L’arbre de régression multivariable et les modèles linéaires généralisés revisités : applications à l’étude de la diversité bêta et à l’estimation de la biomasse d’arbres tropicaux

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.

Reforesting degraded agricultural landscapes with Eucalypts: effects on carbon storage and soil fertility after 26years

In the Western Australian wheatbelt, the restoration of native eucalypt forests for managing degraded agricultural landscapes is a critical part of managing dryland salinity and rebuilding biodiversity. Such reforestation will also sequester carbon. Whereas most investigative emphasis has been on carbon stored in biomass, the effects of reforestation on soil organic carbon (SOC) stores and fertility are not known. Two 26 year old reforestation experiments with four Eucalyptus species (E. cladocalyx var nana, E. occidentalis, E. sargentii and E. wandoo) were compared with agricultural sites (Field). SOC stores (to 0.3 m depth) ranged between 33 and 55 Mg ha−1, with no statistically significant differences between tree species and adjacent farmland. Farming comprised crop and pasture rotations. In contrast, the reforested plots contained additional carbon in the tree biomass (23–60 Mg ha−1) and litter (19–34 Mg ha−1), with the greatest litter accumulation associated with E. sargentii. Litter represented between 29 and 56% of the biomass carbon and the protection or utilization of this litter in fire-prone, semi-arid farmland will be an important component of carbon management. Exch-Na and Exch-Mg accumulated under E. sargentii and E. occidentalis at one site. The results raise questions about the conclusions of SOC sequestration studies following reforestation based on limited sampling and reiterate the importance of considering litter in reforestation carbon accounts.

Relações ecologicas entre caranguejos e composição arbórea em bosques de mangue naturais e restaurados no nordeste brasileiro

This study aimed to compare the development of crab and tree communities of two restored mangrove areas, one planted with Rhizophora mangle and the other naturally recovered, and also to compare the predation of Grapsid crab Goniopsis cruentata and the Ocypodid Ucides cordatus over the propagules of three mangrove trees: Rhizophora mangle, Avicennia schaueriana e Laguncularia racemosa. Specifically, we tested the hypothesis that Goniopsis predation is more important that Ucides predation, and that these consumers have antagonist effects over propagule consumption. In each area, 10 quadrates were selected at random to analyze tree richness, diameter, height, tree biomass and crab richness and density five years after restoration experiment start. Results show that tree height, biomass and crab density were significantly higher in artificially restored area. No significant differences were observed in crab species richness between areas, but higher tree richness was observed in self-recovered area. Results suggest that planting propagules of Rhizophora can significantly increase tree recovering if the aim was increase tree biomass and crab density, which can accelerate return of ecological functionality. Goniopsis is a more important propagule predator than Ucides both in natural and restored areas. The effects of Goniopis were higher in absence of Ucides, due to negative interactions among these two predator species. The preference of Goniopsis by Avicennia and Laguncularia can favor the dominance of Rhizophora observed in Neotropical mangroves. This study suggests that propagule predation by Goniopsis should be controlled in restoration programs, if dominance of Rhizophora is undesirable respect to more rich tree communities

Vegetation, soil and site characteristics of samples taken along the Yamal Transet in 2007-2008

The overarching goal of the Yamal portion of the Greening of the Arctic project is to examine how the terrain and anthropogenic factors of reindeer herding and resource development combined with the climate variations on the Yamal Peninsula affect the spatial and temporal patterns of vegetation change and how these changes are in turn affecting traditional herding of the indigenous people of the region. The purpose of the expeditions was to collect groundobservations in support of remote sensing studies at four locations along a transect that traverses all the major bioclimate subzones of the Yamal Peninsula. This data report is a summary of information collected during the 2007 and 2008 expeditions. It includes all the information from the 2008 data report (Walker et al. 2008) plus new information collected at Kharasavey in Aug 2008. The locations included in this report are Nadym (northern taiga subzone), Laborovaya (southern tundra = subzone E of the Circumpolar Arctic Vegetation Map (CAVM), Vaskiny Dachi (southern typical tundra = subzone D), and Kharasavey (northern typical tundra = subzone C). Another expedition is planned for summer 2009 to the northernmost site at Belyy Ostrov (Arctic tundra = subzone B). Data are reported from 10 study sites - 2 at Nadym, 2 at Laborovaya, and 3 at Vaskiny Dachi and 3 at Kharasavey. The sites are representative of the zonal soils and vegetation, but also include variation related to substrate (clayey vs. sandy soils). Most of the information was collected along 5 transects at each sample site, 5 permanent vegetation study plots, and 1-2 soil pits at each site. The expedition also established soil and permafrost monitoring sites at each location. This data report includes: (1) background for the project, (2) general descriptions and photographs of each locality and sample site, (3) maps of the sites, study plots, and transects at each location, (4) summary of sampling methods used, (5) tabular summaries of the vegetation data (species lists, estimates of cover abundance for each species within vegetation plots, measured percent ground cover of species along transects, site factors for each study plot), (6) summaries of the Normalized Difference Vegetation Index (NDVI) and leaf area index (LAI) along each transect, (7) soil descriptions and photos of the soil pits at each study site, (8) summaries of thaw measurements along each transect, and (9) contact information for each of the participants. One of the primary objectives was to provide the Russian partners with full documentation of the methods so that Russian observers in future years could repeat the observations independently.

Importancia de la caracterización de la masa de raíces en la simulación de ecosistemas forestales

Los análisis de sensibilidad son una herramienta importante para comprender el funcionamiento de los modelos ecológicos, así como para identificar los parámetros más importantes en su funcionamiento. Además, los análisis de sensibilidad pueden utilizarse para diseñar de forma más efectiva planes de muestreo de campo dirigidos a calibrar los modelos ecológicos. En los estudios de ecosistemas forestales, el análisis cuantitativo de la parte subterránea es mucho más costoso y complicado que el estudio de la parte aérea, en especial el estudio de la dinámica de producción y descomposición de raíces gruesas y finas de los árboles. En este trabajo se muestra un ejemplo de análisis de sensibilidad del modelo forestal FORECAST a parámetros que definen la biomasa, longevidad y concentración de nitrógeno en las raíces de los árboles. El modelo se calibró para simular dos rodales de pino silvestre (Pinus sylvestris) en los Pirineos de Navarra. Los resultados indican que la tasa de renovación de raíces finas es el parámetro más influyente en las estimaciones del modelo de crecimiento de los árboles, seguida de la concentración de N en las mismas, siendo la relación biomasa subterránea/total el parámetro al cual el modelo es menos sensible. Además, el modelo es más sensible a los parámetros que definen el componente subterráneo de la biomasa arbórea cuando simula un sitio de menor capacidad productiva y mayor limitación por nutrientes.

Nitrogen balance for an agroforestry system irrigated with saline, high nitrogen effluent

Land disposal is commonly used for urban and industrial wastewater, largely due to the high costs involved in alternative treatments or disposal systems. However, the viability of such systems depends on many factors, including the composition of the effluent water, soil type, the plant species grown, growth rate, and planting density. The objective of this study is to establish whether land disposal of nitrogen (N) rich effluent using an agroforestry system is sustainable, and determine the effect of irrigation rate and tree planting density on the N cycle and subsequent N removal. We examined systems for the sustainable disposal of a high strength industrial effluent. The challenge was to leach the salt, by using a sufficiently high rate of irrigation, while simultaneously ensuring that N did not leach from the soil profile. We describe the N balance for two plant systems irrigated with effluent, one comprising Eucalyptus tereticornis and Eucalyptus moluccana and a Rhodes grass (Chloris gayana) pasture, and the other, Rhodes grass pasture alone. Nitrogen balance was assessed from N inputs in effluent and rainfall, accumulation of N in the plant biomass, changes in soil N storage, N loss in run-off water, denitrification and N loss to the groundwater by deep-drainage. Biomass production was estimated from allometric relationships derived from yearly destructive harvesting of selected trees. The N content of that biomass was then calculated from measured N content of the various plant parts, and their mass. Approximately 300 kg N/ha/yr was assimilated into tree biomass at a planting density of 2500 tree/ha of E. moluccana. In addition to tree assimilation, pasture growth between the tree rows, which was regularly harvested, contributed substantially to N uptake. If the trees were harvested after two years of growth and grass harvested regularly, biomass removal of N by the mixed system would be about 700 kg N/ha/yr. The results of this study show that the current system of effluent disposal is not sustainable as the nitrate leaching from the soil profile far exceeds standards set out by the ANZECC guidelines. Hence additional means of N removal will need to be implemented. Biological N removal is an area that warrants further studies as it is aimed at reducing N levels in the effluent before irrigation. This will complement the current agroforestry system.

Biomass and nutrient cycling in pure and mixed stands of native tree species in southeastern Bahia, Brazil

The objective of this paper is to study selected components of the nutrient cycle of pure and mixed stands of native forest species of Atlantic Forest in southeastern Brazil. Tree diameter, height, above-ground biomass, and nutrient content were determined in 22-year-old stands. Litterfall, litter decomposition, and nutrient concentration were evaluated from August 1994 to July 1995. The following species were studied: Peltogyne angustiflora, Centrolobium robustum, Arapatiella psilophylla, Sclerolobium chrysophyllum, Cordia trichotoma, Macrolobium latifolium. The litter of a natural forest and a 40-year-old naturally regenerated second-growth forest was sampled as well. The mixed-species outmatched pure stands in height, stem volume and total biomass (29.4 % more). The greatest amount of forest litter was observed in the natural forest (9.3 Mg ha-1), followed by the mixed-species stand (7.6 Mg ha-1) and secondary forest (7.3 Mg ha-1), and least litterfall was measured in the pure C. robustum stand (5.5 Mg ha-1). Litterfall seasonality varied among species in pure stands (CV from 44.7 to 91.4 %), unlike litterfall in the mixed-tree stand, where the variation was lower (CV 31.2 %). In the natural and second-growth forest, litterfall varied by 57.8 and 34.0 %, respectively. The annual rate of nutrient return via litterfall varied widely among forest ecosystems. Differences were detected between forest ecosystems in both the litter accumulation and quantity of litterlayer nutrients. The highest mean nutrient accumulation in above-ground biomass was observed in mixed-species stands. The total nutrient accumulation (N + P + K+ Ca + Mg) ranged from 0.97 to 1.93 kg tree-1 in pure stands, and from 1.21 to 2.63 kg tree-1 in mixed-species stands. Soil fertility under mixed-species stands (0-10 cm) was intermediate between the primary forest and pure-stand systems. The litterfall rate of native forest species in a mixed-species system is more constant, resulting in a more continuous decomposition rate. Consequently, both nutrient availability and quantity of organic matter in the soil are higher and the production system ecologically more sustainable.

Biomass of tree species as a response to planting density and interspecific competition

Planting trees is an important way to promote the recovery of degraded areas in the Caatinga region. Experiments (E1, E2, and E3) were conducted in a randomized blocks design, with three, three, and five replicates, respectively. The objectives were to evaluate biomass of the shoots of: a) gliricidia (G) and sabiá (S), as a response to planting density; b) G, S, and neem (N) in competition; c) G, and S in agroforestry. E1 was conducted in split-plots, and planting densities (400, 600, 800, 1000, and 1200 plants ha-1) as subplots. E2 consisted of a factorial comprising the following plots: GGG, NGN, SGS, NNN, GNG, SNS, SSS, GSG, NSN (each letter represents a row of plants). E3 was conducted with G and S in agroforestry experiment. The trees were harvested after 54, 42, and 27 months old, in E1, E2 and E3, respectively. In E1, G presented higher green biomass of the stems and leaf at smaller densities than S, but lower green biomass of branches at most densities. The species did not differ for mean stem dry biomass and leaf dry biomass, but G showed higher branch dry biomass at most densities. Higher planting densities increased green and dry biomass of stems, branches, and leaves in S, but decreased those characteristics in G, with the exception of leaf dry mass, which was not influenced by density. In E2, the behavior of each species was identical in plots containing the same or different species. Griricidia showed the highest green biomass of stems and branches, and the highest values for geren biomass of the leaf were observed for gliricidia and neem. The highest stem, branch, and leaf dry biomass values were obtained for G, S, and N, respectively. In E3, G was superior for stem and leaf green biomass, and for stem and branch dry biomass. There were no differences between species for the other biomass values.

Biomass partitioning and growth efficiency in four naturally regenerated forest tree species

Current forest growth models and yield tables are almost exclusively based on data from mature trees, reducing their applicability to young and developing stands. To address this gap, young European beech, sessile oak, Scots pine and Norway spruce trees approximately 0 to 10 years old were destructively sampled in a range of naturally regenerated forest stands in Central Europe. Diameter at base and height were first measured in situ for up to 175 individuals per species. Subsequently, the trees were excavated and dry biomass of foliage, branches, stems and roots was measured. Allometric relations were then used to calculate biomass allocation coefficients (BAC) and growth efficiency (GE) patterns in young trees. We found large differences in BAC and GE between broadleaves and conifers, but also between species within these categories. Both BAC and GE are strongly age-specific in young trees, their rapidly changing values reflecting different growth strategies in the earliest stages of growth. We show that linear relationships describing biomass allocation in older trees are not applicable in young trees. To accurately predict forest biomass and carbon stocks, forest growth models need to include species and age specific parameters of biomass allocation patterns.

Adaptation of tree growth to elevated CO2: quantitative trait loci for biomass in Populus

Information on the genetic variation of plant response to elevated CO2 (e[CO2]) is needed to understand plant adaptation and to pinpoint likely evolutionary response to future high atmospheric CO2 concentrations.• Here, quantitative trait loci (QTL) for above- and below-ground tree growth were determined in a pedigree – an F2 hybrid of poplar (Populus trichocarpa and Populus deltoides), following season-long exposure to either current day ambient CO2 (a[CO2]) or e[CO2] at 600 µl l−1, and genotype by environment interactions investigated.• In the F2 generation, both above- and below-ground growth showed a significant increase in e[CO2]. Three areas of the genome on linkage groups I, IX and XII were identified as important in determining above-ground growth response to e[CO2], while an additional three areas of the genome on linkage groups IV, XVI and XIX appeared important in determining root growth response to e[CO2].• These results quantify and identify genetic variation in response to e[CO2] and provide an insight into genomic response to the changing environment

Aluminum toxicity to tropical montane forest tree seedlings in southern Ecuador: response of biomass and plant morphology to elevated Al concentrations

In acid tropical forest soils (pH < 5.5) increased mobility of aluminum might limit aboveground productivity. Therefore, we evaluated Al phytotoxicity of three native tree species of tropical montane forests in southern Ecuador. An hydroponic dose-response experiment was conducted. Seedlings of Cedrela odorata L., Heliocarpus americanus L., and Tabebuia chrysantha (Jacq.) G. Nicholson were treated with 0, 300, 600, 1200, and 2400 mu M Al and an organic layer leachate. Dose-response curves were generated for root and shoot morphologic properties to determine effective concentrations (EC). Shoot biomass and healthy leaf area decreased by 44 % to 83 % at 2400 mu M Al, root biomass did not respond (C. odorata), declined by 51 % (H. americanus), or was stimulated at low Al concentrations of 300 mu M (T. chrysantha). EC10 (i.e. reduction by 10 %) values of Al for total biomass were 315 mu M (C. odorata), 219 mu M (H. americanus), and 368 mu M (T. chrysantha). Helicarpus americanus, a fast growing pioneer tree species, was most sensitive to Al toxicity. Negative effects were strongest if plants grew in organic layer leachate, indicating limitation of plant growth by nutrient scarcity rather than Al toxicity. Al toxicity occurred at Al concentrations far above those in native organic layer leachate.