Sotirios Koukoulas

Mapping individual tree location, height and species in broadleaved deciduous forest using airborne LIDAR and multi-spectral remotely sensed data.

Automated feature extraction based on prototypes is only partially successful when applied to remotely sensed imagery of natural scenes due to the complexity and unpredictability of the shape and geometry of natural features. Here, a new method is developed for extracting the locations of treetops by applying GIS (Geographical Information System) overlay techniques and morphological functions to high spatial resolution airborne imagery. This method is based on the geometrical and spatial properties of tree crowns. Airborne data of the study site in the New Forest, UK included colour aerial photographs, LIDAR (Light Detection And Ranging) and ATM (Airborne Thematic Mapper) imagery. A DEM (Digital Elevation Model) was generated from LIDAR data and then subtracted from the original LIDAR image to create a Canopy Height Model (CHM). A set of procedures using image contouring and the manipulation of the resulting polygons was implemented to extract treetops from the aerial photographs and the CHM. Criteria were developed and threshold values were set using a supervised approach for the acceptance or rejection of features based on field knowledge. Tree species were mapped by classifying the ATM data and these data were co‐registered with the treetop layer. For broadleaved deciduous plantations the success of treetop extraction using aerial photographs was 91%, but was much lower using LIDAR data. For semi‐natural forests, the LIDAR produced better results than the aerial photographs with a success of 80%, which was considered high, given the complexity of these uneven aged stands. The methodology presented here is easy to apply as it is implemented within a GIS and the final product is an accurate map with information about the location, height and species of each tree.

Quantifying the spatial properties of forest canopy gaps using LiDAR imagery and GIS

The spatial properties of gaps have an important influence upon the regeneration dynamics and species composition of forests. However, such properties can be difficult to quantify over large spatial areas using field measurements. This research considers how we conceptualize and define forest canopy gaps from a remote sensing point of view and highlights the inadequacies of passive optical remotely sensed data for delineating gaps. The study employs the analytical functions of a geographical information system to extract gap spatial characteristics from imagery acquired by an active remote sensing device, an airborne light detection and ranging instrument (LiDAR). These techniques were applied to an area of semi-natural broadleaved deciduous forest, in order to map gap size, shape complexity, vegetation height diversity and gap connectivity. A vegetation cover map derived from imagery from an airborne multispectral scanner was used in combination with the LiDAR data to characterize the dominant vegetation types within gaps. Although the quantification of these gap characteristics alone is insufficient to provide conclusive evidence on specific processes, the paper demonstrates how such information can be indicative of the general status of a forest and can provide new perspectives and possibilities or further ecological research and forest monitoring activities.

Dynamic simulation and visualisation of coastal erosion

A key requirement for effective coastal zone management is good knowledge and prediction of land erosion rates due to encroachment of the sea. However, in addition to demarcation of the hazard through modelling and mapping, a policy of risk mitigation necessitates significant attention should also be addressed to communicating the transient behaviour of the predictions and associated uncertainty. With climate change and sea level rise implying that historical rates of change may not be a reliable guide for the future, enhanced visualisation of the evolving coastline has the potential to improve awareness of this changing risk. This visual content is developed by linking scientific modelling with the transformation of digital elevation models, and then using GIS to integrate other spatiotemporal content. The resulting high-resolution visualisations may meet demands from decision-makers for tools to communicate scientific results more effectively, due to their realism and apparent authenticity. Nevertheless they can also produces a tension with the underlying scientific content because of the necessary extrapolation of extra detail, and the lack of established procedures to communicate the resulting uncertainty in the visualisation. Coastal managers also have concerns about releasing the visualisations to the general public. These issues are explored through analysis of future cliff erosion in Norfolk on the eastern coast of Great Britain.

Spatial relationships between tree species and gap characteristics in broad-leaved deciduous woodland

Abstract Questions: 1. What are the spatial patterns of all trees, individual tree species, trees within particular height classes, all gaps and gaps with specific properties across the study site in broad-leaved deciduous forest at a range of scales? 2. Are patterns of the above features spatially associated? 3. Are these patterns indicative of gap creation mechanisms and phases of regeneration? Location: Frame Wood, New Forest, UK. Methods: Ripleys K-function analysis was applied to spatial information derived from airborne remotely sensed imagery to characterize the patterns of trees and gaps and to test for spatial interactions between these patterns. The patterns of trees and gaps with specific physical and spatial properties were analysed. Results: The pattern of all tree species combined was random for most scales; Quercus robur followed the same random pattern, while Fagus sylvatica and Betula pendula were clustered over most spatial scales. Large gaps (> 250 m2) and larger trees (> 17.5 m) were randomly distributed, while smaller gaps and smaller trees were clustered. Significant spatial relationships were demonstrated between the patterns of different tree species and between trees within different size classes, as well as between the patterns of trees and gaps with specific properties. Conclusions: Small gap patterns and field evidence indicated that progressive gap enlargement is the most likely creation mechanism for large gaps (> 250 m2). Clustered patterns of younger individuals were indicative of patches of past regeneration. As a complement to field-based data, data derived from remotely sensed imagery provides spatially comprehensive information with which to further investigate woodland stand/community processes and gap dynamics. Abbreviations: ATM = Airborne thematic mapper; CHM = Canopy height model; CSR = Complete spatial randomness; GHD = Gap height diversity; GSCI = Gap shape complexity index; LiDAR = Light detection and ranging.

Evaluating Post-Fire Forest Resilience Using GIS and Multi-Criteria Analysis: An Example from Cape Sounion National Park, Greece

Forest fires are one of the major causes of ecological disturbance in the mediterranean climate ecosystems of the world. Despite the fact that a lot of resources have been invested in fire prevention and suppression, the number of fires occurring in the Mediterranean Basin in the recent decades has continued to markedly increase. The understanding of the relationship between landscape and fire lies, among others, in the identification of the system’s post-fire resilience. In our study, ecological and landscape data are integrated with decision-support techniques in a Geographic Information Systems (GIS) framework to evaluate the risk of losing post-fire resilience in Pinus halepensis forests, using Cape Sounion National Park, Central Greece, as a pilot case. The multi-criteria decision support approach has been used to synthesize both bio-indicators (woody cover, pine density, legume cover and relative species richness and annual colonizers) and geo-indicators (fire history, parent material, and slope inclination) in order to rank the landscape components. Judgments related to the significance of each factor were incorporated within the weights coefficients and then integrated into the multicriteria rule to map the risk index. Sensitivity analysis was very critical for assessing the contribution of each factor and the sensitivity to subjective weight judgments to the final output. The results of this study include a final ranking map of the risk of losing resilience, which is very useful in identifying the “risk hotspots”, where post-fire management measures should be applied in priority.

Multi-temporal land-cover classification and change analysis with conditional probability networks: The case of Lesvos Island (Greece)

This study uses a series of Landsat images to map the main land-cover types on the Mediterranean island of Lesvos, Greece. We compare a single-year maximum likelihood classification (MLC) with a multi-temporal maximum likelihood classification (MTMLC) approach, with time-series class labels modelled using a first-order hidden Markov model comprising continuous and discrete variables. A rigorous validation scheme shows statistically significant higher accuracy figures for the multi-temporal approach. Land-cover change accuracies were also greatly improved by the proposed methodology: from 46% to 70%. The results show that when only two dates are used, the mapping of land use/cover is unreliable and a large number of the changes identified are due to the individual-year commission and omission errors.

Quantifying spatio-temporal patterns of forest fragmentation in Hymettus Mountain, Greece

The rapid land use/cover change (LUCC) and landscape fragmentation occurring around the world is largely attributed to human induced factors. Landscape fragmentation has become a central issue in landscape ecology and conservation policies due to its direct influence on biodiversity which consequently endangers the sustainability of ecological goods and ecosystem services. Thus, fragmentation monitoring and assessment is a critical issue in land use planning and sustainable environmental management in order to avoid any irreversible negative consequences. This research explores the application of methodologies that employ multi-temporal satellite imagery, combined with geographical information systems and landscape metrics, to assess forest fragmentation. The objective is to determine spatio-temporally the LUCCs focusing on the woody vegetation in Hymettus Mountain of Greece over the last decades. The study area, which has been designated as a Natura 2000 site, is situated near the city of Athens. It faces various perturbations triggered by socio-economic factors and the absence of an ongoing contextual appraisal for conservation. To quantify the LUCCs, nine Landsat images spanning 28 years are classified. Post classification comparison is applied to generate transition maps. Additionally, eight landscape metrics are calculated. The change detection results identify hot-spots of forest fragmentation where mitigation measures should be taken, so that further irreversible alteration of the ecosystem is prevented. The landscape metrics advocate that, during the last three decades, the woody vegetation have steadily been more fragmented. The primary direct causes are economic driven intense anthropogenic activities along with frequent wildland fires whereas the indirect cause is the absence of a sustainable environmental management and conservation strategy.

Social capital and citizen perceptions of coastal management for tackling climate change impacts in Greece

Abstract There is a growing consensus among researchers that social aspects and the involvement of local communities play a critical role in public decision-making processes in the coastal zone. Social capital is a parameter which has recently gained significant attention in this context. It is regarded that it has a significant influence on the adaptation capacity of local communities to climate change impacts. The present paper aims to contribute to this field through an examination of citizens’ perceptions of three coastal zone management policies (hold the line, managed realignment and no active intervention) along with the influence of social capital on the level of social acceptability for these proposed policy options. For this purpose, a quantitative empirical study was conducted for the first time in five coastal areas of Greece that are regarded as high flood-risk areas due to sea-level rise. Respondents demonstrated that they are willing to accept changes in their social and natural environments in order to confront sea-level rise and are more positive towards the managed realignment option, as long as this is accompanied by financial compensation for those whose properties will be affected. Regarding the influence of social capital, through the results of an ordinal regression, it was observed that institutional and social trust influence positively citizens’ level of agreement for the managed realignment policy. Furthermore, respondents who believe that a sense of reciprocity exists in their community are also more willing to accept active intervention policies.

A GIS TOOL FOR ANALYSIS AND INTERPRETATION OF COASTAL EROSION MODEL OUTPUTS (SCAPEGIS)

The SCAPE (Soft Cliff and Platform Erosion) model of cliff toe retreat, and a cliff-top recession model, have been linked with a new flexible GIS tool (SCAPEGIS) to provide visualisation and analytical capability for the model results. 45 model runs exploring different sealevel rise and wave climate scenarios and protection choices are available. Outputs are available in the form of maps, dynamic visualisation, and descriptive statistics of key parameters such as cliff toe and cliff top position. It also allows analysis with other datasets such as land use and building location for impact evaluation, and hence supports shoreline management and cliff-top land use planning. Some preliminary results and ideas for further development are presented.

Land cover of Greece, 2010: a semi-automated classification using random forests

ABSTRACT Information about land cover (LC) and land use is fundamental in various areas of research regarding the Earths surface. However, field campaigns are costly and time consuming while existing data sets have strong limitations. Classification of LC by remote sensing, although considered a technically and methodologically challenging task, can facilitate mapping initiatives at various scales. This study suggests an efficient and robust methodology of LC classification with minimal user requirements. The study site is Greece which faces a lack of up to date LC maps at national scale. In this context we employed Landsat imagery, open source software and the random forest classification algorithm to produce a high resolution national LC map for 2010. The algorithm was trained semi-automatically, extracting information from available data sets. The results are promising, achieving an overall accuracy of 83%. The methodology presented minimizes many obstacles that lead to data deficiencies and can act as a baseline for future LC mapping initiatives.