Luigi Boschetti

Remote sensing of fire severity: assessing the performance of the normalized burn ratio

Several studies have used satellite data to map different levels of fire severity present within burned areas. Increasingly, fire severity has been estimated using a spectral index called the normalized burn ratio (NBR). This letter assesses the performance of the NBR against ideal requirements of a spectral index designed to measure fire severity. According to index theory, the NBR would be optimal for quantifying fire severity if the trajectory in spectral feature space caused by different levels of severity occurred perpendicular to the NBR isolines. We assess how well NBR meets this condition using reflectance data sensed before and shortly after fires in the South African savanna, Australian savanna, Russian Federation boreal forest, and South American tropical forest. Although previous studies report high correlation between fire severity measured in the field- and satellite-derived NBR, our results do not provide evidence that the performance of the NBR is optimal in describing fire severity shortly after fire occurrence. Spectral displacements due to burning occur in numerous directions relative to the NBR index isolines, suggesting that the NBR may not be primarily and consistently sensitive to fire severity. Findings suggest that the development of the next generation of methods to estimate fire severity remotely should incorporate knowledge of how fires of different severity displace the position of prefire vegetation in multispectral space.

Southern Africa Validation of the MODIS, L3JRC, and GlobCarbon Burned-Area Products

Three available global multi-annual burned area products (L3JRC, GlobCarbon, and MODIS) are validated for a burning season across southern Africa. Validation is undertaken using the same independent reference data and using the same validation and reporting protocol. The independent reference data were derived by interpreting multitemporal Landsat Enhanced Thematic Mapper Plus data to map the location and approximate date of burning at 11 Landsat scenes distributed across southern Africa and covering approximately 295 000 km2. The accuracy of the products was quantified using metrics derived from confusion matrices to characterize product accuracy for local applications and using metrics derived through a linear regression on a 5 times 5 km grid to characterize product accuracy for coarser scale applications. Quantitative results are described, and the differences between the products are discussed.

Southern African fire regimes as revealed by remote sensing

Here we integrate spatial information on annual burnt area, fire frequency, fire seasonality, fire radiative power and fire size distributions to produce an integrated picture of fire regimes in southern Africa. The regional patterns are related to gradients of environmental and human controls of fire, and compared with findings from other grass-fuelled fire systems on the globe. The fire regime differs across a gradient of human land use intensity, and can be explained by the differential effect of humans on ignition frequencies and fire spread. Contrary to findings in the savannas of Australia, there is no obvious increase in fire size or fire intensity from the early to the late fire season in southern Africa, presumably because patterns of fire ignition are very different. Similarly, the importance of very large fires in driving the total annual area burnt is not obvious in southern Africa. These results point to the substantial effect that human activities can have on fire in a system with high rural population densities and active fire management. Not all aspects of a fire regime are equally impacted by people: fire-return time and fire radiative power show less response to human activities than fire size and annual burned area.

The Global Impact of Clouds on the Production of MODIS Bidirectional Reflectance Model-Based Composites for Terrestrial Monitoring

A global data set of cloud occurrence probability derived from Moderate Resolution Imaging Spectroradiometer (MODIS) Terra and Aqua gridded daily data is analyzed to investigate the probability of obtaining at least a minimum number of cloud-free observations within various compositing periods. The probabilities derived from Terra and Aqua, with morning and afternoon overpass times, respectively, are similar and increase with compositing period. Compositing both Terra and Aqua observations results in considerably higher probabilities of obtaining a sufficient number of observations for bidirectional reflectance model-based compositing. Given that the only alternative to obtaining sufficient samples is to extend the observation period, which can cause significant problems when the surface state changes, it is concluded that using data from the two MODIS sensors provides the most effective way of generating composited products. Findings with respect to the availability of cloud-free composites when n-day composites are generated on a temporally overlapping daily rolling basis, i.e., every day, rather than every n-days, are also discussed for regional and global applications

A MODIS assessment of the summer 2007 extent burned in Greece

Devastating fires affected Greece in the summer 2007, with the loss of more than 60 human lives, the destruction of more than 100 villages and hundreds of square kilometres of forest burned. This Letter presents a map of the extent burned and the approximate day of burning in Greece mapped by the MODIS burned area product for 22 June to 30 August 2007 and the burned areas mapped independently by the European Forest Fires Information Service (EFFIS). The characteristics of the two datasets, and an evaluation of the areas burned comparing the MODIS and EFFIS data for the same temporal interval are described.

A sampling method for the retrospective validation of global burned area products

This work presents a design-based validation and calibration scheme for the Global Burned Area 2000 (GBA2000) products. The objective of such a scheme is to assess the margins of uncertainty associated with the burned area products and to estimate calibration coefficients needed to convert burned pixel counts into areal estimates. As the validation of GBA2000 was performed long after 2000, and given the fact that burned areas are a predominantly nonpermanent land cover change, the reference data are obtained from a set of Landsat-7 Enhanced Thematic Mapper Plus high-resolution remotely sensed data. A stratified sampling scheme is presented, specifically designed for the retrospective validation of burned area data; the scheme is based on combining information from two low-resolution burned area products (GBA2000 itself and Globscar). The resulting stratification has been applied to the whole global GBA2000 dataset, and preliminary validation results are reported for Africa. The conclusions highlight the limits of a retrospective validation exercise, and summarize some of the open issues in the validation of global burned area maps

Tree species mapping with Airborne hyper-spectral MIVIS data: the Ticino Park study case

The present work describes the procedure, which was studied for mapping the spatial distribution of tree forest communities in the Ticino Park located in Northern Italy. Ten overlapping airborne runs of the Multispectral Infrared Visible Imaging Spectrometer (MIVIS) were acquired to cover the entire park extent (920 km2). An integrated supervised classification procedure was developed using band ratios in the red edge portion (REP) of the spectrum and training collected by field survey and visual interpretation. Validation performed with a robust random stratified sampling scheme and taking into account the unequal distribution of the classes showed that, on large‐scale application, high‐resolution remotely sensed images can generate, in a cost‐effective manner, accurate (overall accuracy 75%) local‐scale thematic products.

Global assessment of the temporal reporting accuracy and precision of the MODIS burned area product

A method for the systematic evaluation of the temporal reporting accuracy and precision of burned area products conducted using active fire detections as the reference dataset is described. The method is applied globally to 6 years of Moderate Resolution Imaging Spectroradiometer (MODIS) burned area and active fire product data. The distribution of the time difference between active fire and burned area detections that occur within 90 days is analysed and summary statistics extracted globally. The median time difference in reporting between the MODIS burned area and the active fire product detections is 1 day and the majority of MODIS burned area product detections occur temporally close to an active fire detection: 50% within a single day and 75% within 4 days. Users of the MODIS burned area product with temporal reporting requirements should be aware of these findings if using the approximate day of burning information provided in the burned area product.

Operational Automatic Remote Sensing Image Understanding Systems: Beyond Geographic Object-Based and Object-Oriented Image Analysis (GEOBIA/GEOOIA). Part 1: Introduction

Abstract: According to existing literature and despite their commercial success, state-of-the-art two-stage non-iterative geographic object-based image analysis (GEOBIA) systems and three-stage iterative geographic object-oriented image analysis (GEOOIA) systems, where GEOOIA  GEOBIA, remain affected by a lack of productivity, general consensus and research. To outperform the degree of automation, accuracy, efficiency, robustness, scalability and timeliness of existing GEOBIA/GEOOIA systems in compliance with the Quality Assurance Framework for Earth Observation (QA4EO) guidelines, this methodological work is split into two parts. The present first paper provides a multi-disciplinary Strengths, Weaknesses, Opportunities and Threats (SWOT) analysis of the GEOBIA/GEOOIA approaches that augments similar analyses proposed in recent years. In line with constraints stemming from human vision, this SWOT analysis promotes a shift of learning paradigm in the pre-attentive vision first stage of a remote sensing (RS) image understanding system (RS-IUS), from sub-symbolic statistical model-based (inductive) image segmentation to symbolic physical model-based (deductive) image preliminary classification. Hence, a symbolic deductive pre-attentive vision first stage accomplishes image sub-symbolic segmentation and image symbolic pre-classification simultaneously. In the second part of this work a novel hybrid (combined deductive and inductive) RS-IUS architecture featuring a symbolic deductive pre-attentive vision first stage is proposed and discussed in terms of: (a) computational theory (system design); (b) information/knowledge representation; (c) algorithm design; and

Towards a new paradigm in fire severity research using dose–response experiments

Most landscape-scale fire severity research relies on correlations between field measures of fire effects and relatively simple spectral reflectance indices that are not direct measures of heat output or changes in plant physiology. Although many authors have highlighted limitations of this approach and called for improved assessments of severity, others have suggested that the operational utility of such a simple approach makes it acceptable. An alternative pathway to evaluate fire severity that bridges fire combustion dynamics and ecophysiology via dose–response experiments is presented. We provide an illustrative example from a controlled nursery combustion laboratory experiment. In this example, severity is defined through changes in the ability of the plant to assimilate carbon at the leaf level. We also explore changes in the Differenced Normalised Differenced Vegetation Index (dNDVI) and the Differenced Normalised Burn Ratio (dNBR) as intermediate spectral indices. We demonstrate the potential of this methodology and propose dose–response metrics for quantifying severity in terms of carbon cycle processes.