Open-Source Application for Species Monitoring

This building block democratizes access to cutting-edge technology, enabling scalable and cost-effective wildlife monitoring. Users can upload images or videos, and the application automatically detects and classifies species, providing actionable insights for decision-making.

  • A simple and intuitive user interface to ensure accessibility for non-technical users.
  • Documentation and training resources for users to understand and effectively utilize the application.
  • Community feedback to continually enhance the tool’s usability and features.
  • Usability is key; overly complex interfaces deter users.
  • Offering technical support and clear documentation ensures broader adoption.
  • Integration challenges included aligning the AI model’s output with user-friendly visualization tools; iterative testing was essential to resolve this.
Field Data Collection and Validation Framework

The framework ensures that the AI model is robust and generalizable across different regions and habitats. Data collected is used to test the model’s ability to recognize vulture species in diverse conditions, providing feedback for further optimization.

  • Deployment of drones and camera traps in strategic locations within reserves for optimal coverage.
  • Collaboration with local conservation teams for field logistics and data collection.
  • Consistent testing and refinement of the model based on field results to address discrepancies.
  • Having local partnerships ensures smoother field operations and enhances data collection efficiency.
  • A major challenge was dealing with low-quality or insufficient data; addressing this required setting up more camera traps in diverse locations.
AI-Powered Vulture Species Recognition Model

The purpose of this building block is to automate the traditionally manual process of wildlife monitoring. The model works by analyzing visual data, detecting the presence of vultures, and classifying them into species with high accuracy. This reduces human effort, accelerates data analysis, and ensures consistency in monitoring.

  • A high-quality, annotated dataset with diverse images representing the target species in different environments and conditions.
  • Access to computational resources (Google Colab Pro+)  for training and validating the AI model.
  • Collaboration with conservationists to validate the model’s results in field conditions.
  • Ensure the dataset is representative of real-world conditions to avoid bias in detection (e.g., lighting, angles, habitats).
  • Regular updates to the model with new data improve accuracy and adaptability.
  • Challenges include misclassifications due to overlapping species traits; having experts validate initial results is essential.
City renewal strategy

Three-levels of a Co-living community

By implementing a three-level urban renewal strategy of shared neighborhood, shared courtyard, and shared building, the scattered and limited spaces in the Fayuan Temple area are organized.A top-down communal living system is established to solve the difficulties of residents' relocation, outdated infrastructure, and the protection and utilization of historical buildings. 

Shared Neighborhood

We analyzed the activity paths of different groups of people in the Fayuansi neighborhoods. The clear functional division of different areas leads to lack of intersection among various groups, which invisibly forms social barriers and is not conducive to long-term development of the city.

In the shared neighborhood, the entire block is taken as a shared space and considered as a whole, with a management center as the core to lead the co-living system. By dispersing the public functions throughout the block, the daily life of the local residents in the block has been transformed into a distinctive experiential tourism product. Overlapping activities are happening in composite physical spaces, which generates continuous energy for the neighborhood.

Shared Courtyard

The current living condition inside the courtyards is barely acceptable. Illegal construction such as kitchens and bathrooms has become a common phenomenon and the importance of public spaces are severely downgraded. There is hardly any public spaces left other than basic transportation space. 

Most of the young generations have moved out. Among the remaining residents , elderly and young children are the majority as well as some short-term tenants. The continuous reduction of vitality among the neiborhood has become an attention-grabbing problem.

While improving the overall courtyard environment, priority is given to ensuring the basic living needs of residents. This project adopts a four-step (evaluation, repair, demolition, and addition) measure of courtyard renovation. As the number of vacated households increases, existing illegal buildings will be demolished and public service facilities such as shared kitchens and shared laundry rooms will be added. These facilities will be integrated with internal courtyards and public green spaces to form a pubic core for the neighborhood.

Shared House

The vacant houses in the Fayuan Temple district are all made up with "rooms" with various ownerships. Using "room" as a basic unit for renovation is an effective way to reduce the risk of conflicts regarding property ownerships, and also flexible units can better adapt to the living needs of different types of people. By analyzing the dimensions of the courtyards, a 10-12 square-meter room unit was determined as a standard renewal module. The combination of dual and multiple units enables the possibilities for various functions such as long-term apartment, youth hostel, and Café etc.

 

Public participation

During the 2019 Beijing International Design Week, we invited the local citizens from the nearby neighborhoods to participate in the workshop of "urban additions and subtractions" in order to collect public opinions and expectations for the renovation of Fayuan Temple district.

The followings are the conclusions that we drawn from the interviews and questionnaires collected from the workshop.

  • For the multiple choice question of “What do you want to add to the district?” , pocket parks were selected the most by 19.78% of the participants.
  • Many of the citizens complained that the space within Hutong (the Alley) was cramped and lack of greenery.
  • The inhabitants of Hutong complained that the public restrooms and markets were relatively far, it can hardly meet the daily needs.
  • Many of the inhabitants expected more communication and social space within the neighborhood.
  • A majority of the inhabitants deemed that the amusement facilities for children and fitness equipment for the public were not sufficient.

The visitors of Hutong claimed that the hotels and recreational facilities were not sufficient to support the basic tourism needs.

5. Adaptive Pathway Plan

An Adaptive Pathway Plan is a strategic framework designed to enhance resilience and adapt to long-term changes, particularly in the context of climate change. It involves identifying adaptation challenges and evaluating the effectiveness of various interventions over time. The key components include:

  • Pathways Mapping: The plan illustrates sequences of measures or investments to achieve defined objectives, allowing for adjustments as conditions change.
  • Thresholds and Tipping Points: The approach uses indicators to signal when a change in strategy is needed, ensuring flexibility in decision-making.
  • Removal of Uncertainty: The uncertainty with using climate risk prediction models for decision making has led us to use Resilience instead, therefore removing uncertainty from the decision-making process.
  • Stakeholder Engagement: Involvement of diverse stakeholders ensures that the pathways are context-sensitive and reflect local needs.

Key enabling factors include:

  • Flexibility: The plan must adapt to changing conditions and uncertainties, allowing timely adjustments as new information arises.
  • Stakeholder Engagement: Involving diverse stakeholders ensures the plan addresses various needs, fostering broader support.
  • Clear Triggers: Establishing specific signposts for when to adjust strategies enhances decision-making and responsiveness.
  • Integrated Approach: Aligning the plan with existing policies creates a cohesive strategy that is easier to implement.
  • Ongoing Monitoring: Continuous evaluation of the plan's effectiveness is crucial for informed adjustments and long-term success.

Key lessons learned include:

  • Contextual Adaptation: Tailoring the analysis to specific contexts and needs enhances effectiveness and addresses complexity.
  • Visualization Tools: Diverse visual representations, like metro maps and decision trees, improve understanding and communication of pathways.
  • Stakeholder Engagement: Involving multiple actors is crucial for addressing varied values and objectives, requiring robust governance structures to support ongoing monitoring.
  • Shared Experiences: Documenting and sharing experiences can facilitate wider adoption and application of adaptive pathways in practice.
4. Predictive Scenarios

Kassandra is a predictive system, and it does so by creating ‘scenarios’ in which key parameters are altered individually or collectively and the variation of the Resilience Index is calculated. This is done iteratively until an optimum level is reached.

In addition, the scenarios can be of two types, passive and active. Passive scenarios are those where parameters external to the system are altered, for instance climatic data, whilst active scenarios simulate actual adaptations or management strategies, such as extensive tree planting.

The scenarios are not a forecast but plausible alternative images of how the future can unfold, or, as defined by the IPCC - Intergovernmental Panel on Climate Change.

Key conditions include:

  • Flexible Parameter Adjustment: The ability to easily alter key parameters, both individually and collectively, is crucial for exploring various scenarios and their impacts on the Resilience Index.
  • Comprehensive Scenario Planning: Implementing a structured approach to scenario planning helps ensure that all relevant variables are considered in the analysis.
  • Real-Time Data Integration: Incorporating real-time data feeds allows for dynamic scenario adjustments, improving the relevance and accuracy of predictions.
  • Stakeholder Input: Involving stakeholders in defining scenarios ensures that they reflect real-world concerns and priorities, enhancing buy-in and applicability.
  • Importance of Accurate Models: Initial models that lacked precision led to unreliable scenario outcomes. Ensuring data models are validated and refined improves prediction quality.
  • Parameter Interdependencies: Altering parameters individually sometimes yielded unrealistic results. Recognizing and accounting for interdependencies among parameters enhances scenario realism.
  • Iterative Testing: Conducting iterative tests of scenarios helped identify flaws and areas for improvement. Early iterations often revealed unforeseen implications of parameter changes.
  • Stakeholder Engagement: Gathering input from stakeholders in defining scenarios was crucial. Scenarios that did not align with community concerns faced challenges in acceptance and implementation.
  • Clear Communication: Presenting scenario results clearly and visually improved understanding among stakeholders. Complex data without clear visualizations often led to confusion and misinterpretation.
3. Current Resilience Identification

In this stage Kassandra undertakes an analysis of resilience for all the entities within the Digital Twin based on twelve main Kassandra Parameters, hundreds of sub-parameters and thousands of relationships between these parameters. This highlights immediately areas where resilience might be lower and that might require urgent action.

For the successful implementation of Current Resilience Identification using Kassandra, key conditions include:

  • Comprehensive Data Collection: Gathering extensive data on the twelve main Kassandra Parameters and their sub-parameters is essential for accurate resilience analysis.
  • Robust Analytical Framework: Developing a strong analytical framework to process and interpret the complex relationships between parameters is critical for meaningful insights.
  • Integration of Diverse Data Sources: Ensuring the integration of varied data sources enhances the breadth and accuracy of the resilience assessment.

The key lessons learned during the implementation of Current Resilience Identification using Kassandra are:

  • Iterative Analysis: Initial analyses often uncovered unexpected relationships or gaps in understanding. Iterative approaches allowed for refinement and enhanced accuracy in identifying resilience factors.
  • Visualizations Aid Understanding: Effective visual representations of data relationships significantly improved stakeholder comprehension and engagement in the analysis process.
2. Digital Twin Creation

Kassandra creates or builds upon a Digital Twin of the asset to be studied that uses analysis and simulation tools to take a long-term and whole-system view of an environment.

For the successful implementation of Digital Twin Creation using Kassandra, key conditions include:

  • High-Quality Data: Accurate real-time data from various sources is essential for a reliable Digital Twin.
  • Robust Integration: Seamless integration with existing systems ensures comprehensive environmental views.
  • Interdisciplinary Collaboration: Engaging experts from diverse fields facilitates holistic modelling.
  • User Accessibility: A user-friendly platform encourages stakeholder engagement.
  • Scalability: The framework should be adaptable to future data sources and analytical needs.
  • Continuous Validation: Regularly updating the Digital Twin ensures its accuracy over time.

To avoid common pitfalls, we have found that there is a need to prioritize data quality, adopt flexible development practices, and encourage interdisciplinary collaboration.

  • Data Quality Matters: Ensuring high-quality, accurate data is critical. Inaccurate data inputs led to misleading simulations, undermining trust in the Digital Twin.
  • Iterative Development: Adopting an agile approach allowed for iterative improvements based on user feedback. Initial rigid processes led to missed opportunities for optimization.
  • Interdisciplinary Collaboration: Collaborating with experts from various fields enriched the modelling process. Attempts to work in silos often led to incomplete or unrealistic simulations.
  • Scalability Planning: Planning for scalability from the start ensured the Digital Twin could adapt to growing data and user demands without major redesigns.
  • Regular Validation: Establishing mechanisms for continuous validation helped maintain the Digital Twin’s relevance and accuracy.
1. Data Acquisition and Analysis

Kassandra is a platform designed to enhance climate change decision-making through the power of generative AI. It facilitates the acquisition and consolidation of data from various sources, such citizen engagement workshops, archive searches, surveys, or even IoT devices and urban applications, allowing for a comprehensive view of a city's environmental landscape.

  • Data Acquisition: Kassandra collects diverse data related to climate, resource usage, and urban dynamics, acting as a central hub for this information,
  • Data Transmission: The platform efficiently transmits this consolidated data to a virtual environment, making it accessible and easily understandable for decision-makers.
  • Data Analysis: By integrating with advanced analytics tools, Kassandra supports real-time insights, enabling city planners to visualize trends and make informed decisions regarding resource management.
  • Scalability: The platform’s seamless horizontal scaling allows for accommodating increasing data needs as cities grow and evolve.

The conditions crucial for enabling the success of Kassandra as a platform for climate change decision-making:

  • Data Quality: Ensuring the accuracy, consistency, and completeness of data collected from various sources.
  • Interoperability: Facilitating seamless integration between Kassandra and existing urban systems and technologies.
  • Stakeholder Engagement: Involving community members, policymakers, and experts in the decision-making process to ensure diverse perspectives are considered.

Key lessons learned during the implementation of Kassandra as a climate change decision-making platform include:

  • Importance of Data Governance: Establishing clear protocols for data collection, storage, and sharing is essential. Inadequate governance can lead to data inconsistencies and trust issues among stakeholders.
  • Iterative Development: Adopting an agile approach allowed for continuous improvement based on user feedback and changing requirements. Rigid planning often led to delays and misalignment with user needs.
  • Collaboration with Stakeholders: Engaging local communities, policymakers, and technical experts throughout the process fostered buy-in and created a more relevant tool. Initial efforts that overlooked this collaboration faced challenges in acceptance.
  • Scalability Considerations: Planning for future growth from the outset ensured that the platform could handle increasing data loads and user demands without significant overhauls.