Applying machine learning based on multiscale classifiers to detect remote phenology patterns in Cerrado savanna trees

Plant phenology is one of the most reliable indicators of species responses to global climate change, motivating the development of new technologies for phenological monitoring. Digital cameras or near remote systems have been efficiently applied as multi-channel imaging sensors, where leaf color information is extracted from the RGB (Red, Green, and Blue) color channels, and the changes in green levels are used to infer leafing patterns of plant species. In this scenario, texture information is a great ally for image analysis that has been little used in phenology studies. We monitored leaf-changing patterns of Cerrado savanna vegetation by taking daily digital images. We extract RGB channels from the digital images and correlate them with phenological changes. Additionally, we benefit from the inclusion of textural metrics for quantifying spatial heterogeneity. Our first goals are: (1) to test if color change information is able to characterize the phenological pattern of a group of species; (2) to test if the temporal variation in image texture is useful to distinguish plant species; and (3) to test if individuals from the same species may be automatically identified using digital images. In this paper, we present a machine learning approach based on multiscale classifiers to detect phenological patterns in the digital images. Our results indicate that: (1) extreme hours (morning and afternoon) are the best for identifying plant species; (2) different plant species present a different behavior with respect to the color change information; and (3) texture variation along temporal images is promising information for capturing phenological patterns. Based on those results, we suggest that individuals from the same species and functional group might be identified using digital images, and introduce a new tool to help phenology experts in the identification of new individuals from the same species in the image and their location on the ground. © 2013 Elsevier B.V. All rights reserved.

Source-driven remobilizations of nutrients within stem wood in Eucalyptus grandis plantations

Nutrient remobilizations in tree ligneous components have been little studied in tropical forests. A complete randomized block design was installed in Brazilian eucalypt plantations to quantify the remobilizations of phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), and sodium (Na) within stem wood. Three treatments were studied: control with neither K nor Na addition (C), 3 kmol ha-1 K applied (+K), and 3 kmol ha-1 Na applied (+Na). Biomass and nutrient contents were measured in the stem wood of eight trees destructively sampled at 1, 2, 3 and 4 years after planting in each treatment and annual rings were localized on discs of wood sampled every 3 m in half of the trees. Chemical analyses and wood density measurements were performed individually for each ring per level and per tree sampled. Nutrient remobilizations in annual rings were calculated through mass balance between two successive ages. Our results show that nutrient remobilizations within stem wood were mainly source-driven. Potassium and Na additions largely increased their concentration in the outer rings as well as the amounts remobilized in the first 2 years after the wood formation. The amount of Na remobilized in annual rings was 15 % higher in +Na than in +K the fourth year after planting despite a 34 % higher production of stem wood in +K leading to a much higher nutrient sink. A partial substitution of K by Na in the remobilizations within stem wood might contribute to enhancing Eucalyptus grandis growth in K-depleted soils. © 2013 Springer-Verlag Berlin Heidelberg.

Disturbances, elevation, topography and spatial proximity drive vegetation patterns along an altitudinal gradient of a top biodiversity hotspot

The correlation between vegetation patterns (species distribution and richness) and altitudinal variation has been widely reported for tropical forests, thereby providing theoretical basis for biodiversity conservation. However, this relationship may have been oversimplified, as many other factors may influence vegetation patterns, such as disturbances, topography and geographic distance. Considering these other factors, our primary question was: is there a vegetation pattern associated with substantial altitudinal variation (10-1,093 m a.s.l.) in the Atlantic Rainforest-a top hotspot for biodiversity conservation-and, if so, what are the main factors driving this pattern? We addressed this question by sampling 11 1-ha plots, applying multivariate methods, correlations and variance partitioning. The Restinga (forest on sandbanks along the coastal plains of Brazil) and a lowland area that was selectively logged 40 years ago were floristically isolated from the other plots. The maximum species richness (>200 spp. per hectare) occurred at approximately 350 m a.s.l. (submontane forest). Gaps, multiple stemmed trees, average elevation and the standard deviation of the slope significantly affected the vegetation pattern. Spatial proximity also influenced the vegetation pattern as a structuring environmental variable or via dispersal constraints. Our results clarify, for the first time, the key variables that drive species distribution and richness across a large altitudinal range within the Atlantic Rainforest. © 2013 Springer Science+Business Media Dordrecht.

Ecologia de hemiepífitas estranguladoras no Parque Estadual da Ilha do Cardoso

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Reproductive phenology and fruit production on a land bridge island in the brazilian atlantic forest

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Frugivoria e efetividade de dispersão de sementes dos últimos grandes frugívoros da Mata Atlântica: a anta (Tapirus terrestris) e o muriqui (Brachyteles arachnoides)

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Florística e distribuição de epífitas vasculares em floresta alta de restinga na planície litorânea da praia da fazenda, núcleo Picinguaba, parque estadual Serra do Mar, município de Ubatuba, SP

Pós-graduação em Ciências Biológicas (Biologia Vegetal) - IBRC

Produtividade e nutrição em três acessos de Baccharis trimera em função da adubação organo-mineral, em casa de vegetação

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Efeito de distúrbios ambientais sobre a fauna de cupins (Insecta: isoptera) e seu papel como bioindicador

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Composição da avifauna e frugivoria por ave em um mosaico sucessional na Mata Atlântica

Pós-graduação em Ciências Biológicas (Zoologia) - IBRC

Bases para o manejo sustentável de populações silvestres de Heliconia velloziana Emygdio

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Fragmentação florestal: efeitos em múltipla escala sobre a diversidade de aves em remanescentes florestais no noroeste do Estado de São Paulo

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Fenologia de espécies de floresta atlântica, núcleo Picinguaba, Parque Estadual da Serra do Mar, Estado de São Paulo: comparação entre estratos e influência de borda natural

Pós-graduação em Ciências Biológicas (Biologia Vegetal) - IBRC

Critérios para a amostragem de lianas: comparação e estimativa da abundância e biomassa de lianas no Cerrado

Lianas play a key role in many tropical forests ecosystems, contributing to community abundance, diversity and structure. Techniques to census lianas and estimate biomass have been used mainly in tropical forests. The aim of this study was to verify if different measurement location on the stem and inclusion criteria influence the estimations of liana abundance and biomass in Cerrado. The study was carried out at Itirapina fragment (ca 260 ha) located in Sao Paulo, Southeast Brazil. We measured all apparent ramets of rooted lianas by using different inclusion criteria: >= 0.5 cm in diameter at 30 cm basal area (D_30), and at 130 cm diameter breast height (DBH), in 36 transects (10 x 25 m). We compared the biomass using the allometric equation based on data from tropical florest take at 130 cm (DBH) and other The allometric biomass equation development to estimate the aboveground biomass of live lianas in tropical forests (DAP = 130 cm) was used. We tested the liana biomass, based on measurements taken at (D_30) using two different conversion equations. The results did not show significant differences in liana abundance and above ground biomass estimated applying the different criteria of inclusion However the biomass estimated from data collected at D_30 and converted into DAP showed a tendency to increase in biomass. Our findings suggest that the protocol used to the census of liana and the allometric biomass equation proposed to forests can be applied in the Cerrado.

Frugivory and the effects of ingestion by bats on the seed germination of three pioneering plants

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)