Remote phenology: Applying machine learning to detect phenological patterns in a cerrado savanna

Plant phenology has gained importance in the context of global change research, stimulating the development of new technologies for phenological observation. Digital cameras have been successfully used as multi-channel imaging sensors, providing measures of leaf color change information (RGB channels), or leafing phenological changes in plants. We monitored leaf-changing patterns of a cerrado-savanna vegetation by taken daily digital images. We extract RGB channels from digital images and correlated with phenological changes. Our first goals were: (1) to test if the color change information is able to characterize the phenological pattern of a group of species; and (2) to test if individuals from the same functional group may be automatically identified using digital images. In this paper, we present a machine learning approach to detect phenological patterns in the digital images. Our preliminary results indicate that: (1) extreme hours (morning and afternoon) are the best for identifying plant species; and (2) different plant species present a different behavior with respect to the color change information. Based on those results, we suggest that individuals from the same functional group might be identified using digital images, and introduce a new tool to help phenology experts in the species identification and location on-the-ground. ©2012 IEEE.

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.

Deriving vegetation indices for phenology analysis using genetic programming

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

Phenology of Caryocar brasiliense in the Brazilian cerrado region

The objective was to study the phenology of Caryocar brasiliense Camb. (Caryocaraceae), a type of fruit of the Brazilian cerrado, in three areas (cerrado, pasture and in the campus of the Federal University of Minas Gerais) of the municipality of Montes Claros, State of Minas Gerais, Brazil, during 3 years. The plants of C brasiliense presented linear growth with about 50 cm of height and 35 cm of crown width per year. Soils with loamy texture, dystrophic, rich in aluminum and with low pH favor the production of this species. C. brasiliense produced higher quantity of flowers in the middle third, followed by the apical and basal parts of the crown, while the fruit production was higher in the basal third, followed by the middle and apical parts of the canopy of this species. C brasiliense bloomed during the dry period and produced fruits in the rainy season. Its fruits presented about 6, 8 and 6 cm of width, length and height, respectively, and 160 g of gross weight. The extrativism without control has strong impact on the propagation of C brasiliense because only about 7.96 and 10.65% of plants up to 1.0 m height and 59.58 and 44.73% higher than 3.0 m (reproductive phase) was observed in the cerrado and pasture, respectively. This indicates that fruit collectors, practically, remove all fruits of this tree and thus they reduce, considerably, the propagation of C. brasiliense in the cerrado areas of Brazil. (c) 2006 Elsevier B.V. All rights reserved.

Projections Onto Convex Sets through Particle Swarm Optimization and its application for remote sensing image restoration

Image restoration attempts to enhance images corrupted by noise and blurring effects. Iterative approaches can better control the restoration algorithm in order to find a compromise of restoring high details in smoothed regions without increasing the noise. Techniques based on Projections Onto Convex Sets (POCS) have been extensively used in the context of image restoration by projecting the solution onto hyperspaces until some convergence criteria be reached. It is expected that an enhanced image can be obtained at the final of an unknown number of projections. The number of convex sets and its combinations allow designing several image restoration algorithms based on POCS. Here, we address two convex sets: Row-Action Projections (RAP) and Limited Amplitude (LA). Although RAP and LA have already been used in image restoration domain, the former has a relaxation parameter (A) that strongly depends on the characteristics of the image that will be restored, i.e., wrong values of A can lead to poorly restoration results. In this paper, we proposed a hybrid Particle Swarm Optimization (PS0)-POCS image restoration algorithm, in which the A value is obtained by PSO to be further used to restore images by POCS approach. Results showed that the proposed PSO-based restoration algorithm outperformed the widely used Wiener and Richardson-Lucy image restoration algorithms. (C) 2010 Elsevier B.V. All rights reserved.

Reproductive phenology of Euterpe edulis (Arecaceae) along a gradient in the Atlantic rainforest of Brazil

The palm Euterpe edulis Mart. is one of the dominant tree species in the Atlantic rainforest and considered a key resource for many frugivorous birds. We compared the reproductive phenology of E. edulis in three types of Atlantic rainforest (two lowland forests, restinga and coastal-plain, and a premontane forest) on Cardoso Island (Cananeia, São Paulo, Brazil), aiming to answer the following questions: (i) whether the reproduction of E. edulis is annual and seasonal across the years in the three forest types studied; (ii) what are the environmental factors influencing the reproductive phenology of E. edulis; and (iii) how does the timing of fruiting and fruit production of E. edulis vary among the three forest types? We evaluated the presence of flowers and fruits (immature, unripe and ripe) from August 2001 to July 2004 in 150 individuals (50 per forest), and estimated the number of infructescences with ripe fruits and the production of fruits and seeds by collecting them on the forest floor in the three forest types. Flowering and fruiting of E. edulis were annual and significantly seasonal in the three forest types, with a high synchrony of flowering and medium to low synchrony of fruiting. Flowering peaked in November and December, and immature and unripe fruits peaked in January and March, all during the rainy season. Immature and unripe fruit phases were correlated with the daylength, precipitation and temperature, important factors for fruits development. Ripe fruits peaked in April and May, in the less rainy season, with significant differences in the mean dates among forests. The number of infructescences with ripe fruits and the biomass of fruits and seeds collected on the ground also differed significantly among the forest types, being greater in the restinga and coastal plain forests, respectively. Differences in productivity were related to palm density in each area and the soil fertility. The complementary fruiting pattern of E. edulis in the forests studied may affect the distribution and abundance of certain frugivorous bird species that feed on their fruits.

Phenology, sex ratio, and spatial distribution among dioecious species of Trichilia (Meliaceae)

The flowering, sex ratio, and spatial distribution of four dioecious species of Trichilia (Meliaceae) were studied in a semi-deciduous forest in southeastern Brazil. All reproductive trees (T. clausseni, T. pallida and T. catigua) with dbh greater than or equal to5 cm within a 1-ha plot were collected, sexed, mapped and, for individuals of each species, the distances to the nearest neighbour of the same and opposite sex were measured. For the shrub species T elegans (dbh <5 cm), all reproductive individuals were sampled randomly in 10 samples of 10 x 10 m. The reproductive phenology was observed at weekly to monthly intervals from May 1988 to January 1990. The species are strictly dioecious, did not present any sex-mixed trees or sex switching during the study, and sex ratio did not differ significantly from 1:1. The size distributions and the relative size variation were not significantly different. between sexes. There was no significant segregation or clumping between individuals of either sex and no fruit production without pollination. Onset of flowering and flowering peak were synchronous between male and female plants for all species studied. Flower synchrony was related to outcrossing and pollinator attraction rather than climatic factors.

Temporal variation in the abundance of two species of thrushes in relation to fruiting phenology in the Atlantic rainforest

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

Reproductive phenology of coastal plain Atlantic forest vegetation: comparisons from seashore to foothills

The diversity of tropical forest plant phenology has called the attention of researchers for a long time. We continue investigating the factors that drive phenological diversity on a wide scale, but we are unaware of the variation of plant reproductive phenology at a fine spatial scale despite the high spatial variation in species composition and abundance in tropical rainforests. We addressed fine scale variability by investigating the reproductive phenology of three contiguous vegetations across the Atlantic rainforest coastal plain in Southeastern Brazil. We asked whether the vegetations differed in composition and abundance of species, the microenvironmental conditions and the reproductive phenology, and how their phenology is related to regional and local microenvironmental factors. The study was conducted from September 2007 to August 2009 at three contiguous sites: (1) seashore dominated by scrub vegetation, (2) intermediary covered by restinga forest and (3) foothills covered by restinga pre-montane transitional forest. We conducted the microenvironmental, plant and phenological survey within 30 transects of 25 mx4 m (10 per site). We detected significant differences in floristic, microenvironment and reproductive phenology among the three vegetations. The microenvironment determines the spatial diversity observed in the structure and composition of the flora, which in turn determines the distinctive flowering and fruiting peaks of each vegetation (phenological diversity). There was an exchange of species providing flowers and fruits across the vegetation complex. We conclude that plant reproductive patterns as described in most phenological studies (without concern about the microenvironmental variation) may conceal the fine scale temporal phenological diversity of highly diverse tropical vegetation. This phenological diversity should be taken into account when generating sensor-derived phenologies and when trying to understand tropical vegetation responses to environmental changes.

Reproductive phenology of a northeast Brazilian mangrove community: Environmental and biotic constraints

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Lek phenology of the White-bearded Manakin (Manacus manacus, Aves: Passeriformes: Pipridae) in a subtropical region

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Natural history of Peucetia flava (Araneae, Oxyopidae): seasonal density fluctuation, phenology and sex ratio on the glandular plant Rhyncanthera dichotoma (Melastomataceae)

The seasonal density fluctuation, phenology and sex ratio of Peucetia flava (Oxyopidae) on Rhyncanthera dichotoma (Melastomataceae) were investigated during a year in a swamp in southeastern Brazil. Peucetia flava displayed an unusual non-seasonal life cycle and the population size varied little over the year. The density of Peucetia spiders increased with the increase in abundance of leaves and number of arthropods adhered to glandular trichomes on R. dichotoma leaves. Our findings suggest that seasonal density fluctuation of Peucetia may be influenced by foraging site availability (i.e. leaves) and prey supply (i.e. arthropods adhering to glandular trichomes). The unusual seasonal stability of P. flava may be related to the type of habitat in which this spider occurs (swamp), because of the constant input of allochthonous resources from the water source, i.e. aquatic insects that migrate to the terrestrial environment.

Remote sensing analysis of depositional landforms in alluvial settings: Method development and application to the Taquari megafan, Pantanal (Brazil)

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

Edge effect and phenology in Erythroxylum tortuosum (Erythroxylaceae), a typical plant of the Brazilian Cerrado

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

Effects of environmental conditions associated to the cardinal orientation on the reproductive phenology of the cerrado savanna tree Xylopia aromatica (Annonaceae)

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