Google Street View as an alternative method to car surveys in large-scale vegetation assessments

An evaluation of different approaches which use Google Street View imagery to ground truth land degradation assessments

Member states of the United Nations Convention to Combat Desertification are required to report on the proportion of land that is degraded in their countries, a requirement that is also tied into the UN Sustainable Development Goals (SDGs). National land degradation assessments are often conducted with the use of remote sensing data which are not always ground truthed. Google Street View (GSV) provides high resolution, panoramic imagery across large parts of the world that has the potential to be used to ground truth land degradation assessments. We apply three different methodologies (visual interpretation of GSV images, GSV image classification and vegetation index extraction) to derive vegetation cover estimates from Google Street View imagery for the Hardeveld bioregion of the Succulent Karoo biome in South Africa. Visual estimates of cover best predict known habitat condition values (adjusted R² = 0.86), whilst estimates derived from an unsupervised classification of GSV images also predict habitat condition relatively well (adjusted R² = 0.52). These results show the potential for using GSV imagery, and other large collections of ground-level landscape photographs, as a rough ground-truthing tool, especially in instances where more traditional ground-truthing approaches are not possible.

Eucalypt Recruitment and Invasion Potential in Protected Areas of the Iberian Peninsula under Current and Future Climate Conditions

Eucalyptus globulus Labill. stands have been expanding in protected areas (sites) of the Natura 2000 network in the Iberian Peninsula (Iberia). This expansion is mostly human-driven, but there is increasing evidence of plant recruitment and escape from cultivation areas. Therefore, it is important to assess the recruitment and invasion potential of sites and associated habitats and how future climate may change this potential. Here, we use SDMs to project current and future climatic suitability for E. globulus recruitment in Iberia and combine this suitability with local factors to rate the current recruitment potential of eucalypt stands. This potential is then extrapolated to neighbour areas in Natura 2000 sites to assess the invasion potential. The results show a wide recruitment range along coastal regions of western and northern Iberia (83,275 km2) and a northward contraction under climate change, similar to the trend projected for plantation suitability. Recruitment potential of any level was identified in 989 km2, while invasion potential was identified in 878 km2 across 176 Natura 2000 sites. Heathlands and riparian forests were associated with the largest recruitment and invasion potential areas. This study may help in preventing further negative impacts in protected areas and habitats already affected by E. globulus expansion.

From Forest Dynamics to Wetland Siltation in Mountainous Landscapes: A RS-Based Framework for Enhancing Erosion Control

Human activities have caused a significant change in the function and services that ecosystems have provided to society since historical times. In mountainous landscapes, the regulation of services such as water quality or erosion control has been impacted by land use and land cover (LULC) changes, especially the loss and fragmentation of forest patches. In this work, we develop a Remote Sensing (RS)-based modelling approach to identify areas for the implementation of nature-based solutions (NBS) (i.e., natural forest conservation and restoration) that allow reducing the vulnerability of aquatic ecosystems to siltation in mountainous regions. We used time series Landsat 5TM, 7ETM+, 8OLI and Sentinel 2A/2B MSI (S2) imagery to map forest dynamics and wetland distribution in Picos de Europa National Park (Cantabrian Mountains, northern Spain). We fed RS-based models with detailed in situ information based on photo-interpretation and fieldwork completed from 2017 to 2021. We estimated a forest cover increase rate of 2 ha/year comparing current and past LULC maps against external validation data. We applied this forest gain to a scenario generator model to derive a 30-year future LULC map that defines the potential forest extent for the study area in 2049. We then modelled the distribution of wetlands to identify the areas with the greatest potential for moisture accumulation. We used an S2 mosaic and topography-derived data such as the slope and topographic wetness index (TWI), which indicate terrain water accumulation. Overall accuracy scores reached values of 86% for LULC classification and 61% for wetland mapping. At the same time, we obtained the potential erosion using the NetMap software to identify potential sediment production, transport and deposition areas. Finally, forest dynamics, wetland distribution and potential erosion were combined in a multi-criteria analysis aiming to reduce the amount of sediment reaching selected wetlands. We achieved this by identifying the most suitable locations for the conservation and restoration of natural forests on slopes and in riparian areas, which may reduce the risk of soil erosion and maximise sediment filtering, respectively. The results show a network pattern for forest management that would allow for controlling erosion effects across space and time at three levels: one, by reducing the load that originates upslope in the absence of forest cover; two, by intersecting runoff at watercourses related to sediment transport; and three, by a lack of former barriers, by trapping erosion near to the receiving wetland systems, main river axes and contributing streams. In conclusion, the proposed methodology, which could be transferred to other mountain regions, allows to optimise investment for erosion prevention and wetland conservation by using only very specific areas of the landscape for habitat management (e.g., for NBS implementation).

Natura 2000 Areas, Road, Railway, Water, and Ecological Networks May Provide Pathways for Biological Invasion: A Country Scale Analysis

Invasive species are a major threat to biodiversity worldwide. Controlling their rapid spread can only be effective if we consider the geographical factors that influence their occurrence. For instance, roads, railway networks, green and blue infrastructure, and elements of ecological networks (e.g., ecological corridors) can facilitate the spread of invasive species. In our study, we mapped the occurrence of five invasive plant taxa (tree of heaven, common milkweed, Russian olive, black locust, and goldenrods) in Hungary, using field photos from the EUROSTAT Land Use and Coverage Area Frame Survey (LUCAS) database from the year 2015. Species point occurrence data were compared with the spatial characteristics of linear transport infrastructure and with the green and blue infrastructure. We found that the occurrence of tree of heaven and Russian olive was strongly related to the road and railway network. The average Euclidean distance of LUCAS points infected with these species from railway embankments and roads was much smaller than that of uninfected points. However, black locust and goldenrods were more common only along the road network. According to our results, the occurrence of some investigated invasive plants was over-represented in the HEN and within Natura 2000 areas of Hungary compared to non-infected points. Our results may provide important information for predicting the rate of invasion and for applying targeted management within the HEN, and Natura 2000 protected areas.

Comparison of different methods to assess the distribution of alien plants along the road network and use of Google Street View panoramas interpretation in Sicily (Italy) as a case study

The survey by foot in the field is compared to the survey from a car, the photo-interpretation of Google Street View (GSV) panoramas continuously and at intervals of 1.5 km and the photo-interpretation of Google Earth aerial images on a 10 km stretch of road in Sicily. The survey by foot was used as reference for the other methods. The interpretation of continuous GSV panoramas gave similar results as the assessment by car in terms of the number of species identified and their location, but with lower cost. The interpretation online of aerial photos allowed the identification of a limited number of taxa, but gave a good localisation for them. Interpretation of GSV panoramas, each of 1.5 km, allowed the recognition of twice as many taxa as the interpretation of aerial photos and taking half the time, but did not allow a complete localisation.

None of these methods alone seems sufficient to carry out a complete survey. A mixture of different techniques, which may vary according to the available resources and the goal to be achieved, seems to be the best compromise.

To further test the capabilities of the survey using the interpretation of GSV panoramas every 1.5 km along the roads, we proceeded to study the alien plants along 3500 km of the road network on the island of Sicily. This survey identified only 10% of the known species for the region, but allowed us to trace the distribution of invasive species whose distribution is currently poorly recorded.

Identification of Crop Type in Crowdsourced Road View Photos with Deep Convolutional Neural Network

In situ ground truth data are an important requirement for producing accurate cropland type map, and this is precisely what is lacking at vast scales. Although volunteered geographic information (VGI) has been proven as a possible solution for in situ data acquisition, processing and extracting valuable information from millions of pictures remains challenging. This paper targets the detection of specific crop types from crowdsourced road view photos. A first large, public, multiclass road view crop photo dataset named iCrop was established for the development of crop type detection with deep learning. Five state-of-the-art deep convolutional neural networks including InceptionV4, DenseNet121, ResNet50, MobileNetV2, and ShuffleNetV2 were employed to compare the baseline performance. ResNet50 outperformed the others according to the overall accuracy (87.9%), and ShuffleNetV2 outperformed the others according to the efficiency (13 FPS). The decision fusion schemes major voting was used to further improve crop identification accuracy. The results clearly demonstrate the superior accuracy of the proposed decision fusion over the other non-fusion-based methods in crop type detection of imbalanced road view photos dataset. The voting method achieved higher mean accuracy (90.6–91.1%) and can be leveraged to classify crop type in crowdsourced road view photos.

Ailanthus altissima Forests Determine a Shift in Herbaceous Layer Richness: A Paired Comparison with Hardwood Native Forests in Sub-Mediterranean Europe

Ailanthus altissima is an invasive alien species (IAS) present throughout Europe and included in the list of alien species of Union concern. In sub-Mediterranean areas of central Italy, there is a lack of knowledge about this invasive species and its interactions with the native forest ecosystems. We aim to find what are the main differences in vegetation structure and floristic diversity between A. altissima forests and native forests through the assessment of the principal ecological parameters that differ between the forest types. We performed 38 phytosociological relevés and sampling of ecological parameters in A. altissima forest communities and neighboring native forests. We analyzed how species richness, diversity, life forms, life strategies, structural characteristics, and ecological parameters changed in A. altissima forests compared with native ones. We found that in A. altissima forests, there is a shift in herbaceous layer richness, with a higher presence of annual ruderal herbs and the absence of herbaceous species linked to the forest environment. The ecological parameters that diverge from the native forests were total nitrogen, total carbon, and C/N ratio. A. altissima forest communities could threaten the biodiversity of the native forest ecosystems in the sub-Mediterranean landscape, favoring ruderal species and inhibiting the presence of typical forest species.

Global guidelines for the sustainable use of non-native trees to prevent tree invasions and mitigate their negative impacts

Sustainably managed non-native trees deliver economic and societal benefits with limited risk of spread to adjoining areas. However, some plantations have launched invasions that cause substantial damage to biodiversity and ecosystem services, while others pose substantial threats of causing such impacts. The challenge is to maximise the benefits of non-native trees, while minimising negative impacts and preserving future benefits and options.

A workshop was held in 2019 to develop global guidelines for the sustainable use of non-native trees, using the Council of Europe – Bern Convention Code of Conduct on Invasive Alien Trees as a starting point.

The global guidelines consist of eight recommendations: 1) Use native trees, or non-invasive non-native trees, in preference to invasive non-native trees; 2) Be aware of and comply with international, national, and regional regulations concerning non-native trees; 3) Be aware of the risk of invasion and consider global change trends; 4) Design and adopt tailored practices for plantation site selection and silvicultural management; 5) Promote and implement early detection and rapid response programmes; 6) Design and adopt tailored practices for invasive non-native tree control, habitat restoration, and for dealing with highly modified ecosystems; 7) Engage with stakeholders on the risks posed by invasive non-native trees, the impacts caused, and the options for management; and 8) Develop and support global networks, collaborative research, and information sharing on native and non-native trees.

The global guidelines are a first step towards building global consensus on the precautions that should be taken when introducing and planting non-native trees. They are voluntary and are intended to complement statutory requirements under international and national legislation. The application of the global guidelines and the achievement of their goals will help to conserve forest biodiversity, ensure sustainable forestry, and contribute to the achievement of several Sustainable Development Goals of the United Nations linked with forest biodiversity.

Robinia pseudoacacia-dominated vegetation types of Southern Europe: Species composition, history, distribution and management

Knowledge of the species composition of invaded vegetation helps to evaluate an ecological impact of aliens and design an optimal management strategy. We link a new vegetation analysis of a large dataset to the invasion history, ecology and management of Robinia pseudoacacia stands across Southern Europe and provide a map illustrating Robinia distribution. Finally, we compare detected relationships with Central Europe. We show that regional differences in Robinia invasion, distribution, habitats and management are driven both by local natural conditions (climate and soil properties, low competitive ability with native trees) and socioeconomic factors (traditional land-use). Based on the classification of 467 phytosociological relevés we distinguished five broad vegetation types reflecting an oceanity-continentality gradient. The stands were heterogeneous and included 824 taxa, with only 5.8% occurring in more than 10% of samples, representing mainly hemerobic generalists of mesophilous, nutrient-rich and semi-shady habitats. The most common were dry ruderal stands invading human-made habitats. Among native communities, disturbed mesic and alluvial forests were often invaded throughout the area, while dry forests and scrub dominated in Balkan countries. Continuous, long-term and large-scale cultivation represent a crucial factor driving Robinia invasions in natural habitats. Its invasion should be mitigated by suitable management taking into account adjacent habitats and changing cultivation practices to select for native species. Robinia invasion has a comparable pattern in Central and Southern Europe, but there is a substantial difference in management and utilization causing heterogeneity of many South-European stands.

Using street imagery and crowdsourcing internet marketplaces to measure motorcycle helmet use in Bangkok, Thailand

INTRODUCTION

The majority of Thailand's road traffic deaths occur on motorised two-wheeled or three-wheeled vehicles. Accurately measuring helmet use is important for the evaluation of new legislation and enforcement. Current methods for estimating helmet use involve roadside observation or surveillance of police and hospital records, both of which are time-consuming and costly. Our objective was to develop a novel method of estimating motorcycle helmet use.

METHODS

Using Google Maps, 3000 intersections in Bangkok were selected at random. At each intersection, hyperlinks of four images 90° apart were extracted. These 12 000 images were processed in Amazon Mechanical Turk using crowdsourcing to identify images containing motorcycles. The remaining images were sorted manually to determine helmet use.

RESULTS

After processing, 462 unique motorcycle drivers were analysed. The overall helmet wearing rate was 66.7 % (95% CI 62.6 % to 71.0 %). Taxi drivers had higher helmet use, 88.4% (95% CI 78.4% to 94.9%), compared with non-taxi drivers, 62.8% (95% CI 57.9% to 67.6%). Helmet use on non-residential roads, 85.2% (95% CI 78.1 % to 90.7%), was higher compared with residential roads, 58.5% (95% CI 52.8% to 64.1%). Using logistic regression, the odds of a taxi driver wearing a helmet compared with a non-taxi driver was significantly increased 1.490 (p<0.01). The odds of helmet use on non-residential roads as compared with residential roads was also increased at 1.389 (p<0.01).

CONCLUSION

This novel method of estimating helmet use has produced results similar to traditional methods. Applying this technology can reduce time and monetary costs and could be used anywhere street imagery is used. Future directions include automating this process through machine learning.