d’un modèle faisant usage de la structure d’erreur gamma est supérieur. Cette étude s’insère dans un cadre pratique et se veut un exemple pour les gestionnaires voulant estimer précisément la capture du carbone par des plantations d’arbres tropicaux. Nos conclusions pourraient faire partie intégrante d’un programme de réduction des émissions de carbone par les changements d’utilisation des terres. / In ecology, in ecosystem services studies for example, descriptive, explanatory and predictive modelling all have relevance in different situations. Precise circumstances may require one or the other type of modelling; it is important to choose the method properly to insure that the final model fits the study’s goal. In this thesis, we first explore the explanatory power of the multivariate regression tree (MRT). This modelling technique is based on a recursive bipartitionning algorithm. The tree is fully grown by successive bipartitions and then it is pruned by resampling in order to reveal the tree providing the best predictions. This asymmetric analysis of two tables produces homogeneous groups in terms of the response that are constrained by splitting levels in the values of some of the most important explanatory variables. We show that to calculate the explanatory power of an MRT, an appropriate adjusted coefficient of determination must include an estimation of the degrees of freedom of the MRT model through an algorithm. This estimation of the population coefficient of determination is practically unbiased. Since MRT is based upon discontinuity premises whereas canonical redundancy analysis (RDA) models continuous linear gradients, the comparison of their explanatory powers enables one to distinguish between those two patterns of species distributions along the explanatory variables. The extensive use of RDA for the study of beta diversity motivated the comparison between its explanatory power and that of MRT. In an explanatory perspective again, we define a new procedure called a cascade of multivariate regression trees (CMRT). This procedure provides the possibility of computing an MRT model where an order is imposed to nested explanatory hypotheses. CMRT provides a framework to study the exclusive effect of a main and a subordinate set of explanatory variables by calculating their explanatory powers. The interpretation of the final model is done as in nested MANOVA. New information may arise from this analysis about the relationship between the response and the explanatory variables, for example interaction effects between the two explanatory data sets that were not evidenced by the usual MRT model. On the other hand, we study the predictive power of generalized linear models (GLM) to predict individual tropical tree biomass as a function of allometric shape variables. Particularly, we examine the capacity of gaussian and gamma error structures to provide the most precise predictions. We show that for a particular species, gamma error structure is superior in terms of predictive power. This study is part of a practical framework; it is meant to be used as a tool for managers who need to precisely estimate the amount of carbon recaptured by tropical tree plantations. Our conclusions could be integrated within a program of carbon emission reduction by land use changes.

Airborne discrete-return lidar is an active remote sensing technology capable of obtaining accurate, fine-resolution three-dimensional measurements over large areas. Discrete-return lidar data produce three-dimensional object characterizations in the form of point clouds defined by precise x, y and z coordinates. The data also provide intensity values for each point that help quantify the reflectance and surface properties of intersected objects. These data features have proven to be useful for the characterization of many important forest attributes, such as standing tree biomass, height, density, and canopy cover, with new applications for the data currently accelerating. This dissertation explores three new applications for airborne discrete-return lidar data. The first application uses lidar-derived metrics to predict understory vegetation cover, which has been a difficult metric to predict using traditional explanatory variables. A new airborne lidar-derived metric, understory lidar cover density, created by filtering understory lidar points using intensity values, increased the coefficient of variation (R²) from non-lidar understory vegetation cover estimation models from 0.2-0.45 to 0.7-0.8. The method presented in this chapter provides the ability to accurately quantify understory vegetation cover (± 22%) at fine spatial resolutions over entire landscapes within the interior ponderosa pine forest type. In the second application, a new method for quantifying and locating snags using airborne discrete-return lidar is presented. The importance of snags in forest ecosystems and the inherent difficulties associated with their quantification has been well documented. A new semi-automated method using both 2D and 3D local-area lidar point filters focused on individual point spatial location and intensity information is used to identify points associated with snags and eliminate points associated with live trees. The end result is a stem map of individual snags across the landscape with height estimates for each snag. The overall detection rate for snags DBH ≥ 38 cm was 70.6% (standard error: ± 2.7%), with low commission error rates. This information can be used to: analyze the spatial distribution of snags over entire landscapes, provide a better understanding of wildlife snag use dynamics, create accurate snag density estimates, and assess achievement and usefulness of snag stocking standard requirements. In the third application, live above-ground biomass prediction models are created using three separate sets of lidar-derived metrics. Models are then compared using both model selection statistics and cross-validation. The three sets of lidar-derived metrics used in the study were: 1) a 'traditional' set created using the entire plot point cloud, 2) a 'live-tree' set created using a plot point cloud where points associated with dead trees were removed, and 3) a 'vegetation-intensity' set created using a plot point cloud containing points meeting predetermined intensity value criteria. The models using live-tree lidar-derived metrics produced the best results, reducing prediction variability by 4.3% over the traditional set in plots containing filtered dead tree points. The methods developed and presented for all three applications displayed promise in prediction or identification of unique forest attributes, improving our ability to quantify and characterize understory vegetation cover, snags, and live above ground biomass. This information can be used to provide useful information for forest management decisions and improve our understanding of forest ecosystem dynamics. Intensity information was useful for filtering point clouds and identifying lidar points associated with unique forest attributes (e.g., understory components, live and dead trees). These intensity filtering methods provide an enhanced framework for analyzing airborne lidar data in forest ecosystem applications. / Graduation date: 2013

En écologie, dans le cadre par exemple d’études des services fournis par les écosystèmes, les modélisations descriptive, explicative et prédictive ont toutes trois leur place distincte. Certaines situations bien précises requièrent soit l’un soit l’autre de ces types de modélisation ; le bon choix s’impose afin de pouvoir faire du modèle un usage conforme aux objectifs de l’étude. Dans le cadre de ce travail, nous explorons dans un premier temps le pouvoir explicatif de l’arbre de régression multivariable (ARM). Cette méthode de modélisation est basée sur un algorithme récursif de bipartition et une méthode de rééchantillonage permettant l’élagage du modèle final, qui est un arbre, afin d’obtenir le modèle produisant les meilleures prédictions. Cette analyse asymétrique à deux tableaux permet l’obtention de groupes homogènes d’objets du tableau réponse, les divisions entre les groupes correspondant à des points de coupure des variables du tableau explicatif marquant les changements les plus abrupts de la réponse. Nous démontrons qu’afin de calculer le pouvoir explicatif de l’ARM, on doit définir un coefficient de détermination ajusté dans lequel les degrés de liberté du modèle sont estimés à l’aide d’un algorithme. Cette estimation du coefficient de détermination de la population est pratiquement non biaisée. Puisque l’ARM sous-tend des prémisses de discontinuité alors que l’analyse canonique de redondance (ACR) modélise des gradients linéaires continus, la comparaison de leur pouvoir explicatif respectif permet entre autres de distinguer quel type de patron la réponse suit en fonction des variables explicatives. La comparaison du pouvoir explicatif entre l’ACR et l’ARM a été motivée par l’utilisation extensive de l’ACR afin d’étudier la diversité bêta. Toujours dans une optique explicative, nous définissons une nouvelle procédure appelée l’arbre de régression multivariable en cascade (ARMC) qui permet de construire un modèle tout en imposant un ordre hiérarchique aux hypothèses à l’étude. Cette nouvelle procédure permet d’entreprendre l’étude de l’effet hiérarchisé de deux jeux de variables explicatives, principal et subordonné, puis de calculer leur pouvoir explicatif. L’interprétation du modèle final se fait comme dans une MANOVA hiérarchique. On peut trouver dans les résultats de cette analyse des informations supplémentaires quant aux liens qui existent entre la réponse et les variables explicatives, par exemple des interactions entres les deux jeux explicatifs qui n’étaient pas mises en évidence par l’analyse ARM usuelle. D’autre part, on étudie le pouvoir prédictif des modèles linéaires généralisés en modélisant la biomasse de différentes espèces d’arbre tropicaux en fonction de certaines de leurs mesures allométriques. Plus particulièrement, nous examinons la capacité des structures d’erreur gaussienne et gamma à fournir les prédictions les plus précises. Nous montrons que pour une espèce en particulier, le pouvoir prédictif d’un modèle faisant usage de la structure d’erreur gamma est supérieur. Cette étude s’insère dans un cadre pratique et se veut un exemple pour les gestionnaires voulant estimer précisément la capture du carbone par des plantations d’arbres tropicaux. Nos conclusions pourraient faire partie intégrante d’un programme de réduction des émissions de carbone par les changements d’utilisation des terres. / In ecology, in ecosystem services studies for example, descriptive, explanatory and predictive modelling all have relevance in different situations. Precise circumstances may require one or the other type of modelling; it is important to choose the method properly to insure that the final model fits the study’s goal. In this thesis, we first explore the explanatory power of the multivariate regression tree (MRT). This modelling technique is based on a recursive bipartitionning algorithm. The tree is fully grown by successive bipartitions and then it is pruned by resampling in order to reveal the tree providing the best predictions. This asymmetric analysis of two tables produces homogeneous groups in terms of the response that are constrained by splitting levels in the values of some of the most important explanatory variables. We show that to calculate the explanatory power of an MRT, an appropriate adjusted coefficient of determination must include an estimation of the degrees of freedom of the MRT model through an algorithm. This estimation of the population coefficient of determination is practically unbiased. Since MRT is based upon discontinuity premises whereas canonical redundancy analysis (RDA) models continuous linear gradients, the comparison of their explanatory powers enables one to distinguish between those two patterns of species distributions along the explanatory variables. The extensive use of RDA for the study of beta diversity motivated the comparison between its explanatory power and that of MRT. In an explanatory perspective again, we define a new procedure called a cascade of multivariate regression trees (CMRT). This procedure provides the possibility of computing an MRT model where an order is imposed to nested explanatory hypotheses. CMRT provides a framework to study the exclusive effect of a main and a subordinate set of explanatory variables by calculating their explanatory powers. The interpretation of the final model is done as in nested MANOVA. New information may arise from this analysis about the relationship between the response and the explanatory variables, for example interaction effects between the two explanatory data sets that were not evidenced by the usual MRT model. On the other hand, we study the predictive power of generalized linear models (GLM) to predict individual tropical tree biomass as a function of allometric shape variables. Particularly, we examine the capacity of gaussian and gamma error structures to provide the most precise predictions. We show that for a particular species, gamma error structure is superior in terms of predictive power. This study is part of a practical framework; it is meant to be used as a tool for managers who need to precisely estimate the amount of carbon recaptured by tropical tree plantations. Our conclusions could be integrated within a program of carbon emission reduction by land use changes.