Understanding Neighborhood Dynamics Through Low-cost In-situ Sensing of the Urban Troposphere

The rapid pace of global urbanization has intensified the need to understand the complex and dynamic interactions within urban systems in order to create sustainable and efficient cities and to improve quality-of-life for urban dwellers.  To understand these social, environmental and man-made systems, scientists must collect and analyze massive novel and diverse data of appropriate spatial and temporal resolution to inform individuals, communities and policy-makers and to explore and establish causal relationships between these systems.  Intensive study at the neighborhood scale offers a promising lens to more fully understand the diverse set of interactions within and across heterogeneous communities and facilitates community participation, which is necessary to fully understand problems faced by the community and connects tacit neighborhood knowledge to research outcomes. 

This paper presents initial findings from a pilot deployment of a low-cost in-situ sensor network in New York City for the purpose of understanding how physical and environmental conditions affect individual and social well-being at the neighborhood scale.  Sensor measurements include air temperature, relative humidity, barometric pressure, solar radiation, noise levels and dust particle concentrations and are combined with datasets that describe social, environmental and physical neighborhood characteristics, including socioeconomic and demographic census data and New York City administrative datasets such as tax lot records, the NYC tree census, and 311 call complaint records.  These datasets establish a baseline neighborhood profile in order to compare with other areas and to track and evaluate changes over time.

A key element in this work is the use of low-cost sensing technologies.  These technologies present an important opportunity for high resolution spatial and temporal data collection through large-scale sensor deployment, and allow for a targeted, non-invasive approach to in-situ urban sensing.  The use of these devices, however, requires a complete understanding of their limitations, including data reliability, long-term performance and calibration challenges.  While acknowledging and quantifying these limitations, we demonstrate the ability of these technologies to generate meaningful data that can inform sustainability, resiliency, and quality-of-life metrics for such issues as public health and urban micro-climate dynamics. 

This research is a part of the Quantified Community (QC), a long-term neighborhood informatics research initiative that is comprised of a network of instrumented urban neighborhoods that collect, measure, and analyze data on physical and environmental conditions and human behavior to rigorously study the interaction of poverty, health and the environment in urban neighborhoods.  For this pilot, we partnered with the Red Hook Initiative, a local social services community organization, to install the sensors and engage with the local community to provide additional volunteered data.

In this paper, we discuss the applications and use cases of distributed urban sensor networks, describe our pilot sensor deployment in Red Hook, Brooklyn and present initial analysis of sensor data used to establish a baseline understanding of neighborhood dynamics in Red Hook.  We highlight the use of low-cost devices in order to evaluate their usefulness and identify the importance of community involvement to provide tacit knowledge of the neighborhood. We describe specific use cases for high-resolution in-situ urban sensing, including characterizing the urban heat island and spatial-temporal air quality patterns, and present a framework for utilized collected data to address persistent, and fundamental, questions facing cities and urban neighborhoods.