GeoPython 2021 Schedule

The European Union's Copernicus Sentinel sensors produce large volume Earth Observation data streams, which are available under a full, free and open license. The Copernicus program also supports the establishment of Data and Information Access Services (DIAS) cloud-based processing solutions, some of which are federated in the European Open Science Cloud (EOSC). DIAS platforms closely couple the provision of compute resources with access to very large S3 object storage for data Sentinel archives, which include high resolution Sentinel-1 and -2 sensor data (10 m resolution), with high revisit (5-6 days) and continental coverage.

We demonstrate how we use a combination of geopython modules (GDAL, rasterio, geopandas) with PostgreSQL/Postgis spatial databases to manage the processing of deep time series data stacks with very large vector data sets that outline agricultural parcels in selected EU Member States. Accelerated processing is supported by integration of Numba and orchestration across multiple VMs on the cloud platform using customized Docker containers. Our client interfaces make use of Flask RESTful services and Jupyter Notebooks to support analytical tasks, which can include scipy based image analysis. Time series can also be integrated into machine learning frameworks like TensorFlow and PyTorch. We will demonstrate how our modular set up facilitates the use in monitoring tasks that are required in the Common Agricultural Policy context.

In the course of our presentation, we'll outline specific processing needs, and how we intend to integrate more advanced hardware solutions, such as GPU-based processing (in cupy, Numba), which is still surprisingly sparsely used in the geospatial domain. The relevance of our initiative in the context of European programs such as Destination Earth and European Data Spaces will be shortly addressed as well. Finally, we'll introduce the public github repository where we document our current and ongoing developments.

Konstantinos Anastasakis

Technology expert in the field of Earth Observation (Python, C++)

I am a Senior Scientist at the European Commission's Joint Research Centre (EC-JRC, Ispra, Italy) where I am the lead in the uptake of Copernicus data and processing solutions in agricultural monitoring applications, both in Europe and at global level. I am combining expertise in Earth Observation with applied programming solutions to advance the state-of-the-art. Python has become my preferred programming language, because it facilitates endless data processing solutions that are relevant in our application domain.

Elevation data is often not available for roads or tracks on bridges and inside tunnels. We present a method to interpolate elevation data inside complex bridge and tunnel networks from known elevation values at the access points. For the case of two access points, a simple linear interpolation is sufficient. However, many realistic bridge or tunnel networks (e.g. subway tracks or multi-lane tunnels) contain intersections, railway switches or more than two access points. For the general case, the task is formulated as an optimization problem which reduces to solving a system of linear equations for each network.

The method is implemented as a REST API using Python libraries such as NumPy, NetworkX and Flask. The API is capable of matching input GeoJSON LineString features to tunnel and bridge data from OpenStreetMap and decorating them with interpolated elevation data (for example from SRTM).

As an example of use, we demonstrate how the API integrates with our routing services to yield elevation profiles for trips in public transport. These elevation profiles could assist the planning of electromobility infrastructure such as the location of charging points.

• born 1984 in south west of Germany
• software developer at geOps GmbH, Freiburg, Germany since 2019
• PhD in Physics from University of Freiburg, Germany (numerical simulation)

Spatial data from a variety of sources are increasingly used to target marketing campaigns and prioritize rollout to an optimal audience. In this talk, we will demonstrate how different data sources (e.g. geosocial segments, internet searches, credit card data, demographics and point of interest data) can be blended and how spatial models can be used in identifying “demand hotspots” for Consumer Good Products (CPG). First, I will walk through a methodology on how to select target audiences in New York and Philadelphia for organic / natural products, based on spatial analysis of factors from the different datasets. I will then show a statistical analysis on how features for elimination purposes take place and a classifier is built to examine the impact of each factor on the selection of the “demand hotspots”. I will also present how the conclusions can be extrapolated for further locations, making use of a similarity score index which is based on probabilistic principal components analysis.

I am currently working as Spatial Data Scientist in Carto, where I focus on research and explorations of spatial datasets and techniques. Prior to that, I have been working with IRI as a research engineer, Coca Cola Hellenic Bottling as a Data Scientist, and Research Fellow at University of Surrey. I hold a PhD on Satellite communications from University of Surrey and Msc in Electrical and Computer engineering from National and Technical University of Athens.

A quadtree is a tree data structure in which each internal node has exactly four children. Quadtrees are most often used to partition a two-dimensional space by recursively subdividing it into four quadrants or regions. The regions may be square or rectangular, or may have arbitrary shapes.

This data structure provides a very convenient way to represent some area on the map. A map is divided to a set of tiles. Each of these tiles can also be further divided to four tiles of smaller size and this process can continue recursively until required precision is reached. A tile zoom level is a length of the path from the tile to the root of the quad tree. Each tile can be addressed by its unique key.

Mapquadlib supports different different tiling schemas and some convenient methods to work with them:

• Earth Core (Native) Tiling schema
• Mercator (Bing) Tiling schema
• MOS Quad Block Tiling schema
• HERE Tiling schema
• Nds Quads

We will show some technicalities of different Quadtree schemas and how to work with them using Mapquadlib.

Lead Software Engineer / Data Engineering @ HERE

In 2019, Topi Tjukanov @tjukanov started the #30DayMapChallenge Twitter phenomenon. It is a simple and fun challenge - for each day during the month of November everyone is invited to make a map with any tool they like towards the topic of the day and post it on Twitter under the hashtag #30DayMapChallenge. Last year it has become even more popular and widespread and during November Twitter was flooded with maps from all over the world. I aimed to solve all the challenges with my favourite toolkit, the Python programming language and its geospatial libraries. It quickly becomes obvious that besides crafting a map, coming up with a nice idea and the data acquisition will take a lot of time. Consequently, being effective with the available toolkits becomes very important. From dealing with cartographic projections and various geodata formats with Gdal, GeoPandas and Rasterio to the peculiarities of styling and image composition in Matplotlib, GeoPlot and Datashader, the talk will present the overall map challenge concept and lots of maps, and then dive into gotchas and revelations of making 30 maps in 30 days only using Python.

Alex is a Distributed Spatial Systems Researcher with many years of experience in geospatial data management and web- and cloud-based geoprocessing with a particular focus on land use, soils, hydrology, and water quality data. His interests include OGC standards and web-services for environmental and geo-scientific data sharing, modelling workflows and interactive geo-scientific visualisation. Alex is currently a Marie Skłodowska-Curie Individual Fellow (MSCA) at the University of Tartu where his aim is to improve standardised data preparation, parameterization and parallelisation for hydrological and water quality modelling across scales (H2020 GLOMODAT).

QGIS gives several opportunities to the users/developers to extend its capabilities using Python. During the presentation we walk through most of them. Besides pure Python a basic knowledge of QGIS PyAPI should be picked up. Fortunately there are many free tutorials, handbooks to help with the beginning steps. Tips are presented where to start. We start our journey with the QGIS Python console and script editor. It is an easy to use tool to execute direct Python commands or to write short scripts for yourself and your colleagues. ScriptRunner plugin gives a more comfortable environment for your simple scripts. A GUI helps you to organize scripts and metadata, too. You can write your own functions for Field Calculator. It is the next level of Python intrusion. This helps to make field calculator more powerful to solve your specific tasks. Python actions are the first in our series which extend the user interface for non programmers, too. There are different actions types, but Python actions are the most environment independent ones. They will work on all supported platforms. While the previous Python intrusions are useful for creative users with some programming background, with the next three you can create easy to use modules, programs for any QGIS user. The easiest from the point of view of the programmer is the so called Processing script, which can be used from the processing framework. You do not need to deal with GUI a lot, there is a simple environment to define GUI elements for input parameters of the script, which are managed by the Processing framework. If you would like to create an interactive tool in QGIS with its own GUI elements, you should create a plugin or a standalone application. For this you need some Qt practice, too. Short simple examples are presented for each above mentioned opportinuities.

Land surveyor with GIS and programming experiences, OSGeo charter member, working for Budapest University of Technology and Economics, leader of the Geo4All lab at the Department a Geodesy and Surveying, QGIS plugin developer.

Road traffic injuries are among the ten leading causes of death worldwide and they have a significant effect on the world’s economy. Governments and the private sector are making big efforts to reduce these numbers and, as a result, today we can have real-time information on traffic and weather conditions, in addition to traffic statistics. However, this information is available either post-accident or it is static. Knowing where accidents happen and the conditions under which they happen is very powerful information that can be leveraged to identify hotspots dynamically and take action to anticipate accidents (e.g., city administrations can share this information with their citizens and organize their traffic police accordingly, and logistics companies can use this information to avoid specific routes)

In this talk, we will show how different spatial data sources (road traffic, weather, road signaling, human mobility, points of interest, and working population) affect traffic accidents and how they can be used to identify hotspots dynamically. First, I will walk you through the spatial support selection phase and the process of bringing all data sources to the same support. Once the data is ready to be consumed, I will walk you through a spatial and temporal analysis of accident data using different tools and techniques. This first analysis will already give us some hints on typical characteristics of traffic accident hotspots. I will then present a predictive model using Regression Kriging with Random Forest as regressor that will allow us to predict annual accidents. This predictive model will help us validate our hypothesis of changing conditions affecting traffic accidents and the potential of defining dynamic hotspots. The analysis focuses on the city of Barcelona (Spain), which has a rich Open Data catalog available.

Data Scientist @CARTO

In our line of work we are often asked to develop a simple website with a map and a backend administrator to maintain the map data. It happens often that the backend requires more functionality than the frontend, transforming what was meant to be a “simple map” into a rather bulky budget.

Django is a general purpose web framework for Python that comes with a powerful backend administrator ready to use. It can be easily adapted to fit about every need we may encounter and can be used against legacy databases.

We, at the GIS service of the University of Girona (SIGTE-UdG), often use this approach to build simple admin backends in just a few days, making our custom "simple map" projects really simple and much more affordable while keeping a fully functional backend.

In the talk I plan on sharing our experience with this framework, showing some of the most interesting functionalities and how to make the Django admin geographically aware to fit our purposes.

Part time employed at the GIS service of the University of Girona (SIGTE-UdG) as a full-stack developer and tutor of the Python module at the UNIGIS MSc in Girona. Part time freelance working for Eixos.cat as the main programmer. Ex PHP programmer, now in love with Python, Django and JavaScript.

At GeoCat a team works on a number of python projects. Bridge for QGIS was first released in Bucharest and downloaded 3600 times since then. The underlying module, style 2 style which generates SLD and Mapbox Style from QGIS style has been adopted by a number of other projects. In this presentation we'll share some experiences while working on these topics. We'll look ahead on what's upcoming and would love to hear from you if you run into challenges on the topic of style, data and metadata conversions between products.

Software engineer at GeoCat. PSC member at pygeoapi & GeoNetwork. SDI/INSPIRE expert.

Google Earth Engine (GEE) is a cloud computing platform with a multi-petabyte catalog of satellite imagery and geospatial datasets. It enables scientists, researchers, and developers to analyze and visualize changes on the Earth’s surface. The geemap Python package provides GEE users with an intuitive interface to manipulate, analyze, and visualize geospatial big data interactively in a Jupyter-based environment. The topics will be covered in this workshop include: (1) introducing geemap and the Earth Engine Python API; (2) creating interactive maps; (3) searching GEE data catalog; (4) displaying GEE datasets; (5) classifying images using machine learning algorithms; (6) computing statistics and exporting results (7) producing publication-quality maps; (8) building and deploying interactive web apps, among others. This workshop is intended for scientific programmers, data scientists, geospatial analysts, and concerned citizens of Earth. The attendees are expected to have a basic understanding of Python and the Jupyter ecosystem. Familiarity with Earth science and geospatial datasets is useful but not required.

I am an Assistant Professor in the Department of Geography at the University of Tennessee, Knoxville. My research interests include Geographic Information Science (GIS), remote sensing, and environmental modeling. More specifically, I am interested in applying geospatial big data, machine learning, and cloud computing (e.g., Google Earth Engine) to study environmental change, especially surface water and wetland inundation dynamics. I am a strong advocate of open science and reproducible research. I have developed and published various open-source packages for advanced geospatial analysis, such as geemap, lidar, whitebox-python, and whiteboxR. More information about my research and teaching can be found on my personal website and research blog.

On 11 March 2020, the World Health Organization declared Covid19 a pandemic. Countries around the world rushed to declare various states of emergencies and lockdowns. Canada also implemented emergency measures to restrict the movements of people including the closure of borders, non-essential services, and schools and offices to slow the spread of Covid19. I used this opportunity to measure changes in seismic vibrations registered in Canada before, during, and after the lockdown due to the slowdown in transportation, economic, and construction activities. I analyzed continuous seismic data for 6 Canadian cities: Calgary and Edmonton (Alberta), Montreal (Quebec), Ottawa, and Toronto (Ontario), and Yellowknife (Northwest Territories). These cities represented the wide geographical spread of Canada. The source of data was seismic stations run by the Canadian National Seismograph Network (CNSN). Python and ObSpy libraries were used to convert raw data into probabilistic power spectral densities. The seismic vibrations in the PPSDs that fell between 4 Hz and 20 Hz were extracted and averaged for every two weeks period to determine the trend of seismic vibrations. The lockdown had an impact on seismic vibrations in almost all the cities I analyzed. The seismic vibrations decreased between 14% - 44% with the biggest decrease in Yellowknife in the Northwest Territories. In the 3 densely populated cities with a population of over 1 million - Toronto, Montreal, and Calgary, the vibrations dropped by over 30%.

To enable other students to undertake similar projects for their cities, I created a comprehensive online training module using Jupyter notebooks available on Github. Students can learn about seismic vibrations, how to obtain datasets, and analyze and interpret them using Python. They can share their findings with local policymakers so that they become aware of the effectiveness of the lockdown imposed and are better prepared for lockdowns in the future. When we make data and technology accessible, then lockdowns because of pandemics can be an opportunity for students to take up practical geoscience projects from home or virtual classrooms.

The outputs of my research on COVID19 and Seismic Vibrations are accessible at www.MonitorMyLockdown.com

I have been working on Space, Robotics, and Machine Learning for the past 6 years. I am the 2020 Global Winner of the 2020 NASA SpaceApps Covid19 Challenge for my “Masked Scales” project that used a home-made instrument to measure the impacts of COVID19 in Toronto and converted them into a musical. I won the Gold Medal at the 2020 IRIC North American Continental STEM Fair and all the 3 ribbons at the 2020 Youth Science Canada Online STEM Fair for my project on "Using Machine Learning to Remove Stellar Noise from Exoplanetary Data". I was one of the winners of the European Space Agency's Ariel Telescope Machine Learning Data Challenge and presented my project at the ARIEL: Science, Mission and Community 2020 Conference in the Netherlands. In 2014, I co-founded the HotPopRobot.com Initiative to carry out youth to youth science communication and outreach on space and carry out several free workshops and events each year.

Floods are a global hazard that may have adverse impacts on a wide-range of social, economic, and environmental processes. Nowadays our cities are flooding with increased occurrence due to more severe weather events but also due to anthropogenic pressures like soil sealing, urban growth and, in some areas, land subsidence. Frequency and intensity of extreme floods are expected to further increase in the future in many places due to climate change. The characterisation of flood events and of their multi-hazard nature is a fundamental step in order to maximise the resilience of cities to potential flood losses and damages. SaferPLACES employs innovative climate, hydrological and raster-based flood hazard and economic modelling techniques to assess pluvial, fluvial and coastal flood hazards and risks in urban environments under current and future climate scenarios. SaferPLACES platform provides a cost-effective and user-friendly cloud-based solution for flood hazard and risk mapping. Moreover SaferPLACES supports multiple stakeholders in designing and assessing multiple mitigation measures such as flood barriers, water tanks, green-blue based solutions and building specific damage mitigation actions. The intelligence behind the SaferPLACES platform integrates innovative fast DEM-based flood hazard assessment methods and Bayesian damage models, which are able to provide results in short computation times by exploiting the power of cloud computing. A beta version of the platform is available at platform.saferplaces.co and active for four pilot cities: Rimini and Milan in Italy, Pamplona in Spain and Cologne in Germany. SaferPLACES (saferplaces.co) is a research project founded by EIT Climate-KIC (www.climate-kic.org).

MSc Eng. Stefano Bagli, PhD – is an environmental engineer, data scientist and CEO of GECOsistema srl. He has more than 20 years of experience in environmental modelling and data science (Artificial Intelligence) in the field of air quality, climate services, fate and transport of pollutants in multiple media, hydrology, flood hazard/risk mapping, remote sensing, and developing SDSS embedding models and tools into GIS systems.

UNICEF and Development Seed are working to leverage machine learning, high-resolution imagery, and inexpensive cloud computing to create a comprehensive map of school at the global scale. Accurate data about school locations is critical to provide quality education and promote lifelong learning, UN sustainable development goal 4 (SDG4), to ensure equal access to opportunity (SDG10) and eventually, to reduce poverty (SDG1). However, in many countries educational facilities’ records are often inaccurate or incomplete. Understanding the location of schools can help governments and international organizations gain critical insights around the needs of vulnerable populations, and better prepare and respond to exogenous shocks such as disease outbreaks or natural disasters. Unfortunately, some national governments still don’t know where all the schools in their country are, or have out of date school maps.

Despite their varied structure, many schools have identifiable overhead signatures that make them possible to detect in high-resolution imagery with deep learning techniques. Approximately 18,000 previously unmapped schools across 5 African countries, Kenya, Rwanda, Sierra Leone, Ghana, and Niger, were found in satellite imagery with a deep learning classification model. These 18,000 schools were validated by expert human mapping analysts. In addition to finding previously unmapped schools, the models were able to identify already mapped schools with accuracy between 77 - 95% depending on the country. To facilitate running model inference across over 71 million zoom 18 tiles of imagery development seed relied on our open source tool ML-Enabler.

ML Enabler generates and visualizes predictions from models that are compatible with Tensorflow’s TF Serving. ML-Enabler makes managing the infrastructure for running inference at scale, and visualizing predictions straight-forward from a UI. ML Enabler will spin up the required AWS resources and run inference to generate predictions. ML Enabler helps harness the power of expert human mappers because model predictions can be validated within the UI and validated predictions can be used to generate new training data and re-train the initial model.

Machine Learning Engineer at Development Seed

How can we tell how COVID-19 has affected our economies ? Can we rely on official estimates ? Can we estimate the impact ourselves - with data from space - using satellite images available anyone in real-time using just a computer and a connection to the internet ?

With offices closed, data scarce and reliability of published results in question, the usual tools of obtaining such information is limited. It is in this backdrop that we explore using images of Earth taken at night-time from space. These images are available as GeoTIFF files with precise lon/lat co-ordinates at every pixel. Each pixel in turn contains radiance values that can reflect the economic activity on the surface. Such datasets have been shown to have correlation as high as 99% with actual GDP. Combined with data from power grids, the predictive power can be unparalleled. In this talk, I'll walk through examples of a) how such GeoTIFF files can be processed at scale using standard tools like GNU Parallel, Python Multiprocessing and how the resulting data can be then analysed by assigning radiance values with state/national boundaries and creating models that track changes over time.

Nataraj has 21 years of industry experience in developing the vision, strategy and execution of analytic capabilities in finance and pharmaceutical domains. Experience at IBM, UBS Investment Bank, UBS Wealth Management, Purdue Pharma, Philip Morris. He was the core architect of the RWE and Rx Data Analytics solution at Purdue Pharma which was spun out into a new startup, RxDataScience with over $ 3.5M in seed funding. He currently serves as the VP of Advanced Analytics at the firm.

Nataraj is a published author of multiple books on data science, journal articles and research papers in commercial market research, health outcomes, epidemiology and RWE Analytics. He has been a presenter, keynote speaker and Chairperson at over 20+ machine learning and AI-related healthcare conferences in US, Europe and Asia.

We present the within-field crop yield prognosis model developed by SEGES using machine learning (ML) methods and earth observation (EO) data. Our model and its prognoses were released to the Danish farmers in 2020 via our WebUI solution called CropManager, where the prognosis maps of a 10x10 meter resolution are visualized. This presentation will drill into the specifics of the utilized ML algorithm, ground truth yield data, and the other EO data sources used for creating the prognosis model. Additionally, we also present our use of DevOps solutions, like TeamCity, Octopus, Azure Machine Learning, and Kubernetes, to productionize the ML model.

SEGES is the leading agricultural knowledge and innovation centre in Denmark. We offer sustainable products and services for the agriculture and food sector, by collaborating with international customers, clients and farmers, to build a bridge between research and practical farming. https://en.seges.dk/.

I work as a data scientist at SEGES, where I develop artificial intelligence solutions for the agricultural community in Denmark. My work concerns the development of systems for decision support and automation in many aspects of farming, for instance, management of crops, animals, environment, finances, and legal cases. You are welcome to contact me if you have any interest in applying machine learning to data regarding agriculture, farming, or animals.

LinkedIn: https://www.linkedin.com/in/peter-fogh/

In recent years Ireland has experienced significant and unprecedented flooding events, such as groundwater floods, that extended up to hundreds of hectares during the winter flood season, lasting for weeks to months, and affecting many rural communities in Ireland. This issue was highlighted following widespread and record-breaking flooding in Winter 2015/2016 when little or no hydrometric data of groundwater floods was recorded. Further disruptive groundwater floods in 2018, 2020 and 2021 outlined the need for a systematic and large-scale mapping technique.

In response to these flooding events Geological Survey Ireland started the GWFlood project (2016 - 2019) and the GWClimate project (2020 – 2022) with the aim to establish an automated approach for mapping, monitoring and forecasting groundwater floods at a national scale, and to quantify the impact that Climate Change may have in groundwater systems. The use of remote sensing data, Sentinel-1 satellite imagery from the European Space Agency Copernicus program, was key to overcome practical limitations of establishing and maintaining a national field-based monitoring network. Remote sensing data was complemented with Geographic Information System (GIS) datasets to improve reliability in the final products, and with hydrological models to generate historical groundwater flood records based on meteorological data from Met Eireann and to provide forecast for groundwater floods.

Key deliverables of the GWFlood project included: 1) a national historic groundwater flood map 2) a methodology for hydrograph generation using satellite images, 3) predictive groundwater flood maps, and 4) a groundwater monitoring network to provide baseline data. The GWClimate project is enhancing the tools developed by GWFlood in order to deliver: 1) seasonal peak flood maps, 2) near-real time satellite-based hydrographs, 3) groundwater flood forecasting tools, and 4) maps evaluating the impact of climate change in groundwater systems in Ireland.

Data and maps from GWClimate and GWFlood projects are available at: 1) https://gwlevel.ie, and 2) https://www.gsi.ie/en-ie/programmes-and-projects/groundwater/activities/groundwater-flooding/gwflood-project-2016-2019/Pages/default.aspx

Dr. Joan Campanyà has a background in physics and geophysics and holds a PhD in Earth Sciences from the Universitat de Barcelona. He has contributed to many publications on the use of electromagnetic geophysical methods for characterizing the Earth’s subsurface and evaluating the impacts of Space Weather on ground-based infrastructures. Joan’s current research with Geological Survey Ireland and IT Carlow is focused on using remote sensing data for mapping, monitoring, modelling, and forecasting floods at a national scale, with particular interest in groundwater flooding. The main products from this work are available to the public and have direct implications for government policy and planning.

In this talk, I will discuss some Geodata challenges in the operations side of Moovit. Moovit, an Intel company, is helping to create cleaner, safer cities by guiding people in getting around town using any mode of transport. Today, Moovit serves over 950 million users in 3,400 cities across 112 countries and is the creator of the #1 urban mobility app.

Our Transit Data Repository contains millions of data points, including Real-Time arrivals and Service Alerts from more than 7500 transit operators around the world.

The data comes from multiple sources, is stored in multiple environments, and several formats. Python helps us create cross-platform processes to analyze, and visualize the data. especially for Geodata, we use Python to integrate data in QGIS for GIS experts and to create flexible reports using Jupyter notebooks for no geo experts.

Geo data integration team leader at Moovit

TileDB Embedded is an open-source, universal storage engine with integrations into many tools that already exist within Python such as Dask, xarray and Pandas as well as geospatial specific frameworks such Rasterio, Python-PDAL, GeoPandas and our own open-source library for netCDF and HDF-like data, TileDB-CF-Py.

TileDB Embedded is ideal for geospatial data as it is based on sparse and dense multi-dimensional arrays, implementing indexes such as R-trees and Hilbert curve orderings. It is cloud-native and can encompass multiple geospatial domains. I’ll make the case for universal geospatial data storage in the following parts:

Analysis-ready geospatial data

No more files. A universal format can cover all geospatial data types as sparse or dense arrays allowing rapid slicing, arbitrary metadata, and with versioning and time-traveling built-in. Here, I’ll examine the shared structure of geospatial data that makes it best suited for array-based storage.

Superior interoperability

The tools don’t change. I’ll look at how TileDB arrays work within the Python ecosystem. Leverage PDAL, GDAL and existing tools such as Dask, xarray and Pandas to perform geospatial analysis.

Solution focused

We focus on end-to-end solutions, not format standards. Despite defining a powerful open-spec data format for all geospatial data, our goal is to deliver unprecedented speed for analytics queries and integration with numerous computational tools, via the well-defined APIs of our TileDB Embedded storage engine.

Proven

TileDB Embedded is successfully used by high-profile users and customers to store SAR, hyperspectral, weather, seismic data as well as point cloud data from SONAR and LiDAR sensors, all within a universal data engine that can be used seamlessly from Python. The talk concludes with a co-presented example from TileDB user Capella Space.

Norman is the VP of Geospatial at TileDB. Prior to joining TileDB, Norman focused on spatial indexing and image processing, and held engineering positions at Cloudant, IBM and Mapbox. He has a master's degree in Mathematics from the University of Durham, England. In his free time, Norman likes to garden and fix up his old house.

The geological field experience is traditionally directed on raw data collection with orientation measurements, observations and rock samples descriptions representing some classic examples. Creating geologic and contour maps by hand are also prominent activities within the limited timeframe. We aim to improve this strategy by introducing a seamless and iterative 3D-modelling workflow, in the pursuit of shifting focus back to geological idea and concept integration versus data. Our proposed workflow is intended to work in-parallel, thereby bolstering the efficiency of allotted field time. The workflow for 3D-modelling and visualization combines new and existing Python scripts and using open-source tools to furnish users with a coherent approach for achieving both maps and models. Our approach utilizes GemPy, a 3D geological structural modelling tool, based on the Potential Field (PF) method. Using a sparse dataset, a regional 3D-model was generated and also easily re-generated upon the introduction of new data. Cross-section views of the 3D-model can also be obtained and 2D geological maps may be extracted. With respect to data management, well-known tools such as rasterio, geopandas and numpy were exploited for data imports and processing. The digital elevation model (DEM), field data stored as shapefiles and other required data were organized into a single geopackage, which can be shared and updated as needed. Much effort has been placed on the ease-of-use for data organization. Our proposed approach may have strong impacts on field data collection and decisions, especially in regions with sparse geological knowledge. This notion is supported by promising initial results, well-aligned with inferred regional geology.

Trinidadian geoscientist with a former oil & gas life in Houston, Texas. Currently pursuing a PhD in 3D geological modelling in Mons, Belgium. Unsure of why I have a male Flemish name. Always on the look out to increase workflow efficiency!

In this talk the attendee will be able to understand what is the Open Data Cube (ODC) , why it is important and some use cases. In addition we will demonstrate a typical satellite image processing workflow in order to show the benefits offered by the tool. In the demo:

1. We will deploy an Open Data Cube environment in Docker containers
2. Then, a satellite image is added to the data cube index.
3. Finally, will query the index and retrieve satellite image data to perform a basic NDVI analysis.

Aurelio Vivas graduated as Computing and System Engineer at Universidad del Valle, in Cali, Colombia in 2018. He got a MSc and is currently a PhD student at Universidad de los Andes, in Bogotá, Colombia. Since 2018, he has been a teaching assistant with the System and Computing Engineering Department. He has been able to participate in remote sensing, desktop grid computing, and high-performance molecular dynamics projects. His research interests include Programming Languages, Scientific Parallel Computing and Software-defined Infrastructure.

eemont was created to speed up the writing of Google Earth Engine python scripts and it extends EE classes with new methods such as clouds/shadows masking, image scaling and spectral indices computation.

Let's take a look at the simple usage of eemont:

```python import ee, eemont

ee.Authenticate() ee.Initialize()

point = ee.Geometry.Point([-76.21, 3.45])

S2 = (ee.ImageCollection('COPERNICUS/S2_SR') .filterBounds(point) .closest('2020-10-15') # Extended (pre-processing) .maskClouds(prob = 70) # Extended (pre-processing) .scale() # Extended (pre-processing) .index(['NDVI','NDWI','BAIS2'])) # Extended (processing) ```

And most of these methods are available for a bunch of platforms such as Sentinel 2 and 3, Landsat Series and MODIS products!

I'm a Topographic Engineer graduated from the University of Valle (Cali, Colombia) and a Master in Data Science graduated from the University of Cantabria (Santander, Spain).

I have worked for almost 5 years on topics related to Geomatics: Geographic Information Systems, Photogrammetry and Remote Sensing (the latter being my field of specialization). During this time I have applied data analysis (statistics, programming and Machine Learning) to the geospatial data branch. During the last year I have expanded my field of work to the Data Science area, being my research interests now: Data Science, Remote Sensing, Machine Learning, Computer Vision, Deep Learning and Google Earth Engine.

I have a great liking for the publication of results and I have 10 research publications (6 research articles in journals and 4 research articles in conference proceedings). I have also been a speaker at different conferences hosted by companies such as Google (Geo For Good Summit, 2020), IGAC (International Geomatics Week, 2015 - 2017) and Tecnicaña (IX Atalac-Tecnicaña Congress, 2018).

Artificial intelligence, including machine learning and deep learning, have been increasingly utilized for humanitarian applications, from combating climate change to assessing car accidents. Specifically in the domain of geoscientific analysis, deep learning-based remote sensing has yielded many promising humanitarian applications and results. The occurrence of natural disasters is increasing in frequency and intensity due to climate change, and efficient and accurate computational methods of assessing the building damage caused post-disaster must be in place. This assessment aids in the allocation of resources and personnel. Using Python, we can develop convolutional neural networks and other deep learning architectures to detect and classify levels of infrastructure damage to inform disaster relief and recovery programs. A popular data source for doing so is real-time satellite imagery, which is much more easily gathered than data from on the ground. Other data sources include social media posts.

Thomas Chen is a machine learning/computer vision researcher from the United States. He is particularly interested in using deep learning-based computer vision to assess and gain insights into damage in objects in imagery. More broadly, he is passionate about applying machine learning to real-world issues that face society, including climate change.

Since almost all industries are more or less connected to Location and mapping, it is important to spread awareness and literate developers to understand different aspects of the GIS (Geographic Information System) industry. The first Part of this series focuses on different GIS Data types and how to read them, This includes understanding different data formats such as Shapefiles, GeoJSON, WKT, CSV, TIFF, GeoTIFF, etc.. Users can actually read such files on their computers and be familiar with them. The second part of this series focuses on geospatial analysis with python. Users will first practice working with some core GIS functionalities using GDAL and OGR on the terminal (and later in python). After this, users will be familiarised with the most widely used geospatial python libraries such as pandas, geopandas, fiona, shapely, matplotlib, PySAL, rasterio.

Complete Series is divided into the following sub-topics : 1. Introduction and Installation of all Geospatial libraries in computer and in python environment 2. Working with GDAL and OGR capabilities 3. Spatial Operations and Relationships 4. Vector data analysis and visualization 5. Raster Data analysis and visualization 6. Working with Interactive Map in a python notebook

Pre-requisite for this workshop: 1. Basic knowledge of python 2. Basic knowledge of GIS and GIS Data formats

Access more detailed information at : https://github.com/krishnaglodha/GeoPython_2021#workshop-submission--geospatial-analysis-using-python-101

Hello, world! I'm Krishna, I am a Full-stack GIS Developer who works with Open-source technologies as well as Enterprise Technologies.

Most part of my day goes into thinking how can I make more complex GIS applications using nothing but a cup of coffee and Open-source technologies

Other time I enjoy writing GIS articles, have a look at them here https://krishnaglodha.medium.com/
Read more about me at http://krishnaglodha.com/
LinkedIn : https://www.linkedin.com/in/krishnaglodha

Dissolved oxygen is a key indicator in aquaculture reflecting the change in water quality. In this study, we used the Extreme Gradient Boosting (XGBoost) machine learning algorithm to predict dissolved oxygen in a lagoon (western Greece), considering ten physicochemical and meteorological explanatory variables, that is water temperature, pH value, oxidation-reduction potential, air temperature, salinity, chlorophyll, relative humidity, atmospheric pressure, wind speed and wind direction. XGBoost was trained and then evaluated attaining a root mean square error of 0.66 mg/L in predictions. Results showed that pH, salinity and air temperature were the key factors driving dissolved oxygen variability. Special focus was given on interpreting the outcomes using Additive exPlanations (SHAP) methodology, aiming to elucidate the environmental windows that cause low levels of dissolved oxygen (anoxic conditions), which may have severe impact on the survival rate of aquatic organisms.

Currently, I am a Research Associate at the Hellenic Center for Marine Research (HCMR). My research is focused on modelling of fisheries and marine litter pollution, and Machine/Deep Learning in Marine Science.

The division of the software into components is necessary because large amounts of data can only be processed in data centers while the scientists are in regions with the lowest Internet bandwidth.

The Mission Support System (MSS) software consists of several components. - an enhanced OGC webmap server capable of visualizing the large data generated by complex 3-D atmospheric simulations in a highly configurable manner and delivering the resulting small PNG images over the Internet - a GUI that inserts a flight path on the images transmitted by the OGC webmap server - a socketIO based server that synchronizes the flight paths among the participants simultaneously. - Support processes for data preparation for the webmap server

In order to measure at the most scientifically interesting locations, model predictions of relevant variables such as meteorological parameters, chemical composition, or particle information are necessary to guide the aircraft to the location of interest. The Mission Support System enables the planning of optimal flight path by visualizing the results of model simulations in combination with the selected flight path, allowing for a simple iterative and collaborative improvement process.

I am a programmer from Jülich, Germany. That‘s a small town between Aachen and Cologne.

I work at the Forschungszentrum Jülich GmbH. Employees research in the fields of energy and the environment, information and brain research with the aim of providing society with options for action facilitating sustainable development.

My work is related to atmospheric science.

I have been a fellow of the Python Software Foundation since 2013.

A more detailed interview at blog.pythonlibrary.org

PyCurious is a python package that able to calculate Curie Point Depth (CPD) from magnetic data using Tanaka and Bouligand. In this talk, I will share my experience on working with this package to create the CPD map of Botswana for my MSc research using Bouligand method from an aeromagnetic dataset.

Geoscientist

High Peformance Computing using Quantum Computing is the future and the GIS community should start learning about its concepts and implementation techniques in order to leverage Quantum Computing for solving complex spatial problems. This talk focuses on introducing the basic concepts of Quantum Computing using Qiskit.

Qiskit is a popular open-source, feature-rich, and modular Quantum Computing Python SDK by IBM. It enables working with Quantum Computers at the level of circuits, algorithms, and application modules.

The following topics will be covered during the talk: - Basics of Quantum Computation - Qubits and the Quantum States - Working with Multiple Qubits - Qiskit and Qubit Gates - Hands-on mode - Building small Quantum Circuits - Where to go from here - a roadmap for working with Quantum Algorithms - Applications for the GIS community (Spatial Algorithms and more)

Pre-requisites: - Basics of Linear Algebra - Working knowledge of Python - Basics of Quantum Physics (good to have; the session will cover some basics)

Talk Level: Intermediate

Anmol Krishan Sachdeva is an International Tech Speaker, a Distinguished Guest Lecturer, a Tech Panelist, and has represented India at several reputed International Hackathons. He is a Deep Learning Researcher and has around 8+ publications.

Last year, Anmol helped organize EuroPython 2020, GeoPython & Python Machine Learning Conference 2020, and PyCon India 2020, and all of them were a huge success. He has done MSc in Advanced Computing (ML, AI, Robotics, Cloud Computing, and Computational Neuroscience) from University of Bristol, UK, and currently works at the OLX Group as a Site Reliability Engineer.

In past, he has spoken at renowned conferences and tech forums like KubeCon, PyCon, EuroPython, GeoPython, and got invited as a Chief Guest / Guest of Honour at various events.

He likes innovating, keeping in touch with new technological trends, and mentoring people.

Over the past decade or so agricultural technology has seen remarkable progress giving birth to precision agriculture whereby huge quantities of useful data can be extracted from Internet of Things, sensors, satellites, weather stations, robots, farm equipment, agricultural laboratories, farmers, government agencies etc. In this talk I will highlight various aspects of putting this data to practical use towards enhanced decision-making in crop management practices leading towards crop yield optimization while also helping towards preservation of the environment. This is done by means of modern machine learning methods that treat the data as being part of a a complex system of interconnected variables and conditions with the final output enabling predictions of crop yield to be made. A special emphasis is given to spatio-temporal aspects of agricultural data, and how it is incorporated into our crop yield prediction algorithms.

Some analysis of various agricultural fields and corresponding spatio-temporal variables such as weather, soil properties, topographical factors is also performed in order to motivate the underlying theory for our proposed methodology.

Arjumand Younus is currently working as a postdoctoral researcher at CONSUS in University College Dublin for a massive government-funded project on data-driven agriculture for Ireland. Before that, she was part of the Recommender Systems group at Insight Centre for Data Analytics in University College Dublin. She is also an active member of Women in Research Ireland which is a not-for-profit organisation working to further the cause of minorities and underrepresented groups within research circles of Ireland. Furthermore, she is also the co-founder of the Dublin chapter of Women in Machine Learning and Data Science while is also actively involved with the Python Ireland community. She has a passion for projects belonging to the theme of data science for social good. She is the recipient of Google Women Techmakers scholarship for Europe, Middle East and Africa region. She completed her PhD in Computer Science in 2015 jointly from the National University of Ireland, Galway and University of Milano-Bicocca, Italy; and her Masters in Computer Science from KAIST, South Korea.

We are facing an unprecedented Global Change. We seem to have a not inconsiderable influence on this state and we already know what we can do to mitigate the possible consequences. We are just beginning to understand, if, how and how much these consequences are connected to the Global Change. It is not possible to know all details and there could be a need to adapt to new situations in shorter time steps. Earth Observation Data, especially satellite data, has a great potential to support us during this process. And we must be aware what happened since 1965 when the USGS wanted to start a program for systematic earth observation. With the first Landsat satellite, launched in 1972 started a new era. Now, we are close to the 9th Landsat generation, and the whole archive is open to the public. ESAs Copernicus program is the next game changer, e.g. Sentinel-2A and 2B twin satellites orbiting since 2015 and 2017, using open data policies. Like Sentinel-1A, 1B, 3A and 3B. Tons of data, and we need to get the information out of them in a efficient way. Datacubes are common tools for storing information, also for Earth Observation Data today. The Open Data Cube (ODC) Python library is an impressive open source tool to manage satellite data, and lower the obstacles to bring experts together, which can immediately use analysis ready products or workflows from a large an still growing community. Without producing high costs. Our subnational ODC implementation tries to support decision making in the context of forest monitoring without using proprietary services to lower barriers for state authorities with different data protection regulations.

A plumber who studied Physics and Physical Geography. Interested in nature and ecology as well as open source and open data and their potential to face present and future problems.

The Participatory Terrain model (ParTerra) fuses together data from OpenStreetMap (OSM) and any base elevation dataset to create a high-resolution digital terrain model for any area in the world. A roughness map, specifically designed for hydrodynamic/hydraulic modeling purposes, can also be created. It is designed to facilitate a participatory approach involving relevant stakeholders but can also be used as a stand-alone tool without stakeholder involvement.

This talk will give a brief description of the ideas behind the tool, the way it's implemented in Python and highlight a few use cases from concluded projects at Deltares.

The code is open source, hosted at https://gitlab.com/deltares/parterra/parterra-python. An online application for viewing and analyzing some preliminary results over several regions can be found at https://arjenhaag.users.earthengine.app/view/parterra

Researcher/consultant at Deltares, a Dutch non-profit applied research institute with a focus on water. Personally focusing on hydrology, flood (forecasting) and remote sensing.

The 100-year Global Warming Potential GWP100 is a widely used, simple concept to quantify the impact of climate pollutions. Unfortunately, it does not account appropriately for the different impacts of long-lived climate pollutions (LLCPs) and short-lived climate pollutions (SLCPs).

The modified version GWP* improves upon the GWP100, but also has shortcomings. This talks demonstrates the development of the modified GWP*, that combines both GWP100 and GWP* with a simple linear approach. The modified GWP* provides a better approximation of the system behavior when compared with results calculate with more complex climate models. The talk also introduces the concepts of carbon dioxide equivalents (CO2-e) and CO2-warming-equivalent (CO2-we). The ephemeris is on quantification with Python. Code examples show how the models are implemented.

Sources

This talk is based on the findings in this paper:

Cain, M., Lynch, J., Allen, M.R., Fuglestedt, D.J. & Macey, A.H. (2019). Improved calculation of warming- equivalent emissions for short-lived climate pollutants. npj Climate and Atmospheric Science. 2(29). Retrieved from https://www.nature.com/articles/s41612-019-0086-4

The code is inspired by:

https://gitlab.ouce.ox.ac.uk/OMP_climate_pollutants/co2-warming-equivalence/

and uses the simple emissions-based impulse response and carbon cycle model FaIR:

https://github.com/OMS-NetZero/FAIR

CEO of hydrocomputing

The trackintel framework structures human movement into hierarchical units (i.e., positionfixes, triplegs/stages, trips and tours), and provides functionality to generate everything starting from the raw tracking data. It also provides functions for a complete mobility data processing pipeline: - Preprocessing (filtering, outlier detection, imputation of missing values, quality assessment, data aggregation) - Contextual Augmentation (map matching, trajectory algebra-based context addition) - Analysis (extraction of mobility metrics, preferences, systematic mobility) - Visualization and Communication (generation of maps, charts)

In this talk, we will introduce you to the most important functionalities of the trackintel framework. Moreover, using real-world tracking data, we will provide typical use-cases for different mobility data processing tasks, revealing the usefulness of the framework.

Ye Hong is currently a Ph.D. student at Institute of Cartography and Geoinformation, ETH Zurich. He holds a B.Sc. in GIS and remote sensing from Sun Yat-sen University, China, and an M.Sc. in Geomatics from ETH Zurich. His current work focuses on understanding human mobility behavior and developing models for mobility prediction.

Broad software development skills are becoming a hard requirement for potential (scientific) users of the Free and Open Source Software (FOSS) ecosystem around Geographic Information System (GIS). There is a clear boundary between what can be done with Graphical User Interface (GUI) applications such as QGIS only and contemporary software libraries – if one actually has the required skillset to use the latter. Far more than just rudimentary programming development skills are usually required, however, which are difficult to acquire and distract from the actual scientific work at hand. Installation and deployment on their own of much desired high-performance computing (HPC) software libraries for e.g. general-purpose computing on graphics processing units (GPGPU) or computations on clusters or cloud resources are very often show-stoppers. Recent advances in Python packaging and deployment systems ease the situation and enable a new kind of thinking. Desktop GUI applications can now much more easily be combined with the mentioned type of libraries, which lowers the entry barrier to HPC applications and the handling of large quantities of data drastically. This talk aims at providing an overview over where we are at the moment and what is practically and theoretically possible down the road.

I am a free scientist without specialization. I have more than one and a half decades of experience in various scientific disciplines, related data analysis & computing from embedded systems to super computers as well as the development of instrumentation (hardware & sensors). A lot of my work has evolved around geophysics, [aero-] space engineering and (many) related disciplines. It can be broadly described as data science with strong ties to the mentioned domains. Python has been a critical part of my work for more than twelve years.

PyInterpolate is designed as the Python library for geostatistics. It's role is to provide access to spatial statistics tools used in a wide range of studies.

If you’re:

• GIS expert,
• geologist,
• mining engineer,
• ecologist,
• public health specialist,
• data scientist.

Then this package may be useful for you. You could use it for:

• spatial interpolation and spatial prediction,
• alone or with machine learning libraries,
• for point and areal datasets.

Pyinterpolate allows you to perform:

1. Ordinary Kriging and Simple Kriging (spatial interpolation from points),
2. Centroid-based Kriging of Polygons (spatial interpolation from blocks and areas),
3. Area-to-area and Area-to-point Poisson Kriging of Polygons (spatial interpolation and data deconvolution from areas to points).

R&D Manager specialized in spatial data science and time series analysis. I'm working as a consultant for multiple companies from aerospace, e-commerce, health and well-being industries from Poland and Europe. I love Hackerspace movement and I spent my free time dancing as a bboy.

My blog: https://ml-gis-service.com/

Twitter: https://twitter.com/SimonMolinsky

In this talk will be shown and discussed the process to create a WebApp using Fast API (Python), PostGIS and JavaScript (Vue.JS, OpenLayers and Turf.JS) and also a Widget to generate hexagon or square bins, including explanations about the architecture, some alternatives for this process and others insights that will improve and help real time data analysis using the binning process.

Is widely know that in our pathway to find solutions and solve problems we most of the time face situations that can be solved in many ways. But also, in our path to be more productive we commonly pressure by time, project delivers and many others aspects, so we mainly don't have the time to ask for ourselves: "which is the ways to solve this problem? Which is the best one?". Most of time we just start and go.

Besides the technical talk and explanations also will be show that GIS development is not an easy task for sure as we have many technologies and alternatives to execute our tasks.

Taking this into consideration, we will discuss that we can develop the same app and tool in more than one way, specific describing suggestions for architecture, databases, backend and frontend solutions for the binning process.

Vinícius Cruvinel Rêgo is Civil Engineer, M.Sc. in Geographic Information Systems and Science, MBA in Project Management, MBA in Fullstack Development and Specialization in Distance Education.

Has a multidisciplinary background with main emphasis in project and program management, planning and control, GIS systems development and technical leadership.

Work as a Planning & Control Specialist and Solution Developer in the Management of the Environmental Sanitation Program at CAESB, financed with IDB resources and management carried out by COBRAPE. He is also Titular Professor of the Civil Engineering Course at UNIPLAN teaching subjects from urban mobility and technology.

As a Consultant and Developer he works in projects from many areas like Agriculture, Transport, Water Resources, among others, mainly in the architecture and development of platforms and applications that use Geographic Information Systems (GIS).

Regarding the academic part, he is currently a regular student in the PhD course in Development, Society and Territories at the University of Trás-os-Montes and Alto Douro (UTAD), located in Vila Real, Portugal, conducting researches about urban mobility with focus in micromobility.

This talk will attemp to show, in a didactic and straightforward manner, how you can perform spatial analysis with both of two tools (Spatial SQL and GeoPandas) in order to solve the same given exercise. Several exercises will be discussed in order of ascending complexity, from inside a Jupyter Notebook environment.

This could be of interest for any of the following:

• python and sql people interested in spatial analysis

• postgis people interested in the python environment

• geospatial python people interested in the database environment

Geospatial developer, using exclusively Free & Open Source Geospatial Software for the last 15 years

Unlike types of data that are more commonly dealt with in the industry these days, such as numerical data, text or image data, audio signals need a different approach while trying to extract information and building machine learning models. This talk will highlight the challenges with Audio Classification problems starting with what an audio signal is and what its numerical representation means, how it is widely different from other data types, what feature extraction from audio looks like, how to go about it, what it means and the open source tools in Python that can be leveraged for the same. Digital signal processing, that includes audio processing, is a whole separate field to study and leveraging portions of learning from that in order to build successful models on audio data is an interesting and challenging problem. In addition, Matlab is a popular language of choice with great tools for audio signal processing. Python being a popular language of choice for Machine Learning presents another set of challenges to build successful audio and speech classification solutions in Python alone. Focus will then upon how to build classification models from the features representing the unseen information from audio and speech signals and doing it all leveraging different open source tools available to Python users. This will be followed by a few examples of different audio classification and prediction problem statements and a solution for attempting to solve them using Python using the different features formation techniques and tools discussed earlier in the talk.

Jyotika Singh is Director of Data Science at ICX Media where she works on NLP, feature engineering, supervised & unsupervised machine learning, research, analytics, programming in Python and distributed computing with Spark. She earned her Master’s from the UCLA where she researched signal & speech processing, developed novel approaches to remove noise from speech and worked on a variety of machine learning on image, text, social media, consumer & entertainment data.

Outside her work, she enjoys working on a variety of problem solving techniques on text, audio and image data, has opened multiple github open source projects, such as pyAudioProcessing and pyYouTubeAnalysis, to share her findings and work with the Python and Data Science community.

In her free time, she is big on spending time with family and friends, painting, art & decor, and trying out different sports.

Past chemical emergencies in the United States prompted the initiation of a variety of toxic substance and pollution control programs and regulations, including the Emergency Planning and Community Right-to-Know Act and the Clean Air Act. While these have produced decades-worth of valuable pollution datasets, they are stored on a government website in a collection of CSV tables. This method of accessibility is largely incompatible for public analysis due to the static nature of tables— the need to download them locally and appropriately query them to extract relevant data. For analysis, data is best visualized with dynamic tools and within interactive environments.

This application supports open source software and was developed entirely from the backend database to the front end by relying on publicly-available pollution datasets, PostGIS, GeoDjango, and LeafletJS. With Cal ToxTrack, users can utilize a map and spatiotemporal tools to visualize what chemicals have been released, to what magnitude, and where; practicing their right-to-know.

Full stack web developer with a background in geology, environmental studies, and GIS.

The foundation of our model is the anonymized and aggregated taxi trip data shared by Uber through their Uber Movement platform. This is a large dataset which has both spatial and temporal components to it. We combine this dataset with OpenStreetMap road network and routing data to build a model of travel time that accounts for travel distance, hour-of-the-day and historic traffic. We will demonstrate the process and show a live demo of the model in action.

We will introduce modern geospatial libraries such as geopandas, shapely, folium and services such as Open Source Routing Machine and OpenRouteService that make working with large spatio-temporal datasets easy. Come and learn about how you can use these open datasets and learn about incorporating spatial datasets in your data science workflows.

See the code and demo online!

Vishnu is an MBA student specialising in Analytics and finance from NIT Calicut . His interest are in the field of applied data science and currently works on applying machine learning techniques in the area of urban transformation using open source geospatial datasets and APIs.

Ujaval is the founder of Spatial Thoughts - a learning platform for modern geospatial technologies. He got his Masters in Geospatial Information Engineering from University of Wisconsin-Madison. After joining Google Inc. in California, he moved to India in 2006. He was one of the early employees at Google India and part of the team that launched Google Maps for India. He has worked on multiple Geo teams at Google and led the aerial imagery group in India. He was the lead trainer for Google Earth Engine in India and conducted numerous training sessions across the country.

Ujaval’s interest and expertise is using open-source spatial analysis software to automate GIS and Remote Sensing workflows. He is a world-renowned GIS expert and training facilitator who is passionate about advancing the use of open-source technologies in teaching and research. He is also a visiting faculty/researcher at University of Johannesburg, South Africa.

The purpose of this lecture is finding the effect of different parameters on the prevalence of corona. In this paper, similar behaviors in three parameters of wind speed, temperature and air pressure are clustered between 70 cities and then by clustering the daily changes in the statistics of corona virus cases, cities that participate in different clusters are identified and the impact of each of these parameters on corona outbreak determined.

On the surface, Abouzar Ramezani is an accomplished Geo-Spatial Analyst with over 12 years of experience in Geomatics Science.

A map in a website is the best way to make geographic data easily accessible to users because it represents, in a simple way, the information relating to a specific geographical area and is in fact used by many online services.

Implementing a web map can be complex and many adopt the strategy of using external services, but in most cases this strategy turns out to be a major data and cost management problem.

In this talk we'll see how to create a web map with the Python based web framework Django using its GeoDjango module, storing geographic data in your local database on which to run geospatial queries.

Through this intervention you can learn how to add a map on your website, starting from a simple map based on Spatialite/SQLite up to a more complex and interactive map based on PostGIS/PostgreSQL.

I’m a Python developer who contributes to the Django project and gives talks at tech conferences.

I’ve been a GNU/Linux user since 2000 and I use and promote Free Software.

I have a degree in Computer Science and currently I work remotely, in Pescara (Italy), for 20tab as CTO.

This talk looks at the current status of data visualization capabilities from the perspective of analysts / scientists working with GPS tracking data. We explore interactive visualization possibilities provided by the Pandas ecosystem: starting from GeoPandas & MovingPandas and its visualization options and limitations, followed by a dive into Pandas & Datashader for larger datasets and creating linked plots.

Anita Graser is a spatial data scientist, open source GIS advocate, and author. She works as a scientist at the AIT Austrian Institute of Technology in Vienna and teaches a class on QGIS and Python at UNIGIS Salzburg. She serves on the QGIS project steering committee and is the developer of MovingPandas, a Python library for analyzing movement data. She has published several books about QGIS, including “Learning QGIS” and “QGIS Map Design”. In 2020, she was awarded the Sol Katz Award 2020 for Geospatial Free and Open Source Software.

Migration to Python 3 is over, but that's not the end of the journey. Although your code runs with the currently supported Python 3.6 to 3.9, there may be some pieces of code that look obvious to you, but may surprise younger developers who have never seen Python 2 code.

At the end of 2020 I started looking for Python projects on GitHub and helped them to get rid of those Python 2 relics. I'll show you a few recipes beyond the automatic tools, how to make your code modern and prepared for future updates.

And no, we should not return to Python 2. We should get rid of it completely.

Greedy polyglot, sustainable urbanist, unicode collector, wandering openstreetmapper, and an hjkl juggler. Using Python to make the sun shine, the wind blow, and the gas flow.