Resolution Aerial Photographs Research Articles

Integrating Depth Measurements From Gaging Stations With Image Archives for Spectrally Based Remote Sensing of River Bathymetry

AbstractRemote sensing can be an effective tool for mapping river bathymetry, but the need for direct measurements to calibrate image‐derived depth estimates impedes broader application of this approach. One way to circumvent the need for field campaigns dedicated to calibration is to capitalize upon existing data. In this study, we introduce a framework for Bathymetric Mapping using Gage Records and Image Databases (BaMGRID). This workflow involves retrieving depth measurements made during gaging station site visits, downloading archived multispectral images, and then combining these two data sets to establish a relationship between depth and reflectance. We developed a processing chain that involves using application programming interfaces to obtain both depth measurements made during site visits and images centered on the gage and then linking depth to reflectance via an optimal band ratio analysis (OBRA) algorithm modified for small sample sizes. Applying this workflow to selected gages within two river basins indicated that depth retrieval from multispectral satellite images could be highly accurate, but with variable results from one image to the next at a given site. High resolution aerial photography was less conducive to bathymetric mapping in one of the basin considered. Of the four predictors of depth retrieval performance we evaluated (mean and standard deviation of depth, width, and an index of water clarity), only width was consistently significantly correlated with OBRA R2 (p < 0.026). Currently, BaMGRID is best‐suited for site‐by‐site analysis to support practical applications at the reach scale; continuous, basin‐wide mapping of river bathymetry will require additional research.

Spatial analysis of ecological risk in a coastal dune landscape using high resolution aerial photography

This study aims to investigate the dynamic pattern of landscape ecological units (LEUs) and analyse spatial variations of the ecological risk in Parangtritis coastal dune, Yogyakarta, Indonesia. A quantitative method was used in this research as part of landscape ecological analysis using a geographic information system. LEUs were interpreted by small format aerial photographs (SFAPs) and verified through field survey, then were calculated using the formula within grids to produce the ecological risk index (ERI) in the total area. According to the sub-class and class scenario, many LEUs showed changes in their landscape pattern. The ERI in the study area consisted of five levels (very low to very high), each of which was spatially varied. The ecological risk formed clusters coinciding with certain LEUs where fragility chiefly contributed to the sub-class scenario, while disturbance contributed to the class scenario.

Scattered tree death contributes to substantial forest loss in California

In recent years, large-scale tree mortality events linked to global change have occurred around the world. Current forest monitoring methods are crucial for identifying mortality hotspots, but systematic assessments of isolated or scattered dead trees over large areas are needed to reduce uncertainty on the actual extent of tree mortality. Here, we mapped individual dead trees in California using sub-meter resolution aerial photographs from 2020 and deep learning-based dead tree detection. We identified 91.4 million dead trees over 27.8 million hectares of vegetated areas (16.7-24.7% underestimation bias when compared to field data). Among these, a total of 19.5 million dead trees appeared isolated, and 60% of all dead trees occurred in small groups ( ≤ 3 dead trees within a 30 × 30 m grid), which is largely undetected by other state-level monitoring methods. The widespread mortality of individual trees impacts the carbon budget and sequestration capacity of California forests and can be considered a threat to forest health and a fuel source for future wildfires.

Automated mapping of bedrock-fracture traces from UAV-acquired images using U-Net convolutional neural networks

This contribution presents a novel U-Net convolutional neural network (CNN)-based workflow for automated mapping of bedrock fracture traces from 0.55 cm spatial resolution aerial photographs, acquired by unmanned aerial vehicles (UAV), over the Wiborg Rapakivi granite outcrops in the islands off the coast of the Loviisa Region in Southern Finland. The workflow comprised training a U-Net CNN using a small subset of photographs with manually traced fractures for optimizing the network parameters using the root mean squared propagation optimizer and sigmoidal focal loss function for semantic segmentation of input images and pixel-wise identification of fracture traces. The ridge detection algorithm was then applied to the U-Net prediction results, followed by vectorization of the fracture-traces pixels as vector polylines representing the traces of fractures. Both intensity values of the pixels and topological connectivity were used in the process of vectorization. Quantitatively the results were assessed using various accuracy assessment metrics. Qualitative evaluations of the results were implemented by comparisons of orientations and length-frequency distributions of automatically- and manually-mapped traces.The results show that the model has the class-balanced accuracy score 0.945, predicting 88.79% of the fracture-trace pixels. Bedrock outcrops with well-exposed surfaces present high true positive rates (>99%). The demonstration (test) site has the class-balanced accuracy score of 0.873 and 75% true positive rate. Additionally, the fracture trace networks replicate the orientation distributions of the manually digitized traces. The length distributions of automatic traces, however, differ with varying intensity from the manual trace length distributions. However, this study demonstrates that the workflow can be successfully applied to UAV-acquired images for fast and efficient automated mapping of bedrock-fracture traces during the initial phases of structural characterization of a region.

Dynamic pressure evolution within the 18 May 1980 Mount St. Helens pyroclastic density current: evidence from tree damage

The effects of pyroclastic density currents (PDCs) can be devastating, so understanding their internal dynamics and evolution is important for hazard assessment. We use damaged trees located around Mount St. Helens (USA) as proxy for the dynamic pressure (Pdyn) of the PDC erupted on 18 May 1980. We recorded the location, distribution, and foliage preservation of damaged trees within the medial and distal parts of the devastated forest. Sub-meter resolution aerial photographs from a month after the eruption allow distinction between standing trees that retained foliage from those that were stripped. Heights of standing trees were estimated from the measured lengths of their shadows. The number of standing trees was counted within defined areas along the propagation paths of PDCs. From the measured tree heights, we estimated tree toppling stresses from Pdyn. Overall, Pdyn of the PDC head within the medial to distal portions of the blowdown zone ranged from 10 to 35 kPa. Pdyn likely waned with distance, as shown by the increased number of standing trees in the outer parts of the devastated area. In addition, we find clusters of standing trees on the lee sides of some hills. We propose that these clusters survived because they were primarily impacted by lower dynamic pressures extant within the PDC body, with foliage retention or stripping as a function of the Pdyn evolution in the PDC body. We estimate that Pdyn of the body was less than the estimated maximum Pdyn of the PDC head by 12 ± 4 kPa.

Beach narrowing on prograding coasts: Examples from the tropics to subtropics of eastern Australia

Subtropical to tropical sandy beaches are subject to a wide variety of processes that influence shoreline position, though they receive little research attention compared to temperate coastlines. This presents challenges to understanding the drivers of long-term coastal change, particularly in places like the sub/tropical margins of Australia, where there are significant differences in local to regional geologic, oceanographic, and climatic conditions. Here, we investigate changes in beach planform morphology for a number of beaches in Queensland, Australia, over a period of up to 90 years using high resolution (0.5–5 m pixel size) aerial photographs. In total, 945 images were orthorectified and bounding features for the subaerial beach (high water and dune vegetation lines) were digitised to extract shoreline change statistics for 15 beaches ranging from the tropics to subtropics, between 15°S and 28°S. Three beach types were identified, defined based on lateral boundary features and the scale of shoreline/beach width variability. These include low variability beaches with two bounding rocky headlands (type 1); beaches with one headland and one river boundary, with the most significant shoreline adjustment away from the headland (type 2); and highly variable beaches with two river boundaries (type 3). Our change analysis demonstrates overall seaward displacement of the high water and vegetation lines, but also net losses in beach width due to fast rates of seaward vegetation advance in a tropical setting. These changes are largely consistent along this coastline, despite differences in exposure to deep-water waves and large variation in tidal ranges. Long-term accretion persists here despite rising sea levels over the past century. This result suggests a long-term net positive sediment budget along this section of the coast that is likely a reflection of global adjustments to tropical climate, and associated increases in terrestrial sediment generation and dune vegetation growth rates. Overall, this study provides insights into a variety of coastal settings through the subtropics and tropics and highlights the importance of tracking change along both the landward and seaward edges of the subaerial beach, regardless of exposure and associated wave regimes.

Spatiotemporal Patterns of Leymus arenarius Nebkhas during Early Primary Succession on the Sub-Arctic Coastal Sand Plains of Iceland

Greipsson, S., 2022. Spatiotemporal patterns of Leymus arenarius nebkhas during early primary succession on the subArctic coastal sand plains of Iceland. Journal of Coastal Research, 38(2), 269–278. Coconut Creek (Florida), ISSN 0749-0208. A spatiotemporal analysis of phytogenic mounds (nebkhas) of the pioneer dune building grass Leymus arenarius was assessed during primary succession on three coastal sand plains in Iceland. Extensive sand plains were formed on the study sites due to catastrophic glacial river floods associated with sub-glacial volcanic activities. Present and past spatial patterns of Leymus-nebkhas were examined by the use of satellite images and high resolution aerial photographs and confirmation through field work. The colonization patterns of the nebkhas were studied from the initial phases and up to a maximum period of 54 years on the sand plains Skeidararsandur, Myrdalssandur, and Skogarsandur. The hypothesis was that nebkhas would be clustered and nonrandom in early phases of the primary succession but would have random configurations at later stages. Average density of nebkhas was low within 10 years of colonization on Skeidararsandur (0.55–0.67 dunes ha–1) and the spatial configuration was found to be highly nonrandom (clumped). Following the initial colonization of L. arenarius, the increase of nebkha density with time followed a similar nonrandom pattern on Myrdalssandur with final average density of 6.12 nebkha ha–1. On Skogarsandur, the spatial configuration of nebkhas was random with final average density of 7.08 nebkha ha–1. Nonlinear cubic polynomial regression analysis showed a very strong relationship (R2 = 0.989) for the average density of nebkha for 60 years on Skogarsandur. On the coastal sand plains, the spatial configuration of nebkha was observed to progress with time from nonrandom towards random distribution. This study supports the view of deterministic spatiotemporal pattern of nebkhas during early primary succession on the coastal sand plains. The nebkha density was most likely a function of sand drift on the plains, significantly decreasing (R2 = 0.93) away from the sand source of the glacier river Jökulsa on Skogarsandur. The results are discussed in relation to ecological restoration.

Aleutian Tern Onychoprion aleuticus abundance estimates at four globally significant colonies

Summary Aleutian Tern Onychoprion aleuticus numbers have been in steep decline at known Alaskan breeding colonies in recent decades (IUCN recently uplisted to ‘Vulnerable’). Available data suggest that most of the species may currently breed in Russia. Efforts to document global abundance and trends have been hampered by remoteness of colonies, lack of a formal monitoring programme, and the absence of reproducible population estimates with quantifiable errors, especially for large colonies. We surveyed four historically large colonies in Russia (2018) and Alaska (2019), which together may comprise 30–50% of the global breeding population. At each colony we obtained high resolution aerial photographs using a small Unmanned Aircraft System (sUAS). The large size of the colonies and the minimum altitude required to identify terns made it impractical to collect imagery of the entire colony. Instead, we employed a sampling approach, with sample locations selected based on spatially balanced acceptance sampling. Statistically sampled, low altitude sUAS images provided a fast, reproducible, and rigorous count of abundance for geographically large colonies, with low disturbance, and were generally consistent with concurrent ground-based observations. Concurrence among observers in photo counts indicated high precision in counts of attending birds and unattended nests, although species attribution in mixed tern colonies remains a source of significant uncertainty. Our results indicate that the four colonies surveyed here together supported <2,500 pairs of Aleutian Terns in the survey years. None of the colonies approached their peak size reported previously, likely due to recent predation, long-term decline, cold early season weather, or other factors. If these reduced colony sizes are representative of the current conditions, the implications for the global population would be dire.

Where to start? A new citizen science, remote sensing approach to map recreational disturbance and other degraded areas for restoration planning

This UN Decade on Ecosystem Restoration highlights the capacity of restoration to mitigate trends in biodiversity loss and land degradation. However, many managers lack the tools they need to systematically and comprehensively identify degraded sites to prioritize restoration efforts given limited resources. We developed a novel, inexpensive, low‐tech approach for training and engaging citizen scientists to identify recreational impacts and other degraded areas within a defined unforested area. The mapping process follows four phases: (1) Landscape scans by citizen scientists using Google Earth Pro imagery; (2) A second scan of all marked sites based on high resolution aerial photography; (3) Compilation of basic information about the degraded sites; (4) Addition of associated soil type and plant communities. In the 12,375 ha McDowell Sonoran Preserve (Scottsdale, Arizona), we detected 67 new sites not previously identified by land managers, using an estimated 305 citizen scientist hours and only 30 staff hours. Each site has accompanying information including distance from nearest access point, cause of degradation, and plant and soils detail. After completion, we conducted independent field visits of 33% of the detected sites and verified degradation in all cases. We found that the remotely sensed approach provided better perspective to accurately measure the scale and original source of degradation compared with field visits. The approach can be conducted over a short period of time using citizen scientists, allows managers to undertake landscape level restoration prioritization and planning, and, if repeated, can be used to monitor changes in degradation and restoration over time.

Characterizing Near‐Field Surface Deformation in the 1990 Rudbar Earthquake (Iran) Using Optical Image Correlation

AbstractThe 1990 Rudbar earthquake (Mw 7.3) broke along three right‐stepping segments of the left‐lateral Rudbar fault, in the western Alborz mountains (Iran), producing ∼80 km of surface rupture. Previously reported horizontal (∼60 cm) field offsets were surprisingly low, and the vertical (∼95 cm) unexpected, given the relatively large moment release, shallow source depth, vertical fault geometry, and left‐lateral rake. We characterize the surface displacement from optical image correlation of satellite and aerial images. A maximum displacement of ∼6.5 m is observed in the eastern part of the rupture consistent with both a sharp pulse in the moment release rate ∼10 s after the rupture initiation, and the macroseismic isoseismal curves. We find ∼2.2 m mean surface slip, larger than field measurements, but consistent with empirical scaling laws based upon the seismological moment release. Estimates of fault zone width based on correlation of 1 m resolution aerial photos (and corrected for smoothing bias) are <200 m; these values increase artificially with lower resolution imagery, which is more affected by smoothing bias. The large mismatch between field and correlation displacements suggests the Rudbar earthquake probably featured significant off‐fault deformation, consistent with its low structural maturity. Finally, we document a surface rupture along a structural continuation of the Kabateh segment, east of the Sefid Rud river. Slip along this segment (1–2 m), is less than the neighboring Zard Geli segment (1.5–6 m). The two faults converge at Sefid Rud, which appears to represent a major structural feature controlling fault segmentation in this region.

Canopy chlorophyll content retrieved from time series remote sensing data as a proxy for detecting bark beetle infestation

The European spruce bark beetle (Ips typographus, L.) is an invasive species resulting in a high degree of fragmentation, forest productivity, and phenology. Understanding its biology and its early detection based on its behaviour is essential for its successful management and eradication. This study demonstrates the potential of the canopy chlorophyll content (CCC) product retrieved from remote sensing datasets to detect early bark beetle infestations in the Bavarian Forest National Park. Time series CCC maps were generated through radiative transfer model inversion of images from RapidEye and Sentinel-2 (2011–2018). The CCC products were then classified into stressed and healthy classes using calculated CCC mean and variance values obtained for infected and healthy Norway spruce trees in 2016. The location of infected plots obtained from the interoperation of resolution (0.1 m) aerial photographs was used as reference data to validate the accuracy of the infestation maps generated from CCC. Validation of the infestation maps indicated a classification accuracy of up to 78%. Our results demonstrated that CCC products derived from satellite remote sensing data were a rigorous proxy for early detection of bark beetle infestation. Hence, CCC products may play a significant role in understanding the dynamics of the Infestation and improving the management of bark beetle outbreaks in forest ecosystems. Inclusion of other remotely sensed plant traits as additional parameters in the model, such as dry matter and nitrogen, may further improve the accuracy of early detection of bark beetle infestation using satellite remote sensing.

Monitoring shoreline shift under rapid sea-level change on the Caspian Sea observed over 60 years of satellite and aerial photo records

The Caspian Sea is the world’s largest inland water body and endures rapid fluctuations in sea level and shoreline changes. The present study investigates and analyzes shoreline changes in the two areas of Tajan and Sefidrud on the southern coasts of the Caspian Sea using aerial photos, affixed with TM, ETM, and OLI images from Landsat satellites, as well as images rendered by Google Earth, all within the time period from 1955 to 2018. Shoreline changes were analyzed using the waterline algorithm on high resolution aerial photos and images from the Google Earth engine. Shoreline extraction on Landsat images was conducted after comparing six different indices, of which the wetness index exhibited the least mean error (1.3 m). Calculations on shoreline changes were undertaken in 4 time periods using the DSAS extension in Arc GIS 10.3 including 1955–1977, 1977–1995, 1995–2002 and 2002–2018. Coastal accretion on the Tajan shoreline after decreases in sea level were evident in the first, third, and fourth time periods, while coastal erosion was discernible in the second time period during significant increases in sea level. The situation, however, was different for the coasts of Sefidrud. The shoreline was divided into 14 distinct parts where the shoreline change rate was inconsistent with the actual changes in sea level. The dismatch between sea level and shoreline changes in the Sefidrud region points to the existence of a different set of factors, in addition to the sea level, which significantly affect the status of the coastlines.

Resilience of trees and the vulnerability of grasslands to climate change in temperate Australian wetlands

Observations from wetlands across the globe suggest a consistent pattern of woody encroachment into wetland grasslands, altering habitat structure and ecological function. The extent to which hydrological changes have contributed to woody invasion of wetland grasslands is unclear. Our objective was to compare rates of woody encroachment in Australian floodplain wetlands between wet and dry hydrological phases. We test the hypothesis that contraction of non-woody wetland vegetation (grasses and rushes) would be concentrated in dry phases, co-incident with recruitment of the River Red Gum Eucalyptus camaldulensis lower in the floodplain. We conduct the first detailed mapping of habitat change in two of the largest forested wetlands in inland Australia, comparing wet and dry hydrological phases. Detailed photogrammetry, supported by extensive ground survey, allowed the interpretation of high resolution aerial photography to vegetation community level. We found a consistent pattern of decline in non-woody vegetation, particularly amongst grasses utilising the C4 photosynthetic pathway. The C4 grasses Pseudoraphis spinescens and Paspalum distichum showed steep declines in the Barmah Millewa and Macquarie Marshes respectively, being replaced by River Red Gum E. camaldulensis. C3 sedges proved more resilient in both systems. Our results suggest that a pattern of tree expansion into non-woody wetland vegetation, characteristic of wetlands across the globe, is a major habitat structural change in the Australian floodplain wetlands studied. Projected hydrological impacts of climate change are likely to further restrict wetland grass foraging habitat in these semi-arid floodplain wetlands.

Диагностика и картографирование Местообитаний южных тундр (на примере ключевого участка р. Шапкина)

A classification of habitats is given for the area studied on the right bank of the Shapkina River, (East European tundra, southern subzone). The classification is based on the topographic position of biotopes and their plant communities. The mosaic of tundra vegetation makes it difficult for mapping. Usually in nature the combinations of plant communities are confined to specific types of habitats. To display the regularities of a fine-contour vegetation cover on maps, we used the habitat approach diagnosed by combinations of plant communities. The field survey was done in the summer of 2020; total area of about 150 km2 was surveyed. Ultra-high resolution (3–5 cm/px) aerial photography was carried out for key areas, using a DJI Mavic Pro Platinum quadrocopter (shooting height from 80 to 200 m). 25 geobotanical relevés were completed; in addition 180 short descriptions were made for map verification. All types of habitats in the proposed scheme are correlated with EUNIS units, lists of syntaxa are given. A large-scale map (1 : 50 000) of habitats was prepared. All tundra habitats are divided into two groups. First level of habitat classification confined to the main landscape types: watersheds and river valleys of watercourses with a floodplain regime. Watershed habitats are subdivided into 5 categories (second level of habitat classification), determined by their runoff-geochemical position on the generalized geomorphological profile (from the highest relief elements to the lowest ones), including underlying rocks, moisture regime and migration of elements, exposure features. Due to the small amount of data mapping of river valley habitats was performed only for units of the second level. At the third level watershed habitats are well diagnosed by vegetation at the level of associations, combinations of commuities, soil cover, and microrelief. As a result, most units of the second and third levels are clearly distinguished on aerial photography obtained using a quadrocopter, and also correlate with specific syntaxa. As a result of field data and aerial photographs analyses, 12 categories of habitats, represented by 17 syntaxa, were identified for watersheds at the 3rd level.

Визуальные свойства ландшафтов и методы их оценки с применением ГИС (на примере Белоградчишских скал (Болгария))

Visual characteristics of landscapes are important factors for the assessment of tourist and recreational potential of territories. At present, a number of methodological approaches are applied to assess the visual characteristics of landscapes. They can be divided into traditional, associated exclusively with field research, and innovative, which is based on remote sensing data (RSD) of high spatial resolution and GIS technologies. Field assessment of the visual quality of landscapes utilizes a system of numerous elementary indicators to minimize subjectivity of assessment. They are conducted within separate areas or touristic routes. In its turn, modern GIS and high quality of remote sensing data allow assessing of most indicators of the visual quality of landscapes for any observation point on the entire territory. The main task of our research is to verify the results of automated processing of ultra-high resolution aerial photographs obtained from unmanned aerial vehicles (UAV) by field observations on a touristic route. The research was carried out on the territory of the “Belogradchik Rocks” Geopark (North-West Bulgaria). In our study, we estimated 4 out of 28 aesthetic indicators—the amount of mountain peaks visible from a site, the amount of mountain peaks on the skyline, the percentage of the forest-covered area, and the amount of open spaces in the wooded landscape. The obtained results confirmed that our approach allows calculating these aesthetic indicators at an accuracy level comparable to field observations.

Визуальные свойства ландшафтов и методы их оценки с применением ГИС (на примере Белоградчишских скал (Болгария))

Visual characteristics of landscapes are important factors for the assessment of tourist and recreational potential of territories. At present, a number of methodological approaches are applied to assess the visual characteristics of landscapes. They can be divided into traditional, associated exclusively with field research, and innovative, which is based on remote sensing data (RSD) of high spatial resolution and GIS technologies. Field assessment of the visual quality of landscapes utilizes a system of numerous elementary indicators to minimize subjectivity of assessment. They are conducted within separate areas or touristic routes. In its turn, modern GIS and high quality of remote sensing data allow assessing of most indicators of the visual quality of landscapes for any observation point on the entire territory. The main task of our research is to verify the results of automated processing of ultra-high resolution aerial photographs obtained from unmanned aerial vehicles (UAV) by field observations on a touristic route. The research was carried out on the territory of the “Belogradchik Rocks” Geopark (North-West Bulgaria). In our study, we estimated 4 out of 28 aesthetic indicators—the amount of mountain peaks visible from a site, the amount of mountain peaks on the skyline, the percentage of the forest-covered area, and the amount of open spaces in the wooded landscape. The obtained results confirmed that our approach allows calculating these aesthetic indicators at an accuracy level comparable to field observations.

An object based image analysis applied for volcanic and glacial landforms mapping in Sahand Mountain, Iran

Landform mapping has increasingly become part of the digital domain. While the majority of approaches evaluates Digital Elevation Models (DEM) on a per-pixel basis, some examples exist were object-based image analysis (OBIA) has been applied to terrain data to identify a variety of landforms, including glacial landforms. The main objective of this study is to develop a semi-automated object-based rule set for detecting and delineating volcanic and glacier landforms in the area of the Sahand Mountain, Northern Iran. First, we applied a multi-resolution segmentation algorithm on a freely available Sentinel-2 optical satellite image and then selected the relevant features to define appropriate segmentation scales for each landform category. Object-based rule sets were then developed using spatial (DEM and its derivatives, e.g. slope, aspect, curvature and flow accumulation) and spectral information. Volcanic and glacial landforms were detected and classified into eight classes: caldera, volcanic cone, tuff formation, andesite lava, dacite lava, glacier valley, suspension valley, glacier cirque. An accuracy assessment was applied based on the fuzzy synthetic evaluation technique, together with the error matrix and kappa coefficient, using field data and geomorphological units derived from geological maps and very high resolution aerial photographs. The resulting overall accuracies for each class were 96.2%, 93.3%, 92.4%, 94.2%, 93.01, 95.1, 90.1 and 90.5, respectively. Our research demonstrated that spatial (e.g. density, shape index, length/width) and spectral (e.g. mean, brightness and standard deviation) algorithms together with a grey-level co-occurrence matrix (GLCMs) were efficient features for detecting and mapping volcanic and glacial landforms. We conclude that the OBIA-based algorithms and features provide a high potential for detecting and classifying landforms. Results of this study are of great importance for geomorphology and geology as well as geo-tourism societies and the semi-automated landform mapping contributes to the framework of GIScience.

Spatial metrics from LiDAR roof mapping for fire spread risk assessment of informal settlements in Cape Town, South Africa

The risk of fire spread in informal settlements is significant and can be analysed as a function of the spatial arrangement of dwellings. Spatial metrics representing density and shape of dwellings are proposed as a method to identify settlements at high risk of fire spread. LiDAR data is used to map dwelling roofs for informal settlements in the City of Cape Town, South Africa. The LiDAR roof dataset is validated against a visually interpreted dataset digitised from 6 cm resolution aerial photography and is found to have an overall completeness and correctness accuracy of greater than 75% with systematic underrepresentation of roofs in the LiDAR dataset. Correlation analysis of metrics derived from the LiDAR dataset and the reference dataset indicates that only the edge density and landscape density metrics could be applied with confidence to all the settlements. These two metrics are then applied to the informal settlements of Kosovo and Imizamo Yethu. A high landscape density in combination with a low edge density is found to be indicative of fire spread risk. This study represents a first step in the development of spatial metrics for understanding informal settlement fire spread risk.

Plant Diversity Along the Urban–Rural Gradient and Its Relationship with Urbanization Degree in Shanghai, China

Urbanization is one of the major causes for plant diversity loss at the local and regional scale. However, how plant species distribute along the urban–rural gradient and what the relationship between urbanization degree and plant diversity is, is not very clear. In this paper, 134 sample sites along two 18 km width transects that run across the urban center of Shanghai were investigated. We quantified the spatial patterns of plant diversity along the urban–rural gradient and measured the relationship between plant diversity and urbanization degree, which was calculated using a land use land cover map derived from high spatial resolution aerial photos. We recorded 526 vascular plant species in 134 plots, 57.8% of which are exotic plant species. Six spatial distribution patterns of species richness were identified for different plant taxa along the rural to urban gradient. The native plant species richness showed no significant relationship to urbanization degree. The richness of the all plants, woody plants and perennial herbs presented significant positive relationship with urbanization degree, while the richness of annual herbs, Shannon-Wiener diversity and Heip evenness all exhibited a negative relationship to urbanization degree. Urbanization could significantly influence plant diversity in Shanghai. Our findings can provide insights to understand the mechanism of urbanization effects on plant diversity, as well as plant diversity conservation in urban areas.

Deep learning in the built environment: automatic detection of rooftop solar panels using Convolutional Neural Networks

Mapping the location and size of solar installations in urban areas can be a valuable input for policymakers and for investing in distributed energy infrastructures. Machine Learning techniques, combined with satellite and aerial imagery, allow to overcome the limitations of surveys and sparse databases in providing this mapping at large scale. In this paper we apply a supervised method based on convolutional neural networks to delineate rooftop solar panels and to detect their sizes by means of pixel-wise image segmentation. As input to the algorithm, we rely on high resolution aerial photos provided by the Swiss Federal Office of Topography. We explore different data augmentation and we vary network parameters in order to maximize model performance. Preliminary results show that we are able to automatically detect in test images the area of a set of solar panels at pixel level with an accuracy of about 0.94 and an Intersection over Union index of up to 0.64. The scalability of the trained model allows to predict the existing solar panels deployment at the Swiss national scale. The correlation with local environmental and socio-economic variables would allow to extract predictive models to foster future adoption of solar technology in urban areas.