Karen E. Joyce

A review of the status of satellite remote sensing and image processing techniques for mapping natural hazards and disasters

In the event of a natural disaster, remote sensing is a valuable source of spatial information and its utility has been proven on many occasions around the world. However, there are many different types of hazards experienced worldwide on an annual basis and their remote sensing solutions are equally varied. This paper addresses a number of data types and image processing techniques used to map and monitor earthquakes, faulting, volcanic activity, landslides, flooding, and wildfire, and the damages associated with each. Remote sensing is currently used operationally for some monitoring programs, though there are also difficulties associated with the rapid acquisition of data and provision of a robust product to emergency services as an end-user. The current status of remote sensing as a rapid-response data source is discussed, and some perspectives given on emerging airborne and satellite technologies.

Mangrove Species Identification: Comparing WorldView-2 with Aerial Photographs

Remote sensing plays a critical role in mapping and monitoring mangroves. Aerial photographs and visual image interpretation techniques have historically been known to be the most common approach for mapping mangroves and species discrimination. However, with the availability of increased spectral resolution satellite imagery, and advances in digital image classification algorithms, there is now a potential to digitally classify mangroves to the species level. This study compares the accuracy of mangrove species maps derived from two different layer combinations of WorldView-2 images with those generated using high resolution aerial photographs captured by an UltraCamD camera over Rapid Creek coastal mangrove forest, Darwin, Australia. Mangrove and non-mangrove areas were discriminated using object-based image classification. Mangrove areas were then further classified into species using a support vector machine algorithm with best-fit parameters. Overall classification accuracy for the WorldView-2 data within the visible range was 89%. Kappa statistics provided a strong correlation between the classification and validation data. In contrast to this accuracy, the error matrix for the automated classification of aerial photographs indicated less promising results. In summary, it can be concluded that mangrove species mapping using a support vector machine algorithm is more successful with WorldView-2 data than with aerial photographs.

Bi-directional reflectance of corals

An assessment of the bi-directional reflectance distribution function (BRDF) of corals with different morphologies was undertaken using hyperspectral reflectance measurements. The highest variance in reflectance obtained at different viewing angles was found for the open branching Acropora colony, possibly attributed to branch orientation and internal shadow distribution. Spectral separation within and between features at the nominated sensor-viewing angles was greatest in the near infrared portion of the spectrum. The analysis of coral reef bi-directional reflectance properties and degree of internal shadowing holds potential for future assessment and information extraction relating to coral structural characteristics.

Public Sector Priority Setting Using Decision Support Tools

The public sector is in transition in many countries, including New Zealand, and many government departments have been restructured so as to increase transparency and accountability in their decision-making by implementing fundamental changes to governance, accountability regimes and information systems. New frameworks are also needed for tasks such as priority setting, and this paper describes the framework developed to assist the then NZ Ministry of Agriculture (MAF) in setting priorities for the allocation of research funds. This comprised a participative process including a series of workshops using both GroupSystems software for electronic meeting support and Visual Interactive Sensitivity Analysis (VISA) software to facilitate a multi-criteria decision analysis.

Mapping and monitoring geological hazards using optical, LiDAR, and synthetic aperture RADAR image data

Geological hazards and their effects are often geographically widespread. Consequently, their effective mapping and monitoring is best conducted using satellite and airborne imaging platforms to obtain broad scale, synoptic coverage. With a multitude of hazards and effects, potential data types, and processing techniques, it can be challenging to determine the best approach for mapping and monitoring. It is therefore critical to understand the spatial and temporal effects of any particular hazard on the environment before selecting the most appropriate data type/s and processing techniques to apply. This review is designed to assist the decision-making and selection process when embarking on a hazard mapping or monitoring exercise. It focuses on the application of optical, LiDAR, and synthetic aperture RADAR technologies for the assessment of pre-event risk and post-event damage. Geological hazards of global interest summarized here are landslides and erosion; seismic and tectonic hazards; ground subsidence; and flooding and tsunami.

Live Coral Cover Index Testing and Application with Hyperspectral Airborne Image Data

Coral reefs are complex, heterogeneous environments where it is common for the features of interest to be smaller than the spatial dimensions of imaging sensors. While the coverage of live coral at any point in time is a critical environmental management issue, image pixels may represent mixed proportions of coverage. In order to address this, we describe the development, application, and testing of a spectral index for mapping live coral cover using CASI-2 airborne hyperspectral high spatial resolution imagery of Heron Reef, Australia. Field surveys were conducted in areas of varying depth to quantify live coral cover. Image statistics were extracted from co-registered imagery in the form of reflectance, derivatives, and band ratios. Each of the spectral transforms was assessed for their correlation with live coral cover, determining that the second derivative around 564 nm was the most sensitive to live coral cover variations(r2 = 0.63). Extensive field survey was used to transform relative to absolute coral cover, which was then applied to produce a live coral cover map of Heron Reef. We present the live coral cover index as a simple and viable means to estimate the amount of live coral over potentially thousands of km2 and in clear-water reefs.

How hot? How often? Getting the fire frequency and timing right for optimal management of woody cover and pasture composition in northern Australian grazed tropical savannas. Kidman Springs Fire Experiment 1993-2013

A long-term (1993–2013) experiment in grazed semiarid tropical savannas in northern Australia tested the impact of varying the frequency (every 2, 4 and 6 years) and season (June – EDS versus October – LDS) of fire compared with unburnt controls on woody cover and pasture composition, in grassland and open woodland. Over an 18-year period, woody cover increased by 4% (absolute) in the woodland even with the most severe (i.e. frequent, late dry season) fire treatments. With less severe or no fire, woody cover increased by 12–17%. In the grassland, woody cover remained static when subjected to LDS fires every 2 or 4 years, but increased by 3–6% under other fire treatments, and by 8% when unburnt. Major shifts in understorey species composition occurred at both sites regardless of fire regime. The effect of fire on herbage mass and composition was compounded by higher grazing after fires. The herbage mass of perennial grasses declined and that of annual grasses and forbs increased following early or frequent fires. Brachyachne convergens, Gomphrena canescens and Flemingia pauciflora increased in response to fire while Aristida latifolia and Heteropogon contortus decreased. Four-yearly LDS fire provided the most effective management of woody cover and pasture composition. Although EDS fire is recommended for biodiversity management and reducing greenhouse gas emissions in wet tropical savannas, on grazed pastoral land, it can promote woodland thickening and pasture degradation. Optimal fire management, therefore, depends on vegetation type, land use and the prevailing seasonal timing and frequency of fire.

Mapping Coral Reef Benthos, Substrates, and Bathymetry, Using Compact Airborne Spectrographic Imager (CASI) Data

This study used a reef-up approach to map coral reef benthos, substrates and bathymetry, with high spatial resolution hyperspectral image data. It investigated a physics-based inversion method for mapping coral reef benthos and substrates using readily available software: Hydrolight and ENVI. Compact Airborne Spectrographic Imager (CASI) data were acquired over Heron Reef in July 2002. The spectral reflectance of coral reef benthos and substrate types were measured in-situ, and using the HydroLight 4.2 radiative transfer model a spectral reflectance library of subsurface reflectance was simulated using water column depths from 0.5–10.0 m at 0.5 m intervals. Using the Spectral Angle Mapper algorithm, sediment, benthic micro-algae, algal turf, crustose coralline algae, macro-algae, and live coral were mapped with an overall accuracy of 65% to a depth of around 8.0 m; in waters deeper than 8.0 m the match between the classified image and field validation data was poor. Qualitative validation of the maps showed accurate mapping of areas dominated by sediment, benthic micro-algae, algal turf, live coral, and macro-algae. A bathymetric map was produced for water column depths 0.5–10.0 m, at 0.5 m intervals, and showed high correspondence with in-situ sonar data (R2 value of 0.93).

Assessing Image Processing Techniques for Mapping Landslides

A combination of high relief, steep slopes, intense rainfall, active tectonics and volcanic activity provide the conditioning and triggering for frequent and widespread landsliding events in New Zealand. While many techniques are available to map the extent of landslides, remote sensing and image processing methods have yet to be comprehensively evaluated. This study shows that manual interpretation of high resolution imagery is the most accurate method (92%) although also the most time consuming. NDVI thresholding and supervised classification (spectral angle mapper) provide an adequate accuracy level (76%) with considerable time saving for large area analysis.

Mangrove Tree Crown Delineation from High-Resolution Imagery

Mangroves are very dense, spatially heterogeneous, and have limited height variations between neighboring trees. Delineating individual tree crowns is thus very challenging. This study compared methods for isolating mangrove crowns using object based image analysis. A combination of WorldView-2 imagery, a digital surface model, a local maximum filtering technique, and a region growing approach achieved 92 percent overall accuracy in extracting tree crowns. The more traditionally used inverse watershed segmentation method showed low accuracy (35 percent), demonstrating that this method is better suited to homogeneous forests with reasonable height variations between trees. The main challenges with each of the methods tested were the limited height variation between surrounding trees and multiple upward pointing branches of trees. In summary, mangrove tree crowns can be delineated from appropriately parameterized object-based algorithms with a combination of high-resolution satellite images and a digital surface model. We recommend partitioning the imagery into homogeneous species stands for best results.