Challenge by HUB Ocean and Cognite (Norway):

The environmental effects of marine transport include air pollution, water pollution, acoustic, and oil pollution. The share of maritime transport emissions in global anthropogenic emissions has increased from 2.76% in 2012 to 2.89% in 2018 according to IMO’s Fourth Greenhouse Gas Study. At the same time, the demand for seaborne trade is projected to grow by 39% until 2050. Concepts like “green corridors” (specific trade routes between major port hubs where zero-emission solutions are supported and demonstrated) and “blue corridors” (critical ocean habitats for migratory marine species) are surfacing as potential solutions to enable early adoption of alternative fuels and conservation of vulnerable ocean areas. Various stakeholders such as financial institutions, administrative authorities and ordinary consumers (you and me) increasingly have a need to understand what "green transportation" is so that one can make the right choices and influence the decision makers. Through compelling data storytelling it should, in theory, be possible to demystify this

As a participating team, we want you to develop a solution that make it easier to gain insight into current global, regional or local traffic patterns for ships, how greenhouse gas emissions are trending and how particularly vulnerable areas and marine life are affected. Such a solution can be realized as Python notebooks with visualizations, map layers and animations in GIS tools or web-apps or even mobile apps. To make the task a bit easier:

  • you will be able to take advantage of both relevant data and computing power in both Azure and/or the Ocean Data Platform by HUB Ocean. In the data platform itself, you will find relevant detailed data for around 250k merchant vessels worldwide, as well as processed position data for these ships (from satellite tracking systems), high-fidelity emission estimates and low-fidelity fuel/emission-reports for the period 2020-now.
  • you will be able to take advantage of large global datasets for marine protected areas, biodiversity and ports.
  • you can also combine this data with a massive amount of relatively up-to-date satellite data unlocked in the Microsoft Planetary Computer. This is geospatial data that can be used in different forms or transformed and compiled into even more relevant data sets for more tailored solutions.

HUB Ocean has prepared a repo with example notebooks for how to retrieve and use the global vessel emissions data. The repo can be found at https://github.com/C4IROcean/ClimateHackathon2022 

If you want to use the Ocean Data Connector, HUB Ocean’s cloud hosted Jupyter Lab environment, send an email to thomas.fredriksen@oceandata.earth and request access. In this environment you can select virtual machines up to 8 CPU and 28 GB RAM. All necessary code is preinstalled to access the vessel emissions data, in addition to many other ocean related datasets like biodiversity and marine regions.

A few examples on what you could build or visualize during the hackathon:
  • Ship traffic density and emission trending in UNESCO world heritage sites or special vulnerable areas like West Norwegian Fjords, Great Barrier Reef, Svalbard and more
  • Hot spot maps showing conflicts between marine mammals (blue whales, fin whales, humpback whales and many other amazing species) and ship traffic, for instance by combing observations from the Ocean Biodiversity System (103 million observations) with ship voyage tables
  • Various systemic inefficiencies where ships “hurrying up and waiting” outside major ports like Shanghai, Singapore, Rotterdam, San Francisco or queuing outside the Panama or Suez canals
  • Enrich the World Port Index dataset with more information about new or planned ports and terminals with infrastructure for alternative fuels like Hydrogen or Green Ammonia applying web text mining and natural language processing
  • Detect (by machine learning) classical ferry routes going from port A to B to A and showcase “what if we electrified that route” scenarios, for instance the Seattle – Bainbridge Island, Oslo – Kiel, Sydney – Manly or Dover-Calais
  • Gamification-based solutions where ordinary people can take a picture of a ship, identify it (through computer vision) and learn more about the route it takes, associated emissions and whether it is a "green" or "brown" ship, in other worlds an emission-focused variant of ShipSpotting.com
  • Data storytelling promoting new or planned low- or zero emission ships for instance ships like Color Hybrid (diesel-electric), Kriti Future (ammonia-fuel ready), Havila Capela (hybrid cruise-ship), Tern Island or similar

You will get extra points if you manage to create solutions that give us an increased understanding the ocean’s influence on you, and your influence on the ocean. This can for example be new creative ways to create datasets (for instance image processing of satellite data to derive new datasets) or crowdsourcing data engagement/capturing by means of citizen science. This is the essence of what we call ocean literacy and an important part of the UN Decade of Ocean Science for Sustainable Development.

To solve the challenge, on June 7th we'll provide you with an access to the following datasets:

Ship particulars (dimensions, types, fuels and more) from the Ocean Data Platform

Ship emissions:

Ship voyage tables from:

Ports from:

Biodiversity

Protected areas

Satellite imagery

Relevant background material

Recommended developer tools

  • Ocean Data Platform – Ocean Data Connector (data science environment)
  • Microsoft Planetary Computer Hub
  • Azure services, e.g. PostgreSQL/PostGIS
  • QGIS and relevant plugins
  • Python
  • VS Code
  • Mapbox Studio and Tiling Services

 

Challenge by Inditex (Spain):

Energy and water consumption are key pillars of most productive sectors and powerful drivers of economic growth and social development. Virtually all economic activities require usage of any of them due to the existing intricate connection.

Water holds the key to development of energy infrastructures and remains fundamental throughout the lifecycle of energy infrastructure and resource development, while energy is itself required to make water available for human use and consumption (UNDESA, 2014). The interdependency of both is set to intensify in the coming years, with significant implications for securing these resources (IEA, 2017).

In terms of the textile industry, the vast majority of usage comes from the manufacturing tiers: the extraction of raw materials, spinning, weaving, wet processes, and final assembly of the product. Tracking consumption at each of these stages can be significantly challenging, as the nature of all of these operations is different, and the supply chain is located in a wide variety of markets with different levels of technology developments.

Nevertheless, monitoring must be one of the key starting points of any project focused on reduction. Establishing an accurate baseline is crucial to develop initiatives regarding energy and water efficiency, decrease of water pollution, reduce costs of provision, and abate greenhouse gas emissions.

There is a fair number of existing solutions to measure the utilization of the two dimensions, both on the hardware and the software/ services level. But in most cases, they have been proved to be expensive, difficult to scale and hard to control in a complex and changing environment as can be the broad network of textile supply chains.

Aim and considerations

The aim of this project would be to create an affordable, scalable, and simple solution of measuring and/ or calculating precise energy and water consumption figures of the facilities (and different processes carried out in each of them) involved in the manufacturing of garments, considering not only the potential installation/deployment costs, but also any maintenance fee related to any other stage.

Data accuracy and reliability of the information are essential, not only ensuring that the figures are obtained in a timely manner but also that they are coherent and validated so they can be used as a reference to carry out different initiatives.

Another relevant consideration will be the global approach of the project. Some important origins of production can include Bangladesh, Cambodia, China, India, Morocco, Pakistan, Portugal, Spain Turkey or Vietnam, but it can be extended to a broader variety of locations.

Additional aspects to bear in mind which would be positively valued can be:

  • A systematic life cycle perspective of consumptions, considering both direct and indirect water and energy use environmental impacts at production sites along the supply chain.
  • Development of indicators to detect critical processes and facilities and establish action plans in collaboration with the different parties involved (manufacturers, local and regional authorities…).

The utter goal of this initiative will be to support efforts to set targets, improve engagement with suppliers and simplify current data collection processes.

A few examples on what you could build or visualize during the hackathon:
  • Collaborative platform where information could be shared, while ensuring reliability and accuracy of the information (e.g through Blockchain or other technology).
  • Machine learning-based tool with forecasts and estimation based on simple parameters, previous records, type of processes
  • OCR module for consumption invoices to validate obtained information

Understanding fashion industry processes and challenges regarding energy and water consumption

  • Fashion Industry Charter For Climate Action (2020). Climate Action Playbook. Link
  • Quantis and ClimateWorks Foundation (2018). Measuring Fashion. Environmental Impact of the Global Apparel and Footwear Industries Study. Link
  • Palamutcu, S. (2010). Electric energy consumption in the cotton textile processing stages. Link
  • Radostina A Angelova et al (2021) Consumption of Electric Energy in the Production of Cotton Textiles and Garments. Link

Some existing solutions: Current tools for measurement and calculation of energy consumption, efficiency initiatives, and or GHG emissions in Fashion industry facilities

  • Energy Assessment Tools:
    • Energy Footprint Tool: Assists manufacturing, commercial and institutional facilities to track their energy consumption, factors related to energy use, and significant energy end-use.
    • The Plant Energy Profiler Excel (PEPEx) Tool: An excel based software tool to help industrial plant managers identify how energy is being purchased and consumed at their plant and identify potential energy and cost savings.
    • Energy Tracking Tool: This tool helps to track energy performance and meet energy management goals. It measures energy use, cost, and intensity, as well as greenhouse gas emissions.
  • Textile- and Apparel-Specific Tools
    • World Apparel & Footwear Life Cycle Assessment Database (WALDB)
    • Higg Facility Environmental Module (FEM): Developed by the Sustainable Apparel Coalition, it allows to assess environmental impact of product manufacturing at facilities – from water use, to waste management, to chemical and energy use – in order to identify strengths and uncover areas for improvement. The main flaw from this tool is that it is carried out once a year, not enough for the detail and periodicity needed. Other tools from the Higg initiative include the Higg Monthly Tracking
    • Water Calculation Tool for the Textile Wet Processing Sector: UNIDO and DNV GL developed a water footprint self-assessment tool to assist small and medium size textile enterprises (SMEs) in developing countries to evaluate their water footprint in restricted stages of a product life cycle.

 

Challenge by DNB (Norway):

To decarbonize our society, we need to reduce energy consumption of household and commercial buildings. According to the 2021 Global Status Report for Buildings and Construction by UNEP - UN Environment Programme, buildings constitute approximately 40 % of the Total Global Greenhouse Gas emissions (16 % in Norway).

Currently, there is incomplete national data on the actual energy use from buildings in Norway, and there is no integrated energy consumption datahub that covers the whole country. This is a challenge for DNB, as we need this data to report on the carbon footprint related to the energy use in all buildings financed by DNB.

The more specific energy use will also be important in the client dialogue as it will help DNB to compare building’s energy efficiency to assess risk and evaluate the potential energy efficiency measures. The energy label extracted from EPC (Energy Performance Certificates) is insufficient to provide the necessary information in this context due to the following:

  • Only approximately 30 % of commercial buildings and residential buildings have an EPC label; and
  • The energy label does not necessarily represent the actual energy use/efficiency as the landlord and/or tenant may use more, or less energy than the energy label reflects

The hackathon challenge consists of estimating energy consumption and emissions caused by all buildings in Norway (for both households and commercial) by using different data sources such as building year, corresponding TEK standard, size (sqm), type of building (office, etc.), geographical location, climate history, energy labels, and more. The emissions need to also consider the type of energy mix e.g., Norwegian, Nordic, or European power production as a base to calculate the emissions (CO2-equivalent) more accurately.

Furthermore, to be able to identify/sort candidate buildings that present the largest potential of benefiting from optimizing different parameters in the buildings (Typical parameters are heating, cooling, ventilation, insulation, material, smart charging, boiler, solar panel, etc.

Some starting points:

1. Use data from Table 5-12 in Sintefs report “Potensial- og barrierestudie Energitjenester i næringsbygg”

2. Leverage and combine data from Enova’s report “Byggstatistikk 2017” with the methodology reported in Graddagstall (use browser's translator for English version) and  historical climate data (EN version is available).

 
Other relevant data sources: