ALBERT

Description: This paper presents an approach to detecting patterns in a three-dimensional context, emphasizing the role played by the local geometry of the surface model. The core of the associated algorithm is represented by the cosine similarity computed to sub-matrices of regularly gridded digital surface/canopy models. We developed an accompanying software instrument compatible with a GIS environment which allows, as inputs, locations in the surface/canopy model based on field data, pre-defined geometric shapes, or their combination. We exemplified the approach for a study case dealing with the locations of scattered trees and shrubs previously identified in the field in two study sites. We found that the variation in the pairwise similarities between the trees is better explained by the computation of slopes. Furthermore, we considered a pre-defined shape, the Mexican Hat wavelet. Its geometry is controlled by a single number, for which we found ranges of best fit between the shapes and the actual trees. Finally, a suitable combination of parameters made it possible to determine the potential locations of scattered trees. The accuracy of detection was equal to 77.9% and 89.5% in the two study sites considered. Moreover, a visual check based on orthophotomaps confirmed the reliability of the outcomes.

Description: Global changes impact the human-environment relationship, and, in particular, they affect the provision of ecosystem services. Mountain ecosystems provide a wide range of such services, but they are highly sensitive and vulnerable to change due to various human pressures and natural processes. We conducted a literature survey that focused on two main issues. The first was the identification of quantitative methods aimed at assessing the impact of land use changes in mountain regions and the related ecosystem services. The second was the analysis of the extent to which the outcomes of these assessments are useful and transferable to stakeholders. We selected papers through a keyword-driven search of the ISI Web of Knowledge and other international databases. The keywords used for the search were mountain land use change and ecosystem service. Quantitative approaches to ecosystem service assessment rely on suitable indicators, therefore land use/land cover can be used as an appropriate proxy. Landscape metrics are a powerful analytical tool; their use can increase the accuracy of assessments and facilitate the mitigation of specific phenomena, such as fragmentation or the reduction of core habitat areas. Mapping is essential: it is the basis for spatial analyzes and eases the interactions between stakeholders. Land use/land cover change is a temporal process, so both past and future approaches are meaningful. It is necessary to enhance information transfer from theory to practice. Increasing stakeholder awareness can lead to suitable management solutions, and, reciprocally, stakeholder feedback can help improve current assessment methodologies and contribute to developing new tools that are suitable for specific problems.

Description: Traditional methods for individual tree-crown (ITC) detection (image classification, segmentation, template matching, etc.) applied to very high-resolution remote sensing imagery have been shown to struggle in disparate landscape types or image resolutions due to scale problems and information complexity. Deep learning promised to overcome these shortcomings due to its superior performance and versatility, proven with reported detection rates of ~90%. However, such models still find their limits in transferability across study areas, because of different tree conditions (e.g., isolated trees vs. compact forests) and/or resolutions of the input data. This study introduces a highly replicable deep learning ensemble design for ITC detection and species classification based on the established single shot detector (SSD) model. The ensemble model design is based on varying the input data for the SSD models, coupled with a voting strategy for the output predictions. Very high-resolution unmanned aerial vehicles (UAV), aerial remote sensing imagery and elevation data are used in different combinations to test the performance of the ensemble models in three study sites with highly contrasting spatial patterns. The results show that ensemble models perform better than any single SSD model, regardless of the local tree conditions or image resolution. The detection performance and the accuracy rates improved by 3–18% with only as few as two participant single models, regardless of the study site. However, when more than two models were included, the performance of the ensemble models only improved slightly and even dropped.

Description: Human–wildlife interactions (HWI) were frequent in the post-socialist period in the mountain range of Central European countries where forest habitats suffered transitions into built-up areas. Such is the case of the Upper Prahova Valley from Romania. In our study, we hypothesized that the increasing number of HWI after 1990 could be a potential consequence of woodland loss. The goal of our study was to analyse the effects of landscape changes on HWI. The study consists of the next steps: (i) applying 450 questionnaires to local stakeholders (both citizens and tourists) in order to collect data regarding HWI temporal occurrences and potential triggering factors; (ii) investigating the relation between the two variables through the Canonical Correspondence Analysis (CCA); (iii) modelling the landscape spatial changes between 1990 and 2018 for identifying areas with forest loss; (iv) overlapping the distribution of both the households affected by HWI and areas with loss of forested ecosystems. The local stakeholders indicate that the problematic species are the brown bear (Ursus arctos), the wild boar (Sus scrofa), the red fox (Vulpes vulpes) and the grey wolf (Canis lupus). The number of animal–human interactions recorded an upward trend between 1990 and 2018, and the most significant driving factors were the regulation of hunting practices, the loss of habitats, and artificial feeding. The landscape change analysis reveals that between 1990 and 2018, the forest habitats were replaced by built-up areas primarily on the outskirts of settlements, these areas coinciding with frequent HWI. The results are valid for both forest ecosystems conservation in the region, wildlife management, and human infrastructures durable spatial planning.