MGISA Graduate's Research Reports

Alfred Cal

Flood modeling and impact analysis using LiDAR data: A Case study for Ladyville Village, Belize

Accurate and precise building extraction has become an essential requirement for various applications such as for impact analysis of flooding.  This project seeks to improve on current and past methods of building extraction by using the Principal Components Analysis (PCA) of LiDAR height (nDSM) and aerial photos (in four RGB and NIR bands) in an Object Based Image Classification (OBIA).  This approach uses a combination of aerial photos at 0.1 meter spatial resolution and LiDAR nDSM at 1-meter spatial resolution for precise and high-resolution building extraction.  Because aerial photos provide four bands in the PCA process, this potentially means that the resolution of the image is maintained and therefore building outlines can be extracted at a high resolution of 0.1 meter.  A total of five experiments was conducted using a combination of different LiDAR derivatives and aerial photos bands in a PCA analysis. The PCA of LiDAR nDSM and RGB and NIR bands combination has proved to produce the best result.  The results show a completeness of 87.644 %, and a correctness of 93.220 % of building extraction. This paper provides an improvement on the drawbacks of building extraction such as the extraction of small buildings and the smoothing with well-defined building outline.  

Jennifer Rolph

A GIS-based accuracy evaluation of a high resolution Unmanned Aerial Vehicle-derived Digital Terrain Mode

This project presents an evaluation of the accuracy and error distribution of a high resolution Unmanned Aerial Vehicle (UAV) derived Digital Terrain model (DTM). This exercise involved determining a flight plan to collect imagery from which to extract a georeferenced point cloud containing elevation values through digital photogrammetric processing, resulting in a Digital Surface Model (DSM). The DSM was refined to a DTM reflecting only bare earth values through an editing process of point classification. From the DTM, the surfaces of slope and aspect were derived and assessed using spatial statistics. The accuracy assessment of this spatial data presents a characterization of the error in the differences between the UAV derived elevation, slope and aspect surfaces, referenced to elevations of GNSS RTK Survey Points, as compared to a DTM of 2-meter resolution derived from digital photogrammetry based on large scale aerial survey (SCOOP Data, OMNR 2013). Statistical analysis included an examination of the distribution of error based on distribution plots, descriptive statistics and normalcy testing, reporting of Root Mean Square Error (RMSE), percentiles and confidence intervals, both with and without major outliers. Significance testing of results showed no significant difference in means or distributions between UAV and SCOOP derived DTMs, though the UAV DTM showed lower RMSE and percentile statistics at a higher frequency than the SCOOP DTM in stratified accuracy testing. Statistical analysis and accuracy reporting is stratified based on Non-Vegetated (NVA, Fundamental) and VVA (Vegetated, Supplementary) stratums, in adherence to ASPRS 2014 positional Accuracy Standards for Digital Geospatial Data. This study has shown how a lightweight UAV can contribute to terrain-based studies for small geographic extents, by collecting high density elevation data with good accuracy. The UAV DTM is represented by 0.25 m NVA accuracy at 95% confidence, and 0.10 m accuracy based on 95% RMSE, the level at which the data exhibits a Normal Distribution. In the VVA the accuracy equates to 0.08 m accuracy based on the 90th percentile statistic.

Tyler Pubben

The use of UAV collected imagery in conjunction with Structure from Motion photogrammetry to derive geologic structure information

This research project explores the possibility of using Structure from Motion photogrammetry in conjunction with UAVs and handheld camera systems to improve the quality of data collected during geologic field data collection. Structure from Motion photogrammetry was used to reconstruct georeferenced models of geologic outcrops near Exshaw Alberta from which structural information was successfully measured. The use of photogrammetric methods to generate this point cloud allows for the archival of the structural component of the outcrop as can be done using LiDAR, but also allows for the application of photographic texturing so that the outcrop may be revisited in virtual form. This research evaluates methods to help automate structural measurements using edge detection methods on both standard RGB digital photography and VNIR (Visible to Near InfraRed) to project marker points onto the point cloud. When compared to ground truth data, the measurements obtained through this method are within a 4.22° variance in strike or trend and 3.68° variance in dip or plunge. Of note is the fact that these accuracies are attainable using the camera system’s internal GPS and require no further ground control points. The Python modules developed during this research should allow rapid implementation of this system in a production environment. With further research and development, it may be possible to reliably automate the archiving of geologic outcrops as they exist at the time of data collection while simultaneously extracting detailed structural measurements that would otherwise be difficult or dangerous to obtain in the field.

George Barian

Development of a GIS application for modeling marine oil spills 

The purpose of this research is to develop an oil spill trajectory computer model, based on alternatives to existing comprehensive models, and create with this model an efficient, accurate and easy to use GIS application for modelling marine oil spill trajectories in Salish Sea. The first phase of the project involved analysis of the existing models and establishing performance and design criteria for the oil spill model. In the project’s second and third phases, the oil spill trajectory model and a computer application based on that model were developed. In the fourth and final phase, the developed model was validated, and results were assessed. The research applies a novel alternative approach to marine spill trajectory modelling, based on the Cellular Automata (CA) principle. To our knowledge, it is the first attempt to develop a CA based oil spill trajectory model and application that can be used for modelling real marine oil spills. The developed application may serve as a valuable tool for oil spill emergency responders and the general public in the Salish Sea area.

Aaron Wong

Evaluating the effectiveness of Marine Protected Areas (MPAs) in Belize

Marine Protected Area (MPA) has been a common management tool for conservation and protection of the marine environment from human impact in the recent years. Belize was one of the first countries to implement MPA in the management of marine ecosystem, and annual monitoring was conducted to report the status of the different MPA using a few indicators (coral cover, algal cover, herbivorous fish biomass, and commercial fish biomass). However, reports were not consistently generated to evaluate the implementation of MPA. This research aims to evaluate the effectiveness of marine protected area (MPA) in Belize by analyzing the various indicators mentioned above that are actively collected annually in various synoptic monitoring program (SMP) sites in Belize. Each of the indicators was analyzed to identify whether the indicators are at different levels in different zones of an MPA. Also, the change of each indicator (response ratio) was compared amongst the different zones to see whether the change would be higher in one zone than another. Lastly, the spatial relationship is examined between sites and zones to see whether sites with more favourable condition are concentrated in particular zones. Although no spatial relationship was found between sites and zones and no pattern was identified from the response ratio in different zones, this research found that conservation zone, in general, has a significantly more favorable condition than other zones. This research also demonstrated the importance of enforcement in MPA, when TAMR, an MPA that was not enforced prior to 2015, was excluded from the analysis, coral cover and commercial fish biomass was significantly higher in conservation zone than all other zones. This study can be used by the agencies in Belize to prioritize areas that are most in need of management actions, due to their unfavorable condition compared to other regions within the same MPA. Furthermore, this research will help in convincing more fishers that are opposing the establishment of MPA that MPA are beneficial to the environment and are a means to maintain the sustainability of resources.

Gregory Hay

Geoduck Density Modelling

The Pacific Geoduck, Panopea generosa (Gould, 1950), is a commercially harvested species that is naturally distributed in beds along the entire B.C. coast. A species distribution model (SDM) for the geoduck has been developed to aid and enhance current knowledge of their presence in unknown areas. The model is based on bathymetric variables and substrate. The Broken Group Marine Park located on the West coast of Vancouver Island in Barkley Sound was used as a control site, as no commercial harvesting has taken place. Recent scientific biomass surveys (2012-2014) undertaken throughout the park by industry and the DFO provided the biomass and substrate information. Bathymetry was provided by the CHS for use in the bottom mapping layers. Manipulation of the data layers into an export table for predictive modelling used ArcGIS, with the model being created in R. The model uses a general additive modelling (GAM) algorithm with a poisson distribution and evaluated with the cross-validation method. Three models with different data sets made up the study: a global model encompassing all the beds of the control group, and two local models, each using the data from one geoduck bed. An explained deviance (goodness of fit) of 33% was achieved for the global model, and 68.8% and 70.2% for the local models. The models can predict within a range of 0 - 40 but have difficulty with higher count accuracy. Overall the model creation was a success, and application to beds in the control area can aid in future surveys. It can also be applied to any bed coast wide, data permitting.

Cassandra Clouston

Evaluating Geographic Information System methods to enhance a park visit

Visitors to a park may not be aware of all the activities the park offers, due to the lack of signage or an informational kiosk. Having a system in place to allow visitors to become informed of the park’s assets will enable them to enjoy it more fully. With today’s location aware technology, an interactive and informational experience can be achieved to enhance the public’s knowledge of their surroundings. This research project evaluated Geographic Information Systems (GIS) technology as a basis for enhancing a visitor’s experience for park environments, using The Assiniboine Park English Garden located in Winnipeg, Manitoba as the grounds for this investigation. This research may set a precedent for other parks and green spaces to invest and adopt technology to create an informative and exploratory environment for their visitors. The methodology of this research was guided by multiple factors. The evaluation of existing GIS methods used in similar locations were reviewed to look at the capabilities of current technology: paper maps, QR Codes, Web Map Tours, Web Apps and Native Apps. A determination of how to use GIS to produce a database that would analyze not only the physical entities of the park like pathways and statues, but also the seasonal aspects were undertaken. A questionnaire was conducted with park visitors to gather an understanding of the visitors’ acceptance of the technology proposed. The results of the survey were used to evaluate which of the proposed technologies would be appropriate to implement and guide the prototype development and evaluation. The English Garden staff were supportive in providing their expertise about the existing methods at use in the garden and also as a source for data. This data included AutoCAD files, excel spreadsheets and hand drawn sketches. The data was converted and, along with newly collected data, was organized into a geodatabase for use in Esri ArcGIS software suite.

Orla Eileen Osborne

Foraging trip trends and delineation of foraging areas of Black-legged Kittiwakes (Rissa tridactyla) during the breeding season, on Middleton Island, Alaska

Many threats have led to the decline of seabird populations worldwide, Marine Protected Areas (MPAs) and Marine Spatial Planning (MSP) can protect important seabird foraging areas to assist in the recovery of declining seabird populations. Foraging areas are commonly delineated via GPS tracking data, often from small sample sizes limited to a single stage or year. However, for MPAs and MSP to be effective, we need to establish whether foraging locations collected during a single stage or year are representative of foraging locations across years and stages. This is especially important during periods of abrupt environmental change when seabirds may be most vulnerable. To address this issue, the effect of the North Pacific warming event of 2014 on foraging black-legged kittiwakes (Rissa tridactyla) was examined. Foraging trips before, during, and after the event were used to estimate foraging trip characteristics (trip distance and duration), the dispersal of foraging locations, and the location of foraging areas within and among years. Kittiwake foraging was more dispersed, direct and farther from the colony in years during and immediately after the warming event.  Over a third to over a half of the area used in 2013 was important in subsequent years which indicates that this area was, and may still be, a perennial hot spot.  These findings indicate that a multiyear dataset is necessary for the delineation of a protected area that covers all foraging areas, as they may change substantially from year to year, or a ‘persistent’ area that is used consistently for foraging from year to year. During the chick rearing stage kittiwakes increased their speed and reduced their resting.  The delineation of foraging areas and assessment of foraging effort from this study inform what areas need protection and when. This study not only improves our knowledge of the foraging behaviour of kittiwakes at Middleton Island but also informs data collection and data processing procedures.