Hi Adam, Good idea. It's similar to the idea I posted earlier. Larvae eating fish such as killfish, guppies and mosquito fish, can be deployed to stagnant water. It's the same technique used in Vietnam to combat Dengue epidemic.
Hi Chabota, great idea. You may want to consider Gordon McCambridge's idea on NODE: Remote Data Monitoring for Agricultural Storage and Processing. I think if you integrate with his idea you'll have a richer data to improve the estimates and recommendation for inputs. See also my post on his idea regarding using data analytics.
I think it will be good to apply analytics to both local and remote. For local it doesn't need to do all the processing on the device, given the technical limitation of NODE. The heavy analysis can be done on the server side and then send back the result to the device. So it can also factor in the weather patterns in predicting the temperature and humidity. For remote, I was thinking it will be good to capture data for the whole agricultural cycle, e.g. from quality of seeds, weather patterns, soil moisture, etc throughout the cycle to predict the quality and longevity in storage of the crops.