<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>This project was carried out on <a href="https://www.upwork.com/jobs/~01691cf02b909ebef0" target="_blank" rel="noopener">Upwork</a> for a private client working in infrastructure development, requesting a reference interactive map of UK gas pipelines, related infrastructure and geographical context. The cartographic requirements included that the map should be legible to non GIS experts and include some client provided data layers.</p>
<p>After consulting the client on available options, I delivered a QGIS project map with smart layer grouping for different use cases, and scale dependent symbology. The map included an open vector basemap by Ordnance Survey, OpenStreetMap data, raster overview map, administrative and geographical divisions, Google satellite imagery as an XYZ tiles service and land property boundaries in WMS by the land registry. Data preprocessing included reprojections, styling and clipping to the desired geographical extend. Data descriptions, source, date and copyright information was embedded in the project in layer metadata.</p>
<p>The thematic level included UK gas pipelines, petrol stations (extracted from OpenStreetMap), fire departments, gas processing facilities, agricultural and waste plants, as well as traffic information. Much of the data was put together for the project from open sources including the governmental data portal. The map became a very good visualisation tool serving different contextual information according to scale, and was met with very positive feedback by the client. In total it took under two work days to prepare this map.</p>
<blockquote>
<p style="font-style: italic;" class="mb-0">An excellent experience working with Nick - very responsive; communicated clearly; suggested good ideas and worked with my imperfect brief to deliver a high quality mapped product. Highly recommended!</p>
<p style="font-style: italic;" class="mb-0">Michael Shaw</p>
</blockquote>
</div>

Country-wide infrastructure visualisation

Infrastructure map

<p><span style="font-weight: 400;">Together with tourism specialist Simone Vaccari we are developing a new exploration tool for tourists around Europe.</span></p>
<p><span style="font-weight: 400;">Ptolemy will be a traveler application focused on the cultural, historical and artistic heritage of destinations. By taking advantage of modern web technology and open data, we are aiming to offer a service to people who seek deeper travel experiences.</span></p>
<h2>Go to <a href="https://ptolemy.app" rel="follow noopener" target="_blank">https://ptolemy.app</a> to learn more!</h2>

Ptolemy

<p>East European hiking trails are usually colour-coded according to trail importance, e.g. blue for long distance, red for intermediate, yellow for local etc. What's more, each trail usually has a special icon that characterizes it, shown prominently on signs on the ground. Due to popular demand by users at Outdooractive, who were used to regional web maps showing coloured trail lines separately, I was tasked with making a style amendment to our system to accommodate the use case.</p>
<p>I developed a routine using SQL Postgres preprocessing and Mapbox GL-JS layers to take the existing solutions one step further. The original OpenStreetMap hiking layer was cleaned, aggregated and ordered according to colour and trail type properties. Different front-end layers combine to create the final metro-map style, where colours overlap and automatically buffer, improving clarity. The width of the highlights changes according to trail type and rank on each hiking segment.</p>
<p>The icons were treated in a similar aggregating manner, but using Mapbox's late formatted string expression to show icon clusters with or without labels according to zoom level.</p>
<table style="border-collapse: collapse; width: 99.2278%; height: 29px; background-color: #fbeeb8;" cellpadding="10" border="1">
<tbody>
<tr style="height: 12px;">
<td style="width: 99.5553%; height: 12px;"><span style="color: #000000;">DISCLAIMER</span></td>
</tr>
<tr style="height: 17px;">
<td style="width: 99.5553%; height: 17px;"><span style="color: #000000;">This work was based on infrastructure provided by the cartography team at Outdooractive, including icons and basemap.</span></td>
</tr>
</tbody>
</table>

Print like trail visualizations for the web

cartography

Metro-style hiking trails

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>I was hired by UK location intelligence geocoding provider <a href="https://addresscloud.com/" target="_blank" rel="noopener">Addresscloud</a> to enhance their platform's print report functionality. The company's web app consists of a map-based panel with a list of selected property information on exposure to different types of risk and related metrics.</p>
<p>I developed a javascript utility that generates a printable A4 PDF report of the current map view and related fetched statistics into the layout requested by the client. The code resizes the map canvas into the calculated print resolution size, and builds the PDF layout using jsPDF client-side, importing data like user email from application state.</p>
<p>The original code was optimised with 200kb reduction in bundle size and the user experience was enhanced, including a buffering modal while the generation works, and a responsive two column consistent visual design in the final print.</p>
</div>

Print map reports

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>This project responded to the need for the automatic generation of minimaps showing the service coverage of outdooractive's raster topo layer from different providers. It is a process that was carried out manually with any changes to the status quo, with no standardised style or underlying data.</p>
<p>I created a new QGIS project drawing basemap data from suitable custom-processed Natural Earth and OpenStreetMap tables in a shared PostGIS database. The new layer styles made sure to follow the visual identity of our primary hiking map, prioritising information shown, adding graphic features unavailable to webmapping, tuning down the basemap intensity and highlighting the coverage polygons. The polygons come from a new coverage layer with one feature per provider or map type, that serves as the reference layer for a map atlas cycling though the features and generating predesigned print layouts for every feature.</p>
<p>To account for the reduction in scale, because the coverage polygons are very detailed and lead to artifacts when rendered densely, a PostGIS materialized view was created, simplifying, merging and inflating polygons where needed to aid visualisation. This processing happens on the database server automatically whenever anyone changes the base polygon layer. The print layouts and map symbology were designed for three different image sizes to be used in the outdooractive web knowledge pages.</p>
<p>The overall time to refresh all minimaps dropped from days to a matter of minutes (essentially manipulating the polygons and exporting the print atlas layouts) and the results are currently being introduced into the platform.</p>
</div>

Automatic minimaps

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>While researching webmap vector tile styling and rendering I decided to create a webmap printing service called <a href="http://www.maptakeout.com" target="_blank" rel="noopener">MapTakeout</a> that would offer professionally designed maps for home printing, taking advantage of the amazing Mapbox GL JS and Vector Style Specification. I decided to approach the problem by exporting the webmap canvas and its vector data into PDF layouts (using the jsPDF library) and PNG. This involved computations using window DPI, canvas size, map latitude, and tile size.</p>
<p>As for the question "Why print a webmap?", the specific approach leverages the speed and optimised delivery of global vector tiles to create fast printable output, even though webmap renderers are not at the level of a full-fledged GIS as of 2021. (And they may never be - the question is whether they are sufficient for the desired cartography.) Another problem is that Mapbox GL JS does not yet support projections other than Web Mercator, which is bad for printing.</p>
<p>Seeing that the primary aim was to promote my cartography and experiment with automatic map generation, the current website is a good start. I have, however, observed cartographic limitations with Mapbox vector tiles. This made me consider hosting an own geodatabase online focused on visualisation, bringing together different open sources and preprocessing to push towards better cartography. I am also considering ommiting front-end APIs all together and rendering using a cloud GIS solution.</p>
<p>The website itself was created using Next.js and the Material UI framework. No back-end development has been necessary so far. The great geocoding service Photon by Komoot has been used to power the location search bar and the approximate map title in the exported PDF.</p>
</div>

MapTakeout

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>Visual artist Marina Maniadaki is working on her project <a href="https://www.recreatingtheevidence.co.uk/Lost-Relative-Trail" target="_blank" rel="noopener">Lost Relative Trail</a>, for which she hid ten old family pictures in niches around Venice, and then observed what they meant to the people who found them. She will be registering this experience into a new booklet, and wanted to accompany her descriptions and letters by some hand drawn illustrations of the lost hiding spots in the form of a map.</p>
<p>After settling on the area of interest and map style, we decided that the best format to support the drawing would be fixed backdrops for Marina's lightbox device, providing a geographical base as well as some contextual information from the city. I made four A1 backdrops of the same area out of OpenStreetMap data and Google Satellite Imagery, showing points of interest to the project, a bird's eye view and a blue-red wireframe (both 2D and 3D) to support pencil drawing.</p>
</div>

Lost relative trail

Tourism regions

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p><a href="https://memonomenaperistatika.gr" target="_blank" rel="noopener">Memonomena Peristatika</a> is a website that sprouted as a reaction to a surge of police brutality incidences in Greece in early 2021. It was made to provide a central point of reference for video evidence recorded by victims and bystanders.</p>
<p>I contributed to this Next.js platform on Github by providing a webmap to visualize the long list of reported locations. Because the locations were collected in descriptive form in a spreadsheet, I adapted a Python script for batch geocoding of addresses using the Google Geocoding API by Shane Lynn. I added application-specific configuration globals for relevant spreadsheet column names, filters for the original entries csv (eg. approved), and region bias for geocoding. The script returns the incidences spreadsheet with location coordinates attached.</p>
<p>The map was provided as a react component with a dynamic wrapper to bypass server side rendering. On mount a Leaflet map instance is loaded with the default OpenStreetMap tiles, pulls the enities geojson and creates clustered markers with descriptive popups that link to the relevant video material. The component exposes props for styling its container and manipullating the map object once it is loaded. I added an optional Mapbox Light style which works on input of a valid API-key. The pull-request can be found <a href="https://github.com/balourdos/MemonomenaPeristatika.gr/pull/53" target="_blank" rel="noopener">here</a>.</p>
</div>

Map against violence

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>In the process of working to optimise the QGIS cartographic templates over at Outdooractive, I have been developing custom geometry manipulation functions to aid better automatic map labelling. Inflexibility in the QGIS labelling engine when it comes to complex data-dependent symbology and interactions between layers means sometimes the best available solution is to just export all the labels and do the work manually. Looking for alternatives with the aim to minimise post-processing in Illustrator, I tried bypassing the default labelling QGIS engine options. Instead, I opted for resolving difficult label collision situations using Python in the QGIS geometry generator. Here are two example layers where this was used.</p>
<p>In mountain peak symbology, labels and symbols are inseparable and the priority depends on the prominence of mountains. At the same time, when mountains are isolated, they are usually shown irrespective of height. This makes it hard to use both symbology and labelling interdependently, as well as setting more complex priority criteria. This is why I went for a pure symbology approach that creates a buffer (paper units at scale) around each mountain and tests for intersections with surrounding features. On collision, the lower mountain is hidden. This leads to local chains the end at the tallest mountain that remains visible. This function is run at each render, so the prioritisation is completely responsive. Labels are rendered as Font Markers in the layer symbology, so they act concurrently with the mountain peak symbols.</p>
<p>Another interesting case is contours. As of early 2021, QGIS had no capability to label contours in sensible cartographic ladders (like is found in Maplex). This is a fairly complex computational problem because contour features often do not share as much as it looks, and it becomes hard to assosciate labelling lines with them. This is a good use case for using the geometry generator to transform features before they are labelled. My function uses the current map view to create four global focal points. Each contour linestring is cropped to four segments that are closest to these points. This forms candidate labelling positions for each contour that preserve the original linestring orientation and align nicely with each other. When two segments intersect, they merge to avoid double labels. The images show a default QGIS label placement versus the function results (cropped segments in light green). This is not the perfect laddered labelling solution, but it takes a lot of weight off the cartographer's shoulders.</p>
</div>

Computed symbology

<p>I produced maps for print products of public client organizations around Germany, using the proprietary Outdooractive style and data. This included clipping our data to the ordered map sheets, exporting primary maps series in PDF and postprocessing content, labels and marginalia in Adobe Illustrator. Elevation profiles were also prepared using automated Illustrator actions.&nbsp;</p>
<p>Parallel to the various projects I carried out a complete remake of our master QGIS project file, simplifying the symbology, flattening layers and taking advantage of the full software capabilities. My changes led to a 50% decrease in file size and a much easier to handle style file.</p>
<table style="border-collapse: collapse; width: 99.2278%; height: 29px; background-color: #fbeeb8;" cellpadding="10" border="1">
<tbody>
<tr style="height: 12px;">
<td style="width: 99.5553%; height: 12px;"><span style="color: #000000;">DISCLAIMER</span></td>
</tr>
<tr style="height: 17px;">
<td style="width: 99.5553%; height: 17px;"><span style="color: #000000;">This work was based on infrastructure provided by the cartography team at Outdooractive, including icons and basemap.</span></td>
</tr>
</tbody>
</table>

Activity maps

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>Cartographers and GIS professionals often require custom routines for project-specific work or pre-processing that supports bigger problems. I have been tasked to develop several in-house tools for the Cartography team at Outdooractive, that everyone accesses directly within their QGIS. The scripts live in an online repository so that updates are installed in just one pull.</p>
<h2 class="text-h4 my-5">Topology</h2>
<p>I developed an optimised topology checker for dangling line geometries in trail networks, similar to the native QGIS 'Topology Checker'. The advantage was that dangle points can be exported into permanent files and manipulated accordingly during checking procedures. Additional attributes were added to the geometries to integrate better with other in-house tools. Another checker tool was also created to distinguish between parent trails and export separate dangle files for each. These tools are crucial for fast network checking in large areas, jumping from point to point to inspect.</p>
<h2 class="text-h4 my-5">Data Cleaning</h2>
<p>Apart from custom scripts, I have routinely created QGIS models to get through repetitive tasks. A notable example is cleaning new data topologies from duplicate segments to avoid errors in later processing. This was achieved by running a model in 'Per Feature' mode, making a spatial comparison and extracting clean geometries, before merging all resulting layers back together.</p>
<h2 class="text-h4 my-5">Symbology</h2>
<p>An important way to improve a cartographer's everyday work is to provide custom tools to better and faster visualise data. A routine problem we have at Outdooractive is understanding which trail codes run through which segments of our line network. This usually required setting a new time-consuming filter query in the network layer and visualising by labels and color. I made a geoprocessing tool to provide a fast way to inspect available trails in the layer by manipulating symbology. The tool finds all unique trail codes in the input attribute and generates a new symbology rule based on a filtering query for that code. After execution, the network layer can be symbolised to show only one of the available trails, just by checking that symbol rule in the layers panel. Specific styles can be copied between rules in layer properties.</p>
</div>

Productivity GIS tools

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>Integrating new data from trusted sources in bulk presents the problem of manually checking for and resolving local differences and updates. This is the case for my cartography team at Outdooractive where a network of activity trails (e.g. hiking, cycling) is aggregated from diverse proprietary sources and requires regular maintenance, resulting in heavy repetitive workloads. I created a pyqgis geoprocessing tool that automates name checks and geometry comparisons of new trails against existing ones, dramatically reducing the time needed to adopt changes.</p>
<p>Trail attributes are passed into pandas dataframes for code minimisation. Names undergo a fuzzy string comparison using the python SequenceMatcher and the diverging segments of matching new and old trails are found. Unmatched trails are efficiently compared against the existing network to return potential match suggestions based on geometry. The tool outputs all results in a structured directory of geopackage files as well as a QGIS project file with semantic groups, layer names with metrics and appropriate symbology for further editorial work.</p>
<p>This tool resulted in real workflow improvement, being versatile (custom inputs for encoding, comparison and geometry parameters with recommended defaults) and runnable from inside the Processing Toolbox of QGIS with a user friendly documentation dialog linking to further guidance in a wikipage online.</p>
</div>

Line network processing

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>The insight GIS can unlock for client-targeting companies is important, and in this project I researched available open data sources and statistics to deliver value for the Glasgow branch of a security alarm provider. The desirable outcomes were pinpointed in consultation with the branch manager of the client, and were readjusted at a later stage in the project.</p>
<p>The result was an algorithm to rank statistical areas of Glasgow according to selling potential, taking into account data by the Ordnance Survey open data layers, Open Street Map thematic information, the Scottish Index of Multiple Deprivation and an API for calculating time travel zones for police intervention. Main pillars of the metrics included income, crime, and business location density.</p>
<p>The ranked areas served as backdrop to a map of the city and targetable businesses. Major business groups were found using a clustering algorithm and also ranked by their surrounding area average. The map pinpoints businesses of interest, isolated businesses, and police stations, including their closest postcode. Existing company installations and client were incorporated.</p>
<p>The final product was printed in A0 and was used for day-to-day cold-knocking strategy, and market overview. It received positive feedback for providing branch staff with a clear image of their target area. GIS analysis was carried out in QGIS, with map post-processing in Inkscape.</p>
</div>

Evaluating selling potential using open data

Business strategy map

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>This was a project for a professional running a one-person bakery business delivering her delicious goods to cafes, bakeries and restaurants accross Glasgow on demand. I created a very focused, simple webmap providing an overview of the location of hospitality establishments in her area, as well as her HQ and live location. The map was made using Mapbox GL JS, using custom icons and visibility preferences and was served locally.</p>
</div>

Web map for a baker

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>This thesis was completed towards an MSc in Geoinformation Technology &amp; Cartography at the University of Glasgow in 2019 and is about improving urban monitoring towards representation of cities as living and breathing entities. It was inspired by the difficulties pedestrians are faced with in Athens and other big cities as they go about their everyday lives.</p>
<p>This is the project presentation at the <a href="https://sites.google.com/view/uep2020-conference/home" target="_blank" rel="noopener">5th International Conference on Urban e-Planning</a> and project abstract. The project received the best thesis award for the class.</p>
<div style="padding: 56.25% 0 0 0; position: relative;"><iframe style="position: absolute; top: 0; left: 0; width: 100%; height: 100%;" src="https://player.vimeo.com/video/601045641?h=26409e461d&amp;badge=0&amp;autopause=0&amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen="allowfullscreen" title="Nick Lempesis - Mapping urban problems using remote imagery" data-mce-fragment="1"></iframe></div>
<p>
<script src="https://player.vimeo.com/api/player.js"></script>
</p>
<h2 style="margin: 20px 0;">Abstract</h2>
<p>A good urban quality of life, including urban mobility and public space walkability, is becoming harder to attain in growing contemporary cities facing diverse and complex problems. Pedestrians globally face obstructions, pollution and risk of accident. There are numerous demonstrable benefits in improving the effectiveness of urban management, which is widely under budgetary pressure and often ends up reactive and inconsistent. New standardised urban monitoring methods are needed with a focus on sustainable planning.</p>
<p>This project shows that the combination of automatic object detection in remotely sensed imagery with existing geospatial data can support the solution of urban problems. Remotely sensed imagery provides prompt, uniform data with city-wide coverage, while the recent advances in computer vision enable fast and accessible automatic interpretation, and established cartographic methods can generate and visualise analytical insight. Urban detection promotes cost effective, evidence-based, transparent and participatory planning, and contributes to the United Nations 2030 Agenda goals.</p>
<p>In this proof of concept a workflow was developed for deriving insight on the restriction of pavement accessibility by inconsiderately parked vehicles in target areas. The semi-automatic production was achieved of large scale urban map reports of 200 by 200 m areas, featuring imagery-derived car detections, urban topography and useful geospatial metrics. The maps are usable by urban planners and field surveyors.</p>
<p>Two examples were provided for Paisley and Birmingham. Complementary analysis involved the calculation of pavement &lsquo;load&rsquo;, i.e. overlap with adjacent car detections, and its temporal change. The presented workflow uses orthorectified aerial imagery of 0.25 m spatial resolution and can be replicated using open software. The detection algorithm achieved a workable F-score of 70.72% but could not be adapted for processing 0.46 m satellite imagery. Detection accuracy was found to depend on input image histogram and deteriorate rapidly with loss in image quality. Standard graphics equipment would render a two miles wide urban area in under one hour.</p>
<p>Further research should examine the determinants of object detection accuracy in remotely sensed imagery, and assess other opportunities for cartographic insight. Future implementation should focus on satellite detection, addressing temporal inflexibility and the quantification of urban mobility via walkability indices.</p>
</div>

urban

Mapping urban problems

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>As a topographic team at the University of Glasgow we used a DJI Phantom 4 UAV and the Pix4D photogrammetric software to perform image acquisition and create a densified Digital Surface Model point cloud of the university sports grounds using Structure from Motion.</p>
<p>Ground truth was captured at custom made high visibility targets using a Real Time Kinematic GNSS Leica antenna. The broader aim was to evaluate and report on achievable data quality and outline implications of the method, but the accuracy achieved for the time taken by the survey was impressive.</p>
</div>

Structure from motion

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>This A3 storytelling map of Europe was the result of an exercise in cartographic presentation. I created a journalistic style layout backed by statistical data and press research, dealing with a big contemporary topic, the refugee crisis resulting from of the Syrian Civil War. I chose this topic as exemplary of the unique clarity of perspective that maps can bring about in dividing societal issues.</p>
<p>The design was meant as an overview of migration that does not isolate the phenomenon from its socio-political context or present it merely in one level. Migrant routes are not symbolised by aggressive arrows, but shown through refugee camp density and hosted populations. A single family journey case study serves to decouple migration from statistical abstraction.</p>
<p>The map features manual shaded relief to avoid cluttering, blending modes, inset information at points of interest, and a double graduated circle/colour symbology to detect "openess" to refugees. The project was completed using QGIS and Inkscape, on a tight deadline in less than a week.</p>
</div>

A storytelling map focused at the Syrian refugee crisis

Immigration map

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>This project called for the creation of a geographical database and associated output templates for use in administration of a rural crofting estate in Drumbuie, Kyle, NW Scotland. The target use consisted of collecting and distributing rent, managing land tax, maintaining a record of occupied land, managing unoccupied land, delivering services to owners and tenants, and solving land related disputes.</p>
<p>I created a database schema including parcels, owners, tenants and tenancy entities. Local topographic data was obtained from the Ordnance Survey MasterMap series and symbolised appropriately. Some preprocessing was necessary to derive a close approximation to actual land parcels.</p>
<p>A conceptual, logical and physical model and documentation accompanied the database. The implementation used the ArcGIS suite (ArcMap, ArcCatalog) with the Related Tables function revealing linked objects.</p>
</div>

Land plot database

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>This project delivered an A3 format topographic map design at 1:100k scale presenting the main points of interest on the island of Arran in Scotland. The map is usable in brochure or poster form and is aimed at daily tourism. Basic cultural and physical layers were manually generalised from the Ordnance Survey Strategi dataset, with some additional information vectorised from the 1:250k OS raster map.</p>
<p>The colour scheme attempts a reference to the winter island landscape, with warm coloured cultural features in contrast. Hypsometric layers desaturate past 500m, distinguishing mountainous areas and giving a purple note, also faintly visible in the real landscape. A light texture has been used in woodland.</p>
<p>A supporting map specification on colours and data sources came with the map. Data manipulation and prioritisation was done in QGIS and postprocessing, layout, icon design, typography and colour balancing were carried out in Inkscape.</p>
</div>

Touristic map design for a Scottish island

Arran topography

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>I performed a viewshed analysis for a proposed windfarm development in rural Wales, using local topographic maps and the ArcMap visibility tools. Following research on windfarm construction and related environmental issues and effects on the community, I created a poster for a hearing at the local council promoting the project.</p>
<p>The design legibly communicates zones of theoretical visibility for different parts of the turbines, as well as 3D landscape reconstruction scenes made using ArcScene.</p>
<p>A video presentation of the proposal accompanied the poster. The video was created using true 3D graphics and camera animations in Blender with the BlenderGIS addon, with editing done in NCH VideoPad. Several technical difficulties were faced in the process but the end result was adequately realistic.</p>
<div style="padding: 56.25% 0 0 0; position: relative;"><iframe style="position: absolute; top: 0; left: 0; width: 100%; height: 100%;" src="https://player.vimeo.com/video/691858258?h=d6e7c339e0&amp;badge=0&amp;autopause=0&amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen="allowfullscreen" title="Visual Impact Modelling" data-mce-fragment="1"></iframe></div>
<p>
<script src="https://player.vimeo.com/api/player.js"></script>
</p>
</div>

Windfarm visual impact

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>As part of a geomatics exercise aiming at familiarisation with computational transformations of coordinates, a set of digitiser coordinates of the Madagascar coastline, captured from a scanned map, was computed into projected coordinates in the Mercator and Lambert Conformal Conic (LCC) projections, in Python.</p>
<p>The process was carried out for the minimum solution using two ground control points, and using a Least Squares adjustment for four control points. A map report created.</p>
</div>

Coordinate transformations

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>Cartographic design was a big part of my MSc in Geoinformation Technology &amp; Cartography at the University of Glasgow. Several exercises and small projects on layout, label placement, symbology, generalisation, statistical representation, and many hours of personal experimentation helped me build a cartographic skill set.</p>
</div>

Cartographic process

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>This project was a multi-criteria site evaluation on the Greek islands of Milos and Tinos, in view of recent discussions on windfarm development increasing local energy autonomy and reliability.</p>
<p>The analysis and cartography were carried out in ArcMap. A vector based approach was taken with discrete criteria like minimum wind speed, while Fuzzy Suitability was used to combine continuous criteria like visual impact in raster format. The workflow sequence was automated into a ModelBuilder geoprocessing tool.</p>
<p>The map features a multi-directional hillshade made using Terrain Tools, mosaiced with the thematic layer to avoid transparency colour muting. A simple corresponding web map on ArcGIS online was also made.</p>
</div>

Windfarm siting algorithm

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>I have carried out satellite imagery classification using a number of methods and software. For Coursera's GIS Specialisation, I classified Landsat 8 images of central Chile from 2014 using false-colour composites to find the extent of wildfires the country was fighting at the time.</p>
<p>This was an unsupervised classification in ArcMap. In my MSc course on remote sensing I calculated the seasonal shrinkage of Lake Chilwa in Malawi using ERDAS Imagine and European Space Agency's SNAP software. The classification used a Random Forest algorithm after a comparative literature review on classification options. I tested for accuracy and produced an error matrix for manual samples.</p>
</div>

Image classifications

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>My MEng degree introduced me to technical thinking and involved numerous realistic projects in subjects including urban space repurposing, building refurbishment, structural design, sustainability, engineering application programming.</p>
<p>What I consider really valuable is the repeated friction we had with the application of standards, structured problem-solving and the search for widely adoptable practices.</p>
</div>

other

Technical projects

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>This dissertation for MEng Civil Engineering at the University of Bristol included a collaborative part worked on by a team of four and an individual technical annex shown as a separate project. As our supervisor was active in geotechnical studies in the Caribbean at the time, we were made aware of increasing housing safety hazards in the suburban slopes of the island nation of Saint Lucia. Regular landslides are observed mainly affecting the informal settlement that results due to hurried, unregulated urban sprawl and rapid economic migration.</p>
<p>We set out to create the basis for a disaster-resilient, cost-effective, modular, and culturally relevant single-household house design for use in accessible governmental social-housing schemes. The result was a robust and configurable timber frame unit inspired by beehive structures, resting on deep bored concrete pile foundations.</p>
<p>The design attempted to minimise vulnerability to seismic and hurricane hazards, as well as exposure to landslides. Reliance on network utilities was controlled, with the structure being environmentally friendly and economically viable at GBP 40 thousand and an 11 day construction time.</p>
<p>We made use of Finite Element Analysis (MIDAS) and European construction standards for structural design, a translational slope stability model for geotechnics, as well as the SketchUp Vray engine for realistic 3D visualisation. We also provided a flowchart method for geotechnical engineers to assess potential slopes for safe development. The project received overall positive feedback.</p>
</div>

House design in St Lucia

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>This MEng Civil Engineering work utilised my dissertation's resulting house design as a unit basis in a communal neighbourhood plan. A relevant water supply network design was also provided to form a framework that can be used by the government of Saint Lucia to combat unregulated urban sprawl and substantially improve the living standards of underprivileged people. To be complete, this framework will have to include water sourcing and quality considerations, drainage and sewage plans, power distribution, maintenance procedures and inclusive community action and empowerment plans to keep the neighbourhood desirable to its residents.</p>
<p>For the urban plan I researched arrangements that minimise fast traffic and infrastructure requirements. A dense footway network emphasises social cohesion and pedestrian movement and minimises road construction in steep slopes.</p>
<p>For hydraulic design I used the EPANET software and provided adequate redundancy for infrastructure failure.</p>
</div>

Urban planning for Castries

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>This is a multi-stage project that drew from a real dam development scenario, carried out for my MEng at the University of Bristol in 2015. It involved a field trip to Powys, Wales and evaluation of potential dam sites, a collaborative dam design, and the writing of an advisory brief including a project, business and environmental strategy plan for permission approval from the National Assembly of Wales.</p>
<p>The design report included a site selection evaluation, aqueduct routing, structural and hydraulic designs and technical drawings, and an environmental impact asssessment. The advisory brief included a stakeholder analysis, value management strategy and risk assessment, procurement and financing plan, full construction costing and health and safety quality management.</p>
</div>

Dam design consultation

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>This Civil Engineering research project aimed at contributing to the improvement of individual mobility and standards of living in rural African communities in a sustainable and cost-effective manner.</p>
<p>Bamboo-framed bicycles present an opportunity to create a viable local industry that reduces dependency on Chinese and Indian imports and provides an accessible means of transportation. Bamboo is a robust structural material if properly used, native to sub-Saharan Africa.</p>
<p>The Engineers Without Borders at the University of Bristol had tested one such bike in Gambia in 2013 and found a weakness in how frame tubes were connected using raisin soaked hemp.</p>
<p>A literature review was carried out to identify opportunities for avoiding the current joinery technique. In a team of two we designed, built and tested (among many other configurations) a new joint made of locally sourced hardwood timber that could potentially offer a solution.</p>
<p>We tested our bamboo's material properties as well as the final joint performance under various loading scenarios. We used OasysGSA for Finite Element Analysis to find standard bicycle loads on tube connections.</p>
<p>The joint solution was judged successful in structural adequacy, serviceability, accessibility, constructability and cost-effectiveness but can be improved in terms of efficiency and wood selection.</p>
</div>

Joint design for bamboo bicycles

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">Over a one-week surveying field trip with the University of Bristol, I contributed to a team of three in the creation of a hand drawn hypsometric topographic map of a medium sized area in coastal south Wales, using standard survey equipment.</div>

Topographic survey in Wales

<div data-v-5fd2d85c="" class="text-body-1 mx-auto">
<p>I developed an interest in Smart City applications through an optional course at the University of Bristol, where I researched smart technologies and their role in service innovation to meet social, economic and environmental policy objectives. In one computational MATLAB project I analyzed and visualized energy use data from 4000 households to demonstrate behavioral patterns and explore how smart meters can reduce energy consumption and improve efficiency.</p>
<p>I am also interested in the promotion of pedestrian prioritization in urban environments. In an urban redesign of the main campus street and junctions, I suggested a shared use scheme where the street is paved and unmarked, leading to a coexistence of pedestrians with restricted vehicle users. Pavements were also reconfigured to facilitate preferred pedestrian routes. These interests led me to seek smart solutions to everyday urban problems in my MSc thesis.</p>
</div>

Nick Lempesis

Maps, GIS and geospatial web

Projects