Jay has posted 1 posts at DZone. View Full User Profile

Java Sun SPOTS in the Rain Forests of India

  • submit to reddit
Ashoka Trust for Research in Ecology and Environment (ATREE)  is built on a strong academic foundation and actively pursues investigations in the natural and social sciences, which form the basis for ongoing projects that are dedicated to outreach and education. The Kalakad Mundanthurai Tiger Reserve (KMTR) , and the Biligiri Rangaswamy Temple Wildlife Sanctuary, two important protected areas in India’s Western Ghats, have been long-standing foci of ATREE’s work in forest ecology. They have also undertaken a pioneering canopy research program at KMTR which is opening a new world of ecological understanding.

We at the Canopee Project have identified that collecting data on these covariates could be very easy, most precise and regular if an efficient Sun SPOT system could be developed. The project specifically aims at developing a sensor solution which will be tested and deployed at KMTR. See Canopy Research for more details.

We've recenlty been to the KMTR Rain forest for a possible deployment solution using Sun SPOTs. Here is a brief account of how Sun SPOTs actually fared in these XTreme conditions!

The Existing System

As a part of the project, they need to measure various environmental variables, such as temperature, light, rainfall, soil moisture, if possible CO2 etc.

Currently the parameters are measured using hobo data loggers developed by Onset in US from various parts of the forest, that could be ground, forest canopy, disturbed forests etc. The data loggers currently give readings at 1 hour interval for light, temperature and RH:

Actual Deployment Place highlighted in Red

The Environmental Conditions:

1. Temperature Range:      8 - 38 Degree C

2. Humidity:                     15-100%

3. Rainfall Range:             1000-4500mm

Requirements Hobo Datalogger circuit

Since they do long term monitoring at the site for vegetation, phenology and soil moisture levels the equipments need to run unattended for months together. The equipments should also be small, completely water proof - usually moisture levels are 100% for a large part of the year. Many dataloggers have failed due to moisture. Hence, a reliable, moisture resistant sensor network is needed which will measure the required parameters efficiently.

They deploy the data loggers on tree tops about  22 - 25 mt high. They periodically have to climb up trees and bring the dataloggers down. So, it would be really useful if the system could send data down wirelessly if base station connected to the computer is in range.The data can then be stored in database and analyzed to generate reports. But theory sounds really good. What about practical deployment and feasibility? We analyzed many important aspects. Here is what we found.

Can Sun SPOTs help ?                                   The Data Logger Circuit

Issue I: How to protect the Sun SPOT in such hostile conditions?

                                 Box Open                                                                                    Box Closed

The environment where the actual deployment is to be made is very hostile. The main problems are:

  • Humidity
  • Wind
  • Rain
  • Animals - specially monkeys. One of the data loggers was actually chewed away and destroyed by a Nilgiri Langoor!

Sun SPOT and HOBO Together!                                                         Deployed on Tree Top

But the solution was already in place! They have developed a nice little box that keeps the data logger away from Rain and water splashings from sides. Also they lock it away so that the animals can't get their hands on it. The device is suspended from the middle and slits are provided so that the temperature and moisture are not altered. 

Issue II: The accuracy of the data?

We've developed a prototype of the probable system for testing. The Sun SPOT was kept on the tree for 1 day and programmed to take 48 readings of light and temperature at intervals of 1/2 hour each. Later it would send the data to database through basestation.

The actual result was fantastic! We placed the Sun SPOT besides the hobo data logger inside the box and the readings found were almost matching. The slight deviation of 0.5 units was completely acceptable.

Here is the actual comparison:

Readings with Data Logger:      Download Readings sheet

From ATREEPics

Readings with Sun SPOT:   Download Readings Sheet

From ATREEPics

Issue III:  Range of Sun SPOTs!

We did rigorous testing for the Range of the Sun SPOTs. And the results were fantastic again:

Height above sea level





Range Found

1250 m

60 – 65 %


Clear Ground


70 m

1250 m

60 – 65 %

40 Degrees vertically Downwards

Some obstacles like shrubs and few trees


38 m

1300 m

75 – 80 %


Clear Ground but very windy


90 m

1200 m + 22 m on tree top



Inside dense forest with thick foliage between ground and tree top.


22 m but not very clear. We had to move about to get clear range.

20 m on tree top



Inside forest with little foliage between ground and tree top.


20m clear transmission found.




 Dr T Ganesh (ATREE Scientist) and Us.                                       Last ditch Code Changes!



The project is ON!

Article Resources: 
Published at DZone with permission of its author, Jay Mahadeokar.

(Note: Opinions expressed in this article and its replies are the opinions of their respective authors and not those of DZone, Inc.)


Geertjan Wielenga replied on Wed, 2009/01/21 - 4:02pm

Brilliant article!

Harris Goldstone replied on Wed, 2009/01/21 - 4:25pm

One of the best things I've seen on JL.

Arvind Gupta replied on Thu, 2009/01/22 - 12:11am


java maverik replied on Fri, 2009/01/23 - 12:29am


 I did't understand one thing "The Sun SPOT was kept on the tree for 1 day and programmed to take 48 readings of light and temperature at intervals of 1 hour each" can somebody explain it how does this happened.


Jay Mahadeokar replied on Fri, 2009/01/23 - 1:40am in response to: java maverik

Oops! My mistake - It should be: "The Sun SPOT was kept on the tree for 1 day and programmed to take 48 readings of light and temperature at intervals of 1/2 hour each"  Anyway you can see the thing in the graph shown as Sun SPOT readings...   :)

Jay Mahadeokar replied on Fri, 2009/01/23 - 1:53am in response to: Jay Mahadeokar

Made the correction. I have also provided the Reading spreadsheets for you reference.

sanja bon replied on Thu, 2009/08/13 - 6:53pm


I have just found out about your article, I have read it and it  pretty interesting and amaizing. Since you've done all the work with two types of data loggers, I would be gratefull if you can help me with the sunSPOT loggers.

So, I'm working on same kind of scenario, Early Forest Fire detection and till  now, I have made an JAVA application  for the sunSPOTs, when they detect some parametars above threshold, they broadcast that info to the Data Base. I'm using 3 sunSPOTs, but I have problem with the sunSPOTs when I deploy the application and enable OTA command server for over-the-air  communication but the sunSPOTs don't want to work if they are not connected to an usb cable to the computer. Did you work with sunSPOTs connected to the lap-top through an usb cable?

I am most kindly asking you if you could help me with this problem.

Thank you in advance,


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.