ADH Monitor App

I collaborated with forensic entomologist Lue Cuttiford to develop a Java based Web Application to aid in the conduct of experimental work.

Writing on her website, Lue describes the problem she faced:

Part of my PhD thesis work involving human remains is being carried out at a University of Texas facility around 2 hours drive away from the A&M campus. The remains need to be sampled very regularly at specific thresholds of Accumulated Degree Hours (ADH) in order to have a standard set of data that can be used for validation. As such it became necessary to try and find a way to monitor ADH in real time, so that the graduate students helping out could easily be notified of an upcoming threshold for sampling.

Accumulated Degree Hours are the number of hours passed multiplied by the temperature in Celcius, minus some baseline threshold. In the case of my research using blowflies, we know that most species of blowfly in Texas will not progress through the typical developmental cycle in temperatures below 10°C since they are warm weather species. Therefore, our baseline is 10°C. So for a typical Texas summer day, the ADH at 35°C between 2pm and 3pm would be 1 hour x (35°C-10°C) = 25. What this means is that the average larvae will have undertaken 25ADH of development within the hour passed. Over a number of days, the ADH will increase and the larvae will pass through each moult or stage of development and eventually pupate and then eclose into an adult fly. Laboratory datasets of common species are available for most of the common forensic blowfly species and these will show the mean ADH required for a species to reach each stage of development, so that field specimens may be compared. This is the most commonly accepted method for deriving time to colonisation on human remains.

Since one of the factors for ADH is temperature, devising a system of sampling at specific ADH thresholds is problematic because those sampling times are solely dependent on the weather. It is highly unlikely that accurate sampling times could be estimated in advance, even with the very best forecasting. For the purposes of my study, I needed to know exactly when the next threshold was about to arrive so that I could carry out sampling at the same ADH periods for each set of remains.

Whilst considering how to tackle this rather tricky problem, I wondered whether it would be possible to use an application to monitor local weather stations and to tell me in real time how many ADH had passed since a new donation (body) had been placed outside. I enlisted the help of a talented programmer, Dr Robin Fencott, to write the program and build an interface that myself and my assistants could use during active experiments.

The ADH Monitor runs ‘in the cloud’, autonomously collecting temperature data and computing ADH values for a set of user defined experiments. A simple interface allows researchers to enter data about the experiment using a standard web-browser. Experiment data includes the start time, baseline ADH threshold and temperature values for which an e-mail notification should be sent. These e-mail notifications inform researchers that data sampling is required at a specific set of remains. The system will continue to send out notifications until a member of the research team takes appropriate action.

Multiple researchers can log into the system simultaneously to add experiments and review the progress of those currently ongoing. In this way, the ADH Monitor acts as a remote, online and collaborative research tool which provides up-to-date status information about on-going experiments to all research team members.

In the initial prototyping stage, temperature data was collected from a data-logger at the Texas A&M San Marcos facility. The system was then expanded to collect temperature data from the publicly available Wunderground (www.wunderground.com) weather data API. Using multiple data sources provides a fail-safe should one source become unavailable, and using global weather data sources such as Wunderground allows the system to potentially calculate ADH values for any geographic location for which regular temperature information is available.

The ADH Monitor was used by a team of 3 researchers to track and provide sampling notifications for live experiments between September and November 2013. This initial success demonstrates the potential of the system, however we intend to extend this proof of concept into a scalable system which could be made publicly available to researchers working in agriculture.

The ADH Monitor runs ‘in the cloud’, autonomously collecting temperature data and computing ADH values for a set of user defined experiments. A simple interface allows researchers to enter data about the experiment using a standard web-browser. Experiment data includes the start time, baseline ADH threshold and temperature values for which an e-mail notification should be sent. These e-mail notifications inform researchers that data sampling is required at a specific set of remains. The system will continue to send out notifications until a member of the research team takes appropriate action.

Multiple researchers can log into the system simultaneously to add experiments and review the progress of those currently ongoing. In this way, the ADH Monitor acts as a remote, online and collaborative research tool which provides up-to-date status information about on-going experiments to all research team members.

In the initial prototyping stage, temperature data was collected from a data-logger at the Texas A&M San Marcos facility. The system was then expanded to collect temperature data from the publicly available Wunderground (www.wunderground.com) weather data API. Using multiple data sources provides a fail-safe should one source become unavailable, and using global weather data sources such as Wunderground allows the system to potentially calculate ADH values for any geographic location for which regular temperature information is available.

The ADH Monitor was used by a team of 3 researchers to track and provide sampling notifications for live experiments between September and November 2013. This initial success demonstrates the potential of the system, however we intend to extend this proof of concept into a scalable system which could be made publicly available to researchers working in agriculture.

Below is a screenshot of the ADH monitor, capturing vital information about each experiment and the general layout of the user interface.

 

 



Comments are closed.