Poster presentation: Development of Sensor-caching for Environmental GIS with GK-12 classrooms

Tags: , , ,

Fellows Chris Shing and Louis Gutierrez presented their poster at the Association of American Geographers Annual Conference held in NYC in February 2012. Their poster was part of the paper session, Geospatial Technologies in the K-12 Classroom.

Community Sensing Project – Indoor air quality in Langui

Tags: , ,

by Louis Gutierrez

I had the great pleasure of traveling to Peru this Summer for 11 days, thanks to GK-12, 3Helix and RPI. The trip was an amazing adventure, full of beautiful landscapes, exotic cuisines, and most importantly, lots of opportunities to learn.

South of Cusco, 13500 ft high up in the mountains, is situated the lively village of Langui. With a Church that’s over 200 yrs old, and a picturesque Lake that is bustling with life, this place is paradise in many ways. The freezing cold nights, and hypothermia inducing morning showers, can be sobering, and can quickly remind that, despite it’s infinite beauty, it is still a difficult environment.

Thanks to the hard work of Prof. Montoya, Univ of Colorado-Boulder, and RPI, situated at the top, looking down on Langui, is La Casa Ecológica (the Ecological House). This amazing structure represents the vision of Prof. Lupita: a house that meets fundamental needs in a safe, sustainable environmentally sounds way. The house has the following:

  • Solar Panels for Electricity
  • Trombe Walls for Heat
  • Efficient Stove which drastically reduces indoor emissions

Ultimately the goal is to make this the model house for residents in Langui.

The Problem: Indoor Air Pollution

The World Health Organization  reports that an estimated 3 billion people in the world prepare food over an open fire inside their homes( This not only serves as a means to cook food, but also as source of heat during the winter months and cold nights. Unfortunately, an estimated 1.5 million deaths are attributed to the toxic byproducts of indoor stoves, the bulk of which includes women and children.

In the last decade various organizations, NGOs, and Universities have moved to design low cost, energy efficient stoves which can drastically help to reduce indoor air pollution. However, despite these breakthroughs in home cooking and heating, communities (specifically in rural villags) have been slow to adapt this technology, and chronic disease due to indoor air pollution remains an enormous challenge around the world. In Langui, traditional cooking is done over an open fire using non-renewable fuels. Efficient stoves are specifically engineered to draw pollutants out of the house and maximize fuel usage. Usually they have a chimney and metal sheet that heats up.

Kristen Matsumura’s Project

Kristen Matsumura is a CU-Boulder Graduate Student in Civil Engineering, and her project was to look at improved stoves that were recently installed in 15 rural homes in Ayaviri. Her project was to collect feedback: including motivations for use, comprehension of the benefits of the efficient design, and the challenges with adopting the technology. She also worked to measure the performance of stoves to ensure that the provided improvements in cooking efficiency and pollution reduction were meeting established standards.

Proposed Project

To meet the challenge of lobbying rural communities to transition over to cleaner burning stoves, we propose placing the same science and a similar, but lower costing, technology in the hands of community leaders that Scientists have been relying on for their own environmental data.

Conducting a professional grade environmental study can be costly, intricate and the results can be confusing for laypeople to interpret. Moreover, limited resources, both human and technological, means that only a very small fraction of the 3 billion households relying on wood/solid fuel burning stoves will ever have the opportunity to directly measure the effects of indoor air pollution.

The goal of this project is to develop low cost, low energy, mobile environmental sensors which can be operated by persons with little to no formal training. The units upload data to a centrally hosted  web-server or locally hosted software for data analysis and visualization.

Wood and other solid fuel burning stoves have an opportunity cost, one that impacts economics, culture and health. By introducing Culturally Situated Sensors, the expectation is that community leaders will be able to conduct their own

environmental study, review the data, and ultimately make a more informed decision.

Click here to read the full project proposal.

Hardware Component

The system is built on an Arduino Mega and has support for temperature/humidity sensor, time and two of the following sensors: (dust/particle, soil moisture, CO2, Carbon Monoxide, Benzyne). A separate interface device has an lcd screen and gps. The unit is used to initialize the sensors. Information about CSCS technolgy can be found here.

The system produces a csv file which has gps coordinates, sensor types and values over a given time frame. The csv file is saved on an SD card which can be read by any device enabled with an SD reader. Click here to see the data plotted on the CSCS website.

In Langui we utilized different types of sensors, ranging from professional grade to DIY (Do It Yourself) prototypes, which measured Dust/Particles and CO. Here is a list of the sensors used:

Information about the UCB Particle Sensor can be found here.
Information about the DustTrack can be found here.

Culturally Situated Community Sensors Peru Project: Successes and Challenges

The most important aspect of the data gathered from Langui is how we can use it to improve upon the current model.

Battery Life
Battery life was surprising long, eight D batters provided of 36hrs of sensing. This far exceeded expectations, and could allow us in the future to reduce the amount of batteries needed for the device to function.

Inconsistent Data
As shown in the plots from the CSS website, the data does not seem to follow the professional grade sensors. This indicates that additional work is needed to make sure the sensors are functioning correctly. One major problem was that having multiple sensors complicates the process, so reducing the the amount of sensors can make the device more reliable.

Upload Time
It turns out that uploading 3600 tuples of data to a database stored on an RPI server takes quite a bit of time (45min). One solution for this is to reduce the amount of data points (sense less frequently), reduce the data being read or designing a new system for uploading the data.

Future Direction

The most important result from this experimentation is the new direction it has inspired.

Sensor cashing presentation by Chris Shing

eco house

eco house

Village of Langui
eco house

Culturally Situated Sensors: 0.0

Tags: , , , , , , , ,

Wednesday July 20th, 2011 marked the first trial run of the Culturally Situated Sensor Project*. The trial came in the form of a Workshop facilitated by Science and Technology Graduate Student Kirk Jalbert which took place at Shiprock, NM in partnership with Dine College and the Navajo Reservation. Thanks to the hard work of Kirk, the workshop was executed flawlessly and . The only problem that surfaced was on Tuesday, when a small error in the format for the date values caused issues with reading in the data to the web platform.

Just to give a brief overview of the Culturally Situated Sensor Project:

Hardware Component

Version 0.0
The system is built on an Arduino Mega and has support for temperature/humidity sensor, time and two
of the following sensors: (dust/particle, soil moisture, CO2, Carbon Monoxide, Benzene). A separate
interface device has an lcd screen and gps. The unit is used to initialize the sensors.

The system produces a csv file, which includes gps coordinates, sensor types and measurements over
a given time frame. The csv file is saved on an SD card, which can be read by any device enabled with
an SD reader.

Software Component
Version 0.0
The software is designed to allow for authenticated users to upload browse and upload an appropriate
csv file. The data is uploaded and then marked on a map according to gps coordinates, and sensor data
is plotted on a graph. A Beta demo can be seen at here.

Despite problems with software/hardware communication, we are all very happy with the results, and from all accounts, the workshop was a success. It offered an opportunity for students to experience GIS, and Participatory Sensing, at a more tangible, visceral level. We are grateful to Kirk Jalbert, Chris Shing and Mitch Sikapizye for all there hard work over the last few weeks. The results from the workshop can be seen here.


Within the next few weeks we will be conducting the 2nd and 3rd trials of the community Sensor Project, once again in partnership with the Navajo Reservation, and with a new partner in the community of Langui, which is located outside of of Cuzco, Peru. At the Navajo Reservation Kirk will be focusing on outdoor air quality (VOC, CO), as well as soil moisture. I will be traveling to Langui to work on indoor air quality, concentrating specifically on smoke emitted by indoor stoves (both efficient and traditional). We will be working along side Professor Montoya and her team from University Colorado Boulder.

*Name is still open for debate; other aliases: Culturally Situated Community Sensors, Community Sensors, RPI Community Sensors