MeteoMobil

Overview

• Recent research is focusing extensively on building Cloud-based open-source solutions for big geospatial data analytics in the Cloud.
• Avalanches of georeferenced mobility (human, vehicles, etc.,) and micro-blogging data (e.g., tweets) are being collected and processed daily.
  • However, mobility data alone is not enough to unleash the opportunities for insightful analytics that may assist in mitigating the adverse effects of climate change.
  • For example, answering complex queries such as follows: “what are the Top-3 neighborhoods (districts, boroughs, etc.,) in Buenos Aires in Argentina in terms of vehicle mobility where the index of Particulate Matters (PM10) air-borne pollutant is greater than 40”.
  • Similar queries are necessary for emergent health-aware smart city policies.
    • For example, they can provide insights to municipality administrators so that they foster the design of future city infrastructure plans that feature citizen health as a priority.
    • For example, by restricting the number of vehicles accessing highly-polluted areas of the city during peak hours.
  • Also, such information can be used to build mobile interactive maps for daily dwellers so that to inform them which routes to avoid passing-through during specific hours of a day to avoid being subjected to high-levels of vehicle-caused air-borne pollutants (such as PM10).
• However, answering such a query would require joining real-time mobility and environment (e.g., meteorological) data.
• Stock versions of the current Cloud-based open-source geospatial management systems do not include intrinsic solutions for such scenarios.
• MeteoMobil is a system for the combined analytics of integrated information representing mobility and environment conditions.
• We have implemented our system atop Apache Spark for efficient operation over the Cloud.
• We have tested the system and shown its capabilities in supporting climate change analytics.
  • Our results show that MeteoMobil can be efficiently exploited for advanced climate change analytics

publication: presented at IEEE CAMAD 2021, To Appear