High-resolution Digital Surface Research Articles

Rapid viewshed analyses: A case study with visibilities limited by trees and buildings

High-resolution digital surface models improve the precision of viewshed analyses, but these data can be challenging to use for large study areas. We developed a 2D viewshed (2DView) approach, that uses land cover to estimate viewshed extent. We aimed to develop a practical approach to map viewing areas for >1000 waterbodies spread over 6000 km2. In this paper, we present a case study in which we use the 2DView method for a viewshed analysis where trees and buildings, rather than terrain, are the primary obstacles to visibility. We evaluate the 2DView through comparison with DSM-based viewsheds.We used 30 waterbodies for our evaluation. We applied our 2DView analysis using high-resolution (i.e. 1 m) land cover and building footprint data to represent trees and buildings, respectively. For conventional 3DView, we used 1) true digital surface models and 2) bare-earth elevation models modified by adding fixed heights (i.e. 30 m) for trees and buildings. The 2DView method had good agreement with the 3DView method with average intersection-of-unions ranging from 90 to 95%. The 2DView method tends to overestimate viewshed extents; thus, it can serve as a pre-screening tool to reduce the instances when more accurate 3DView analyses are required.

Antecedent rainfall as a critical factor for the triggering of debris flows in arid regions

Abstract. In arid regions, debris flows (DFs) are occasionally observed when heavy rainfall hits steep slopes with unconsolidated materials. Despite the related hazards, not much is yet known about the critical rainfall conditions for DF initiation in dryland areas mainly due to a lack of observations and data. In this study, we use high-resolution digital surface models and orthophotos to detect DFs on the arid slopes of the Dead Sea's northwestern margins, as well as weather radar rainfall estimates for the characterization of the triggering and non-triggering storms. We identify deposits from over 40 relatively small DFs that occurred between 2013 and 2019. We divide them into four groups based on their spatial distribution and triggering period, and we identify the most likely triggering storms based on weather radar data and additional information. We show that the DFs were likely triggered by an intense convective cell (lasting less than 45 min) which was preceded by non-negligible rainfall amounts (about 10 mm) delivered during the storm. Comparing triggering and non-triggering storms, we observed that rain intensity alone is insufficient to explain the phenomena and discuss the possibility that antecedent rainfall could represent a critical factor for the triggering of DFs on steep slopes of arid environments.

Study on the Prediction of Debris Flow Risk Using LS-RAPID

In this study, the LS-RAPID program was applied to evaluate the debris flow deposition area using the field surveyed topographical data in the areas of debris flow occurrence. In 2020, aerial photogrammetry was performed using an unmanned aerial vehicle in Seonseri, Osan-myeon, and Gokseong-gun. The erosion and deposition areas owing to debris flow were estimated using the orthographic images generated from photogrammetry and high-resolution Digital Surface Mode (DSM), and the extent of debris flow was accurately estimated. For debris flow simulation, the LS-RAPID program, with the apparent friction angle of the Coulomb's friction model as an important factor, was applied. The simulation results were significantly affected by the quality of the topographical data. It is essential to apply DSM data for obtaining a highly accurate simulation of the deposition area where damage generally occurs. To evaluate the performance of the simulation, the LSSI index with the deposition area as a factor was applied; it was found to be 65%, showing a superior predictive performance. In conclusion, it is possible to predict the debris-flow damage using LS-RAPID simulation, and the DSM data on deposition areas are required to generate a reliable disaster map.

Morphometric analysis of hillslope evolution in the Kadisha River Basin based on archived aerial photographs

The morphometric analysis of hillslope evolution, using the traditional method of interpreting archived aerial photographs, requires both considerable time and subjectivity. It is well-established that the Structure from Motion (SfM) approach, a photogrammetric method producing high resolution Digital Surface Models (DSM), is more effective. This study aims to determine how hillslope evolution in the upper basin of Kadisha watershed can be better analysed for an accurate morphometric analysis. It investigates and validates the influence of the applied Morphometric Slope Index (MSI) on morphology, depth, and erosion volume, as well as the processes of the computed pre-erosion surfaces in GIS environment. To enable testing the hypothesis that the characteristics of drainage networks are highly affected by the MSI, the following drainage network parameters were quantified in the thirty-two sub-basins: MSI, Drainage Density (D d ), average slope (ASL), depth, and eroded volume (V). Correlation between morphometric features and erosion was identified to reveal the mechanism of hillslope evolution. The results largely support what was hypothesized: MSI was useful in capturing single slope morphometric characteristics in sub-basins. The process revealed that pre-erosion MSI is a desired slope morphometry indicator and a hillslope evolution predictor in the study area.

Alfalfa Biomass Estimation Using Crop Surface Modeling and NDVI

<b>Highlights</b> <list list-type=bullet><list-item> A UAV with NDVI and RGB cameras captured aerial images of alfalfa plots before and after harvest. </list-item><list-item> Structure from Motion (SfM) and crop surface models were used to determine change in alfalfa crop height. </list-item><list-item> An equation to predict dry matter fraction (DMF) as a function of NDVI was developed. </list-item><list-item> Dry alfalfa yield models using DMF and change in crop height were developed. </list-item></list> <b>Abstract.</b> Alfalfa is an important forage crop grown for hay, silage, and pasture production. Accurate yield estimations before harvest or grazing are crucial for producers to optimize forage utilization. The goal of this research was to develop an aboveground dry biomass prediction function for alfalfa using remote sensing from an unmanned aerial vehicle (UAV). A DJI Mavic Pro equipped with RGB and NDVI cameras captured aerial images. High-resolution orthomosaics and digital surface models were obtained using Structure from Motion (SfM) with Agisoft Metashape. An equation to estimate wet biomass was developed using three variables: change in canopy height (∆H) from digital elevation and crop surface models, and NDVI based canopy density index (CDI) data. The dry biomass yield was estimated as the product of a wet biomass prediction function and a correlation equation to estimate the dry matter fraction (DMF). The best correlation equation for wet biomass (BMwet) only required SfM methods to measure ΔH. The linear regression equation for BMwet had an R2 of 0.963 and the mean coefficient of variation (CV) was ±23%. The best prediction equation for DMF was a quadratic equation with an R2 of 0.642 with a mean CV of ±11%. It was also determined that the DMF varied significantly by season of harvest. As a result, the dry biomass could be estimated using the DMF equation or seasonal mean DMF values. Comparison of the observed dry biomass measurements with the empirically determined biomass prediction function that used the NDVI data to estimate the DMF showed excellent agreement. The model underpredicted the dry biomass by 1.1% with a standard error of the estimate of 236 kg DM/ha (CV = ±26%). The model predictions using the seasonal mean DMF values were not significantly different.

The use of unmanned aerial vehicle (UAV) data in village development plans: A case study of Aksaray Yaylak Village

With the increasing population, the problem of people's need for shelter has also emerged. In many settlements, structures were built without the necessary planning. Three-dimensional (3D) models of these lands should be created in order to make land management plans. Technological improvements in remote sensing have made easier the terrain modelling. To make a better land management plan of rural areas, we need high resolution maps. Unmanned aerial vehicle (UAV) is one of the methods in map production, which can obtain spatial data with high accuracy. In this study, we created 3D model of Yaylak village located in Aksaray District using UAV Photogrammetry. We created high resolution Digital Surface Map and orthophoto map of the village using the pictures obtained from e-Bee. Using these maps, the administrator of the region will decide healthy decision. This study shows us UAV can be used in preparing the necessary maps for designing the settlement area.

Surface Water Extent Mapping in Denmark: Comparing Airborne Thermal Imagery and Satellite Earth Observation

Mapping and prediction of inundated areas are increasingly important for climate change adaptation and emergency preparedness. Flood forecasting tools and flood risk models have to be compared to observe flooding patterns for training, calibration, validation, and benchmarking. At the regional to continental scales, satellite earth observation (EO) is the established method for surface water extent (SWE) mapping, and several operational global-scale data products are available. However, the spatial resolution of satellite-derived SWE maps remains a limiting factor, especially in low-lying areas with complex hydrography, such as Denmark. We collected thermal imagery using an unmanned airborne system (UAS) for three areas in Denmark shortly after major flooding events. We combined the thermal imagery with an airborne lidar-derived high-resolution digital surface model of the country to retrieve high-resolution (40 cm) SWE maps. The resulting SWE maps were compared with low-resolution SWE maps derived from satellite earth observation and with potential flooded areas derived from the high-resolution digital elevation model. We conclude that UASs have significant potential for SWE mapping at intermediate scales (up to a few square kilometers), can bridge the scale gap between ground observations and satellite EO, and can be used to benchmark and validate SWE mapping products derived from satellite EO as well as models predicting inundation.

Accelerating Tsunami Modeling for Evacuation Studies through Modification of the Manning Roughness Values

The role of the Manning roughness coefficient in modifying a tsunami time series of flow depth inundation was studied in Iquique, Chile, using a single synthetic earthquake scenario. A high-resolution digital surface model was used as a reference configuration, and several bare land models using constant roughness were tested with different grid resolutions. As previously reported, increasing the Manning n value beyond the standard values is essential to reproduce mean statistics such as the inundated area extent and maximum flow depth. The arrival time showed to be less sensitive to changes in the Manning n value, at least in terms of the magnitude of the error. However, increasing the Manning n value too much leads to a critical change in the characteristics of the flow, which departs from its bore-like structure to a more gradual and persistent inundation. It was found that it is possible to find a Manning n value that resembles most features of the reference flow using less resolution in the numerical grids. This allows us to speed up inundation tsunami modeling, which could be useful when multiple inundation simulations are required.

An open access geospatial database for the sub-Antarctic Prince Edward Islands

Researchers of projects at the sub-Antarctic Prince Edward Islands are increasingly considering geospatial data as an essential component in answering scientific questions. A need exists for high-resolution geospatial data in both multi- and transdisciplinary research to better analyse fine-scale biotic–abiotic interactions of the Islands’ landscape and ecosystems within the context of climate change and the impacts of invasive species. However, much of the geospatial data that currently exist have limitations in spatial coverage and/or resolution, are outdated, or are not readily available. To address these issues, we present an online geospatial database for the Prince Edward Islands (both islands) produced from a high-resolution digital surface model and satellite imagery. This database contains vector files, raster data sets, and maps of topographical and hydrological parameters. It is freely available to download from Figshare – an open access data repository. We encourage the South African polar science community to make use of similar platforms for improved data sharing practices. Significance: A topographical and hydrological geospatial database – produced from a 1 m x 1 m digital surface model of the Prince Edward Islands – is provided. These fine-scale geospatial data allow for a more comprehensive assessment of biotic–abiotic interactions at an island scale. Also included are locality maps specifying place names and established long-term marine mammal monitoring beaches and coastal zones for improved cross-referencing. The dataset is downloadable from an open access data repository and intended to promote open science and data sharing practices. Open data set: https://doi.org/10.6084/m9.figshare.19248626

Mapping snow depth and spatial variability using SFM photogrammetry of UAV images over rugged mountainous regions of the Western Himalaya

Accurate snow depth observations are one of the most important inputs in the process of avalanche hazard estimation, forecasting, and mitigation. This study presents the high-resolution snow depth mapping of avalanche-prone regions of the Manali–Dhundi area, Himachal Pradesh (HP), India, using, repeated UAV Photogrammetric surveys. The images captured by an RGB camera mounted on a fixed-wing UAV (e-Bee X) were used to generate the high-resolution Digital Surface Model (DSM) and Orthophotos of the study area. Mapping of snow depth of the study area is crucial for avalanche hazard assessment and safe movement of the personnel and vehicles passing through the highways in this area during winters. A total of 12 km2 area covering six major avalanche sites was surveyed. A snow depth map was generated by differencing co-registered snow-covered DSM and bare surface DSM. UAV-derived snow depth values have been validated with the point snow depth measurements of Wireless Sensor Networks (WSN), manual observation station, and snow depth around a structure of known height in the survey area. The UAV-retrieved snow depth values were found in good correlation with field-measured snow depth (R = 0.95). The maximum and minimum observed snow depth values at different locations on open slopes and flat areas were 1.62 m and 0.79 m. The estimated Mean Error (ME) and RMSE based on these measured values were found 0.15 and 0.21 m, respectively. The study reveals that the snow depth map generated from the UAV survey provides reliable estimates of snow depth values for flat areas and open slopes but the method is not suitable for forested areas. The study also infers that the statistical measures (i.e., µ, σ, RMSE, etc.) derived from the complete survey dataset without considering topographic factors and features of the region will not provide realistic estimates of snow depth and its accuracy. The study suggests that regular monitoring (before and after every snow storm) of the snow depth using a UAV-based method is quick, cost-effective, and provides accurate snow conditions for operational use and planning in snow-bound regions.

High-resolution Digital Surface Model of the 2021 eruption deposit of Cumbre Vieja volcano, La Palma, Spain

Identifying accurate topographic variations associated with volcanic eruptions plays a key role in obtaining information on eruptive parameters, volcano structure, input data for volcano processes modelling, and civil protection and recovery actions. The 2021 eruption of Cumbre Vieja volcano is the largest eruptive event in the recorded history for La Palma Island. Over the course of almost 3 months, the volcano produced profound morphological changes in the landscape affecting both the natural and the anthropic environment over an area of tens of km2. We present the results of a UAS (Unoccupied Aircraft System) survey consisting of >12,000 photographs coupled with Structure-from-Motion photogrammetry that allowed us to produce a very-high-resolution (0.2 m/pixel) Digital Surface Model (DSM). We characterised the surface topography of the newly formed volcanic landforms and produced an elevation difference map by differencing our survey and a pre-event surface, identifying morphological changes in detail. The present DSM, the first one with such a high resolution to our knowledge, represents a relevant contribution to both the scientific community and the local authorities.

Applications of Image-Based Computer Vision for Remote Surveillance of Slope Instability

Landslides and slope failures represent critical hazards for both the safety of local communities and the potential damage to economically relevant infrastructure such as roads, hydroelectric plants, pipelines, etc. Numerous surveillance methods, including ground-based radar, InSAR, Lidar, seismometers, and more recently computer vision, are available to monitor landslides and slope instability. However, the high cost, complexity, and intrinsic technical limitations of these methods frequently require the design of alternative and complementary techniques. Here, we provide an improved methodology for the application of image-based computer vision in landslide and rockfall monitoring. The newly developed open access Python-based software, Akh-Defo, uses optical flow velocity, image differencing and similarity index map techniques to calculate land deformation including landslides and rockfall. Akh-Defo is applied to two different datasets, notably ground- and satellite-based optical imagery for the Plinth Peak slope in British Columbia, Canada, and satellite optical imagery for the Mud Creek landslide in California, USA. Ground-based optical images were processed to evaluate the capability of Akh-Defo to identify rockfalls and measure land displacement in steep-slope terrains to complement LOS limitations of radar satellite images. Similarly, satellite optical images were processed to evaluate the capability of Akh-Defo to identify ground displacement in active landslide regions a few weeks to months prior to initiation of landslides. The Akh-Defo results were validated from two independent datasets including radar-imagery, processed using state of the art SqueeSAR algorithm for the Plinth Peak case study and very high-resolution temporal Lidar and photogrammetry digital surface elevation datasets for the Mud Creek case study. Our study shows that the Akh-Defo software complements InSAR by mitigating LOS limitations via processing ground-based optical imagery. Additionally, if applied to satellite optical imagery, it can be used as a first stage preliminary warning system (particularly when run on the cloud allowing near real-time processing) prior to processing more expensive but more accurate InSAR products such as SqueeSAR.

Morphometric Analysis of Saz-Çayırova Drainage Basin using Geographic Information Systems and Different Digital Elevation Models

Drainage basin/watershed analysis based on morphometric parameters has an essential role in watershed management and planning. Reliable delineation of watersheds and drainage networks is critical for hydrological and geomorphological studies. Since access to high-resolution digital elevation models (DEMs) and digital surface models (DSMs) is costly, many researchers need to evaluate low-resolution open-source products. Several data sources produced from different surveying techniques are used in the morphometric analysis. In this study, five different datasets such as Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) GDEM, Shuttle Radar Topography Mission (SRTM) DEM, Advanced Land Observing Satellite (ALOS) DSM, National Aeronautics and Space Administration (NASA) DEM, and a DEM from topographic maps (TOPO DEM), were investigated based on morphometric parameters. The tests was carried out in the Saz-Çayırova Basin, which is one of the critical urbanization and industrialization regions of Kocaeli, Turkey. In this study, the TOPO DEM, whose horizontal resolution is 30 m, was produced from 1:25K scaled digitized topographical maps. It was used for comparative analysis, as in all DEMs sources. The morphometric parameters' result of the TOPO DEM was used as the reference data for comparing the results of the other DEMs sources. In addition, the mean absolute percentage error (MAPE) was used to compute the accuracy between the freely available DEMs and the TOPO DEM for each morphometric parameter. The outcomes of this study reveal that the most consistent results with the TOPO DEM are provided by SRTM DEM, following the NASA DEM.

Intra-gully mapping of the largest documented gully network in South Africa using UAV photogrammetry: Implications for restoration strategies

Gully erosion can reach alarming dimensions, and in several cases contributes significantly to soil loss and sediment yield in catchments. The studied example is one of the largest known gully networks in the world (its surface area is approximately 0.5 million m2 and volume 5 million m3), more than twice as large compared to the largest gullies described in peer-reviewed literature. To improve gully management strategies, it is not only critical to understand the factors that lead to erosion, but also to identify the extent of in-channel erosional and depositional features. Few studies consider in-channel patterns of erosion and deposition within gully systems. This study utilises a fixed-wing Unmanned Aerial Vehicle (UAV) and Structure-from-Motion (SfM) photogrammetry to create a high resolution orthomosaic and Digital Surface Model (DSM) to determine the position and spatial extent of erosional and depositional features within a large gully network in South Africa. Formation of such a large gully in relation to a relatively small drainage area is unusual but can be explained by the presence of dispersive soils and regional landscape incision. Longitudinal and cross-sectional profiles indicate that the gully network consists of three channel types (i.e. V-shaped, U-shaped and trapezoidal), each with a unique combination of erosional and depositional features. This study demonstrates the value of identifying and mapping of intra-gully features to better understand sediment transfer within the gully network to implement appropriate restoration strategies, which in this instance should focus on depositional zones to enhance channel decoupling.

Developing a Guideline of Unmanned Aerial Vehicle’s Acquisition Geometry for Landslide Mapping and Monitoring

Remote sensing data and techniques are widely used for monitoring and managing natural or man-made disasters, due to their timeliness and their satisfactory accuracy. A key stage in disaster research is the detailed and precise mapping of an affected area. The current work examines the relationship that may exist between the acquisition geometry of Unmanned Aerial Vehicle (UAV) campaigns and the topographic characteristics of an investigated area, toward landslide mapping and monitoring that is as accurate as possible. In fact, this work, concerning the systematic research of the acquisition geometry of UAV flights over multiple active landslides, is conducted for the first time and is focused on creating a guideline for any researcher trying to follow the UAV photogrammetric survey during landslide mapping and monitoring. In particular, UAV flights were executed over landslide areas with different characteristics (land cover, slope, etc.) and the collected data from each area were classified into three groups depending on UAV acquisition geometry, i.e., nadir imagery, oblique imagery, and an integration of nadir and oblique imagery. High-resolution orthophotos and Digital Surface Models (DSMs) emerged from the processing of the UAV imagery of each group through structure-from-motion photogrammetry (SfM). Accuracy assessment was carried out using quantitative and qualitative comparative approaches, such as root mean square error calculation, length comparison, and mean center estimation. The evaluation of the results revealed that there is a strong relationship between UAV acquisition geometry and landslide characteristics, which is evident in the accuracy of the generated photogrammetric products (orthophotos, DSMs). In addition, it was proved that the synergistic processing of nadir and oblique imagery increased overall centimeter accuracy.

Solar Energy Modeling and Mapping for the Sustainable Campus at Monash University

An increase in energy demands and positive public acceptance of clean energy resources have contributed to a growing need for using solar energy in cities. Solar photovoltaic (PV) deployment relies on suitable locations with high solar energy potential. In the urban context, building rooftops are often considered one of the most available locations for solar PV installation. This work demonstrates a new geospatial-method for spatiotemporal modeling and mapping solar energy potential based on a high-resolution (0.2 m) digital surface model (DSM) and solar radiation dataset. The proposed method identifies building rooftops with a high solar energy potential by using the Solar Analyst (SA) model. The results show that 93.5% of the rooftop area has high solar energy potential in the study area. The annual averaged sum of solar irradiation values is estimated to be 1.36 MWh/m2. In addition, the study showed that sloped rooftops facing to the north received up to 30% more incoming solar radiation than other rooftops with different geometry and orientation. The results are validated using recorded energy output data from four existing solar PV systems in the study area. The return on the initial investment of PV systems installation is estimated to be from four to five years.

Village-level poverty identification using machine learning, high-resolution images, and geospatial data

Tracking progress in poverty alleviation and promptly identifying the distribution of poor areas are critical for strategic policy interventions, especially for regions with poor statistical systems. The massive satellite imagery and geospatial data provide great opportunities for timely and cost-effective socioeconomic evaluations. However, existing research on poverty identification is mostly based on satellite images, and the potential of combined multi-source geospatial data on poverty identification has not been fully explored. Here, we propose an approach that evaluates how village-level poverty can be identified by integrating high-resolution imagery (HRI), point-of-interest (POI), OpenStreetMap (OSM), and digital surface model (DSM) data. The study area included 338 villages from Yunyang County, located in Hubei Province, central China. We extracted the explanatory variables indicating access to facilities and services, agricultural production conditions, village construction, and the spatial distribution of village settlements from the HRI, POI, OSM, and DSM data. The random forest algorithm was then used to model the relationship between village-level poverty and explanatory variables. The results demonstrated a 54% accuracy in the prediction of village-level poverty; the best prediction performance (72%) was observed for the villages categorized as poor. The built-up land proportion and the time cost to the facilities and services contributed the most to the identification of village-level poverty, while the proxy variables of agricultural production conditions contributed the least. This study provides an approach to village-level poverty identification using satellite imagery and geospatial data and proves that the data employed in this study could identify the poorest areas that are highly coupled with natural geographical conditions and backward public services.

Efficient dead pine tree detecting method in the Forest damaged by pine wood nematode (Bursaphelenchus xylophilus) through utilizing unmanned aerial vehicles and deep learning-based object detection techniques

Pine wood nematode (Bursaphelenchus xylophilus) is an invasive pathogen in South Korea, where it has caused pine wilt disease (PWD) with extremely high mortality of native pine species (Pinus densiflora, Pinus thunbergii, and Pinus koraiensis). Since the disease spreads by its vectors, native pine sawyer beetles (Monochamus alternatus and Monochamus saltuarius), the cost of monitoring the expansion has been rapidly increasing. Furthermore, it is even more costly to eliminate new and isolated infections since unremoved infected trees act as new sources of infection through the preferred oviposition of the beetles on such trees. The methodology of combining unmanned aerial vehicle (UAV) and object detection based on deep learning provides the opportunity to solve such problems, as UAV with RGB camera can provide high spatial resolution aerial image and digital surface model (DSM), which can be used for object detection with excellent results. In this study, we evaluated the performance of this method to detect dead pine trees in PWD-damaged areas. In particular, to ensure low omission error of monitoring, YOLOv3 was employed for object detection as the model design is focused on minimizing the omission error. We also modified the model so that the positions and crown diameter could be estimated. Four detection models were trained using four different combinations between aerial images (R, G, B) and DSM from UAV. Among them, the model from RGB showed the highest performance (recall: 0.9909, precision: 0.8438) and was selected as the optimal model. Our results suggest that our method can contribute to low-cost and effective monitoring of the dead pine trees while maintaining low omission error, which is critical for PWD management.

Deep Learning for Strawberry Canopy Delineation and Biomass Prediction from High-Resolution Images.

Modeling plant canopy biophysical parameters at the individual plant level remains a major challenge. This study presents a workflow for automatic strawberry canopy delineation and biomass prediction from high-resolution images using deep neural networks. High-resolution (5 mm) RGB orthoimages, near-infrared (NIR) orthoimages, and Digital Surface Models (DSM), which were generated by Structure from Motion (SfM), were utilized in this study. Mask R-CNN was applied to the orthoimages of two band combinations (RGB and RGB-NIR) to identify and delineate strawberry plant canopies. The average detection precision rate and recall rate were 97.28% and 99.71% for RGB images and 99.13% and 99.54% for RGB-NIR images, and the mean intersection over union (mIoU) rates for instance segmentation were 98.32% and 98.45% for RGB and RGB-NIR images, respectively. Based on the center of the canopy mask, we imported the cropped RGB, NIR, DSM, and mask images of individual plants to vanilla deep regression models to model canopy leaf area and dry biomass. Two networks (VGG-16 and ResNet-50) were used as the backbone architecture for feature map extraction. The R 2 values of dry biomass models were about 0.76 and 0.79 for the VGG-16 and ResNet-50 networks, respectively. Similarly, the R 2 values of leaf area were 0.82 and 0.84, respectively. The RMSE values were approximately 8.31 and 8.73 g for dry biomass analyzed using the VGG-16 and ResNet-50 networks, respectively. Leaf area RMSE was 0.05 m2 for both networks. This work demonstrates the feasibility of deep learning networks in individual strawberry plant extraction and biomass estimation.

Assessment of Storm Surge History as Recorded by Driftwood in the Mackenzie Delta and Tuktoyaktuk Coastlands, Arctic Canada

The southern Beaufort coastline in Canada experiences significant storm surge events that are thought to play an important role in coastal erosion and influence permafrost dynamics. Unfortunately, many of these events have not been documented with tide gauge records. In this paper, we evaluate coastal driftwood accumulations as a proxy for estimating maximum storm surge heights and the history of these events. We use historical air photos and data derived from Unoccupied Aerial Vehicle (UAV) imagery to resurvey four coastal stranded driftwood study sites that were first appraised in 1985–86 and assess two new regional sites in the Mackenzie Delta. Maximum storm surge heights were found to be similar to observations carried out in the 1980s, however, we refine the elevations with more accuracy and reference these to a vertical datum appropriate for incorporating into sea level hazard assessments. Detailed mapping, historical air photo comparisons and the UAV acquired imagery at a site close to Tuktoyaktuk demonstrate that the highest storm surge at this site (1.98 m CGVD2013) occurred in association with a severe storm in 1970. This event shifted driftwood and floated material slightly upslope from an older event thought to occur in 1944 that reached 1.85 m (CGVD2013) elevation. The quality and accuracy of the high-resolution Digital Surface Model (DSM) and orthophoto derived from Structure from Motion (SfM) processing of the UAV photographs allowed mapping of four distinct stratigraphic units within the driftwood piles. Based on variations in anthropogenic debris composition, weathering characteristics and history of movement on aerial photographs, we conclude that no storm surge events at Tuktoyaktuk have exceeded ∼1.3 m (CGVD2013) since 1970. While there has been some speculation that ongoing climate change may lead to more frequent large magnitude storm surges along the Beaufort coast, our study and available tide gauge measurements, suggest that while moderate elevation storm surges may be more frequent in the past several decades, they have not approached the magnitude of the 1970 event.