Toward Satellite-Based Land Cover Classification Through Optimum-Path Forest

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

Modelagem da biomassa acima do solo (BAS), por meio de imagens de alta resolução espacial e índices de vegetação

Pós-graduação em Agronomia - FEIS

Estimation of croplands using indicator kriging and fuzzy classification

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Semiautomatic object-space road extraction combining a stereoscopic image pair and a TIN-Based DTM

In this letter, a semiautomatic method for road extraction in object space is proposed that combines a stereoscopic pair of low-resolution aerial images with a digital terrain model (DTM) structured as a triangulated irregular network (TIN). First, we formulate an objective function in the object space to allow the modeling of roads in 3-D. In this model, the TIN-based DTM allows the search for the optimal polyline to be restricted along a narrow band that is overlaid upon it. Finally, the optimal polyline for each road is obtained by optimizing the objective function using the dynamic programming optimization algorithm. A few seed points need to be supplied by an operator. To evaluate the performance of the proposed method, a set of experiments was designed using two stereoscopic pairs of low-resolution aerial images and a TIN-based DTM with an average resolution of 1 m. The experimental results showed that the proposed method worked properly, even when faced with anomalies along roads, such as obstructions caused by shadows and trees.

Vibrational and rotational spectroscopy of (CD3OD)-C-13

We have used Fourier Transform spectral data on the C-O stretching mode of (CD3OD)-C-13 in order to perform a vibro-rotational analysis for this molecule. We have estimated a few molecular parameters of the ground and C-O stretching vibrational modes. Based on these parameters, and by using the Kwan-Dennison model, we propose assignments for a number of far-infrared laser transitions of (CD3OD)-C-13.

Diagnóstico da qualidade ambiental nas áreas verdes públicas em Presidente Prudente (SP)

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Impact of the Manaus urban plume on trace gas mixing ratios near the surface in the Amazon Basin: Implications for the NO-NO2-O-3 photostationary state and peroxy radical levels

We measured the mixing ratios of NO, NO2, O-3, and volatile organic carbon as well as the aerosol light-scattering coefficient on a boat platform cruising on rivers downwind of the city of Manaus (Amazonas State, Brazil) in July 2001 (Large-Scale Biosphere-Atmosphere Experiment in Amazonia-Cooperative LBA Airborne Regional Experiment-2001). The dispersion and impact of the Manaus plume was investigated by a combined analysis of ground-based (boat platform) and airborne trace gas and aerosol measurements as well as by meteorological measurements complemented by dispersion calculations (Hybrid Single-Particle Lagrangian Integrated Trajectory model). For the cases with the least anthropogenic influence (including a location in a so far unexplored region similar to 150 km west of Manaus on the Rio Manacapuru), the aerosol scattering coefficient, sigma(s), was below 11 Mm(-1), NOx mixing ratios remained below 0.6 ppb, daytime O-3 mixing ratios were mostly below 20 ppb and maximal isoprene mixing ratios were about 3 ppb in the afternoon. The photostationary state (PSS) was not established for these cases, as indicated by values of the Leighton ratio, Phi, well above unity. Due to the influence of river breeze systems and other thermally driven mesoscale circulations, a change of the synoptic wind direction from east-northeast to south-southeast in the afternoon often caused a substantial increase of ss and trace gas mixing ratios (about threefold for sigma(s), fivefold for NOx, and twofold for O-3), which was associated with the arrival of the Manaus pollution plume at the boat location. The ratio F reached unity within its uncertainty range at NOx mixing ratios of about 3 ppb, indicating "steady-state" conditions in cases when radiation variations, dry deposition, emissions, and reactions mostly involving peroxy radicals (XO2) played a minor role. The median midday/afternoon XO2 mixing ratios estimated using the PSS method range from 90 to 120 parts per trillion (ppt) for the remote cases (sigma(s) < 11 Mm(-1) and NOx < 0.6 ppb), while for the polluted cases our estimates are 15 to 60 ppt. These values are within the range of XO2 estimated by an atmospheric chemistry box model (Chemistry As A Box model Application-Module Efficiently Calculating the Chemistry of the Atmosphere (CAABA/MECCA)-3.0).

El Niño southern oscillation and its effect on fog oases along the Peruvian and Chilean coastal deserts

Fog oases, locally named Lomas, are distributed in a fragmented way along the western coast of Chile and Peru (South America) between ~6°S and 30°S following an altitudinal gradient determined by a fog layer. This fragmentation has been attributed to the hyper aridity of the desert. However, periodically climatic events influence the ‘normal seasonality’ of this ecosystem through a higher than average water input that triggers plant responses (e.g. primary productivity and phenology). The impact of the climatic oscillation may vary according to the season (wet/dry). This thesis evaluates the potential effect of climate oscillations, such as El Niño Southern Oscillation (ENSO), through the analysis of vegetation of this ecosystem following different approaches: Chapters two and three show the analysis of fog oasis along the Peruvian and Chilean deserts. The objectives are: 1) to explain the floristic connection of fog oases analysing their taxa composition differences and the phylogenetic affinities among them, 2) to explore the climate variables related to ENSO which likely affect fog production, and the responses of Lomas vegetation (composition, productivity, distribution) to climate patterns during ENSO events. Chapters four and five describe a fog-oasis in southern Peru during the 2008-2010 period. The objectives are: 3) to describe and create a new vegetation map of the Lomas vegetation using remote sensing analysis supported by field survey data, and 4) to identify the vegetation change during the dry season. The first part of our results show that: 1) there are three significantly different groups of Lomas (Northern Peru, Southern Peru, and Chile) with a significant phylogenetic divergence among them. The species composition reveals a latitudinal gradient of plant assemblages. The species origin, growth-forms typologies, and geographic position also reinforce the differences among groups. 2) Contradictory results have emerged from studies of low-cloud anomalies and the fog-collection during El Niño (EN). EN increases water availability in fog oases when fog should be less frequent due to the reduction of low-clouds amount and stratocumulus. Because a minor role of fog during EN is expected, it is likely that measurements of fog-water collection during EN are considering drizzle and fog at the same time. Although recent studies on fog oases have shown some relationship with the ENSO, responses of vegetation have been largely based on descriptive data, the absence of large temporal records limit the establishment of a direct relationship with climatic oscillations. The second part of the results show that: 3) five different classes of different spectral values correspond to the main land cover of Lomas using a Vegetation Index (VI). The study case is characterised by shrubs and trees with variable cover (dense, semi-dense and open). A secondary area is covered by small shrubs where the dominant tree species is not present. The cacti area and the old terraces with open vegetation were not identified with the VI. Agriculture is present in the area. Finally, 4) contrary to the dry season of 2008 and 2009 years, a higher VI was obtained during the dry season of 2010. The VI increased up to three times their average value, showing a clear spectral signal change, which coincided with the ENSO event of that period.

The development and deployment of FOTOS: a fast observation of trace organics system

Flüchtige organische Bestandteile (engl.: VOC) sind in der Atmosphäre in Spuren vorhanden, spielen aber trotzdem eine wichtige Rolle in der Luftchemie: sie beeinflussen das Ozon der Troposphäre, städtischen Smog, Oxidationskapazität und haben direkte und indirekte Auswirkungen auf die globale Klimaveränderung. Eine wichtige Klasse der VOC sind die Nicht-Methan-Kohlenwasserstoffe (engl.: NMHC), die überwiegend von anthropogenen Quellen kommen. Aus diesem Grund ist für Luftchemiker ein Messinstrument nötig, das die VOC, die NMHC eingeschlossen, mit einer höheren Zeitauflösung misst, besonders für Echtzeitmessungen an Bord eines Forschungsflugzeuges. Dafür wurde das System zur schnellen Beobachtung von organischen Spuren (engl.: FOTOS) entworfen, gebaut für den Einsatz in einem neuen Wissenschaftlichen Flugzeug, das in großen Höhen und über weite Strecken fliegt, genannt HALO. In der Folge wurde FOTOS in zwei Messkampagnen am Boden getestet. FOTOS wurde entworfen und gebaut mit einem speziell angefertigten, automatisierten, kryogenen Probensystem mit drei Fallen und einem angepassten, erworbenen schnellen GC-MS. Ziel dieses Aufbaus war es, die Vielseitigkeit zu vergrößern und das Störungspotential zu verringern, deshalb wurden keine chemischen Trocknungsmittel oder adsorbierenden Stoffe verwendet. FOTOS erreichte eine Probenfrequenz von 5.5 Minuten, während es mindestens 13 verschiedene C2- bis C5-NMHC maß. Die Drei-Sigma-Detektionsgrenze für n- und iso-Pentan wurde als 2.6 und 2.0 pptv ermittelt, in dieser Reihenfolge. Labortests bestätigten, dass FOTOS ein vielseitiges, robustes, hochautomatisiertes, präzises, genaues, empfindliches Instrument ist, geeignet für Echtzeitmessungen von VOC in Probenfrequenzen, die angemessen sind für ein Forschungsflugzeug wie HALO. Um die Leistung von FOTOS zu bestätigen, wurde vom 26. Januar bis 4. Februar 2010 ein Zwischenvergleich gemacht mit dem GC-FID-System am Meteorologischen Observatorium Hohenpeißenberg, einer WMO-GAW-globalen Station. Dreizehn verschiedene NMHC wurden innerhalb des Rahmens der GWA Data Quality Objectives (DQO) analysiert und verglichen. Mehr als 80% der Messungen von sechs C3- bis C5-NMHC erfüllten diese DQO. Diese erste Messkampagne im Feld hob die Robustheit und Messgenauigkeit von FOTOS hervor, zusätzlich zu dem Vorteil der höheren Probenfrequenz, sogar in einer Messung am Boden. Um die Möglichkeiten dieses Instrumentes im Feld zu zeigen, maß FOTOS ausgewählte leichte NMHC während einer Messkampagne im Borealen Waldgebiet, HUMPPA-COPEC 2010. Vom 12. Juli bis zum 12. August 2010 beteiligte sich eine internationale Gruppe von Instituten und Instrumenten an Messungen physikalischer und chemischer Größen der Gas- und Partikelphasen der Luft über dem Borealen Wald an der SMEAR II-Station nahe Hyyttiälä, Finnland. Es wurden mehrere Hauptpunkte von Interesse im Mischungsverhältnis der Alkane und im Isomerenverhätnis von Pentan identifiziert, insbesondere sehr unterschiedliche Perioden niedriger und hoher Variabilität, drei Rauchschwaden von Biomassen-Verbrennung von russischen Waldbränden und zwei Tage mit extrem sauberer Luft aus der Polarregion. Vergleiche der NMHC mit anderen anthropogenen Indikatoren zeigten mehrere Quellen anthropogener Einflüsse am Ort auf und erlaubten eine Unterscheidung zwischen lokalen und weiter entfernten Quellen. Auf einen minimalen natürlichen Beitrag zum 24h-Kreislauf von NOx wurde geschlussfolgert aus der Korrelation von NOx mit Alkanen. Altersschätzungen der Luftmassen durch das Isomerenverhältnis von Pentan wurden erschwert durch sich verändernde Verhältnisse der Quellen und durch Besonderheiten der Photochemie während des Sommers im hohen Norden. Diese Messungen zeigten den Wert des Messens leichter NMHC, selbst in abgelegenen Regionen, als einen zusätzlichen spezifischen Marker von anthropogenem Einfluss.

Geomorphological and statistical analysis on Ravenna dune fields changes, based on Terrestrial Laser Technology

Coastal sand dunes represent a richness first of all in terms of defense from the sea storms waves and the saltwater ingression; moreover these morphological elements constitute an unique ecosystem of transition between the sea and the land environment. The research about dune system is a strong part of the coastal sciences, since the last century. Nowadays this branch have assumed even more importance for two reasons: on one side the born of brand new technologies, especially related to the Remote Sensing, have increased the researcher possibilities; on the other side the intense urbanization of these days have strongly limited the dune possibilities of development and fragmented what was remaining from the last century. This is particularly true in the Ravenna area, where the industrialization united to the touristic economy and an intense subsidence, have left only few dune ridges residual still active. In this work three different foredune ridges, along the Ravenna coast, have been studied with Laser Scanner technology. This research didn’t limit to analyze volume or spatial difference, but try also to find new ways and new features to monitor this environment. Moreover the author planned a series of test to validate data from Terrestrial Laser Scanner (TLS), with the additional aim of finalize a methodology to test 3D survey accuracy. Data acquired by TLS were then applied on one hand to test some brand new applications, such as Digital Shore Line Analysis System (DSAS) and Computational Fluid Dynamics (CFD), to prove their efficacy in this field; on the other hand the author used TLS data to find any correlation with meteorological indexes (Forcing Factors), linked to sea and wind (Fryberger's method) applying statistical tools, such as the Principal Component Analysis (PCA).

Dynamics of Shifting Cultivation Landscapes in Northern Lao PDR Between 2000 and 2009 Based on an Analysis of MODIS Time Series and Landsat Images

The rotational nature of shifting cultivation poses several challenges to its detection by remote sensing. Consequently, there is a lack of spatial data on the dynamics of shifting cultivation landscapes on a regional, i.e. sub-national, or national level. We present an approach based on a time series of Landsat and MODIS data and landscape metrics to delineate the dynamics of shifting cultivation landscapes. Our results reveal that shifting cultivation is a land use system still widely and dynamically utilized in northern Laos. While there is an overall reduction in the areas dominated by shifting cultivation, some regions also show an expansion. A review of relevant reports and articles indicates that policies tend to lead to a reduction while market forces can result in both expansion and reduction. For a better understanding of the different factors affecting shifting cultivation landscapes in Laos, further research should focus on spatially explicit analyses.

A STUDY OF SO2 EMISSIONS AND GROUND SURFACE DISPLACEMENTS AT LASTARRIA VOLCANO, ANTOFAGASTA REGION, NORTHERN CHILE

Lastarria volcano (Chile) is located at the North-West margin of the `Lazufre' ground inflation signal (37x45 km²), constantly uplifting at a rate of ~2.5 cm/year since 1996 (Pritchard and Simons 2002; Froger et al. 2007). The Lastarria volcano has the double interest to be superimposed on a second, smaller-scale inflation signal and to be the only degassing area of the Lazufre signal. In this project, we compared daily SO2 burdens recorded by AURA's OMI mission for 2005-2010 with Ground Surface Displacements (GSD) calculated from the Advanced Synthetic Aperture Radar (ASAR) images for 2003-2010. We found a constant maximum displacement rate of 2.44 cm/year for the period 2003-2007 and 0.80- 0.95 cm/year for the period 2007-2010. Total SO2 emitted is 67.0 kT for the period 2005-2010, but detection of weak SO2 degassing signals in the Andes remains challenging owing to increased noise in the South Atlantic radiation Anomaly region.