Golden Software Blog

Helping you learn more about the latest product information, tips, tricks, techniques, and customer stories so you can visualize data and communicate results with ease.

Customer Spotlight: Dr. Eric Delmelle Uses Voxler 3D Data Visualization For Space-Time

Golden Software customers possess a broad assortment of backgrounds from earth sciences and engineering to education and politics. This vast background results in a variety of uses for Golden Software's products. Each customer uses the software in a unique way, and we are pleased to share these stories. This blog article features Dr. Eric Delmelle, a professor of GIS and Health Geography in the Department of Geography and Earth Sciences at the University of North Carolina, and his use of Voxler, among other applications, to visualize space-time patterns of human behaviors and human health issues.

Dr. Delmelle's research focuses on answering fundamental epidemiological questions where spatial and spatiotemporal methodology is a critical avenue for analysis. He uses robust geocomputational methodologies that "deepen our understanding on the dynamics of infectious and non-infectious diseases". Dr. Delmelle is dedicated to the development of new visualization techniques that detect space-time patterns at different scales and leverage state of the art computational techniques to generate predictive models that could ultimately have influence on health decisions in the public sector.

A great recent example of Dr. Delmelle's work was based around evaluating the impact of space-time patterns of dengue fever outbreaks in Cali, Colombia. Dr. Delmelle and his colleagues were recently published in the International Journal of Geographical Information Science, in an article titled, "Visualizing the impact of space-time uncertainties on dengue fever patterns". In their study, the group used Voxler's robust 3D display capabilities to visualize their results in a 3D framework, which aided in the discovery of new space-time patterns of dengue fever outbreaks and gave insight on the dynamics of vector-borne diseases.

Dengue fever is transmitted to humans by mosquitos in warm climates, often causing severe outbreaks in an area's populations and considered a serious health problem for problematic areas like Cali, Columbia. In Dr. Delmelle's research, fever cases were collected using georeferenced locations during an epidemic in 2010. Data clusters were generated from the collected cases over a 6-month period and visualized by Dr. Delmelle using a space-time kernel density estimation technique in Voxler. In the images below, Dr. Delmelle models the space-time kernel density estimation in Voxler of geocoded dengue fever cases reported in the 6-month study period with reference to the geographic area of study. The density of fever cases is rendered using VolRender modules and the highest density cases are highlighted in purple. Isosurface module "shells" are generated to delineate the highest densities of the reported cases across Cali.

Voxler Isosurface Map - Dengue Fever Outbreak

Dr. Delmelle's 3D results of the space-time kernel density estimation of dengue fever outbreaks in Cali, Columbia. The highest density values are shown in purple.

The 3D visualization of the findings of Dr. Delmelle's research in Voxler resulted in being beneficial in a number of different ways. First, by mapping the 3D shape of the clusters, the dynamics of the dengue fever outbreak over time become more clear, which leads to a better understanding of the geographical pattern of the fever outbreak in terms of resurgence and eradication. The geocoding and temporal visualization of the fever data can "provide a better confidence for public health managers to decide where and when to allocate resources".

We are pleased Dr. Delmelle has integrated Voxler into his development of new visualization techniques for space-time data, and it's exciting to see his application of Voxler's tools in his work. You can find links to Dr. Delmelle's work in "Visualizing the impact of space-time uncertainties on dengue fever patterns" in the International Journal of Geographical Information Science, 28(5), 1107-1127.



No comments made yet. Be the first to submit a comment
Tuesday, 17 January 2017

Captcha Image

Keep Reading

04 January 2017
Company News

The new year is here! 2017 is a book of clean, blank pages waiting to be filled and brimming with potential.

With the beginning of the new year, I, like many of you, set aside time for reflection. It is easy for me to forget just how much was accomplished over the past 365 days when I’m in the midst of projects and tasks of my own. As I look back on 2016, I am pleased with just how much Golden Software has accomplished.

This isn’t to say 2016 wasn’t without hardship. We felt the effects of industry downturns. We had to make difficult cuts and evaluate where every penny was spent. We sidelined projects to focus on endeavors that would bring the most value to customers. Through it all, Golden Software is much stronger and better off.

Looking ahead, 2017 is gearing up to be an exciting year. The accomplishments of 2016 have set the bar high, but I am confident we will keep the momentum going.

What won’t change this coming year is our continued focus on delivering valuable features, improvements, and support to our customers. This single purpose guides all Golden Software projects.

What would you like to see from Golden Software in 2017? Please share your thoughts! I welcome your feedback.

01 December 2015
Real Life Applications

We recently received an update on the case of the missing diver. The blog has been rewritten to reflect this new information.

November 20, 2011 - Four recreational divers went looking for wreckages at the bottom of De Nieuwe Meer in Amsterdam, Netherlands, an approximately 30 meter deep lake. The group split into two pairs and set out on their adventure. They were well-equipped and all wore full face diving masks.

Around 11:30 AM, an emergency call was made by the diving group to report a fellow diver was missing. The woman diver, who was paired up with the missing 55 year old man from Amstelveen, reported looking back for him, but all she could see was a cloud of sediment. The two had just climbed over a 6 foot cooling water pipe. They were aware of the pipe, which was used to suck up water from the lake to cool several city buildings, as they had crossed it during previous dives. Her diving partner never surfaced.

Following the call, emergency responders arrived on scene including three different fire departments, the police, and ambulance service. The fire department immediately deployed divers and boats to locate the missing diver. Thereafter, the police deployed cadaver dogs who were trained to locate people beneath bodies of water.

Rescuers operating the sidescan SONAR. Picture from AT5.

According to Dutch protocol, the first hour a diver is missing, the "Golden Hour," is the most critical as chances are greater that the person will be found alive. Thereafter, chances of a lifesaving operation diminish. For this search and rescue, the Golden Hour was extended because the diver was wearing a full face diving mask, and his oxygen tank had approximately 3 hours of dive time. Even if the diver was found unconscious, the full face mask should prevent water from entering the lungs and should keep the oxygen flowing.

Due to dense fog, the search was called off on the first day; however, the other two male divers who were part of the original group, continued to search for their missing friend, but to no avail. The following day, National Police divers and specially trained cadaver dogs joined the search. Due to the cold water temperatures and the length of time the diver was underwater, it was unlikely he would be found alive. Another special police task force joined the retrieval mission and came prepared with side scan sonar to collect data of the lake bottom. On the third, fourth and fifth days of the search, the Royal Dutch Navy brought in autonomous underwater vehicles (AUVs) but, even after these efforts, was unable to locate the body.

On November 25th, the site was opened to any public entity wishing to participate in the search. It was at this point when Henk de Vries and the former Metaldec Survey BV entered the picture. Metaldec Survey used Golden Software products Didger, Voxler, and Surfer for a wide variety of projects including visualization work for several police departments. They had previously helped the police and fire departments to set protocols for forensic drowning scenes and were often known to volunteer their efforts for various police projects.

At the request of private police agents, Mr. de Vries and his team got to work. The original diving group had been equipped with devices that took depth recordings every 30 seconds during their dive. Mr. de Vries used this data, along with echo sounding data, and created a 2D contour map and 3D depth map of the lake with the help of Didger and Surfer. These maps were used to identify the dive path taken by the missing man to determine approximate location where the diver had gone missing.

Thereafter, a higher resolution bathymetric map was created by the Metaldec Survey team. In this refined area, Metaldec set up their side scan and Echo sounder sonar equipment to collect further data. The side scan and bathymetric map played a crucial role in the search endeavours as SAR divers could “see” the lake’s bottom. They were then able to safely plan their dive by using the given depths as detailed on the bathymetry map and address any dangers such as barbed wire, cables, or fishing nets prior to the dive. All dives that utilize a bathymetry map prior to the dive can save lives as it reduces the chance of decompression sickness, also known as the bends, since divers can avoid unexpected and drastic dive ascents and descents and can also calculate the maximum bottom time.


Additionally, the Sonar Tow-Fish, an expensive tool used to collect sonar data, needed the bathymetric information so it could be towed 20 feet above the bottom of the lake. Without the bathymetric map, it would have been almost impossible to coordinate when the Tow-Fish should be raised or lowered depending on the lake bottom.


3D Surface Map of De Nieuwe Meer created with Surfer

Mr. de Vries overlaid his sonar and bathymetry data on a Google Earth image of the lake and made a surprising, and concerning, discovery. The pipeline running through the middle of the lake did not cross the entire lake as everyone had assumed. Instead, it stopped at an inlet near the center of the lake. No previous divers had ever mentioned this, but it was assumed all SAR divers could be in danger of being sucked into the pipe. All diving was immediately halted.


Sonar images overlain on Google Earth image of De Nieuwe Meer

Interestingly, this inlet area was near a location where the SAR dogs were alarmed and where the original diving group believed their fellow diver disappeared. It was initially thought the man was sucked into the 6 foot wide pipe inlet. After further analysis of the sonar images and discussions with the pipe owner, it became clear the inlet was closed off by a wire cap. The inlet actually consisted of 6 upright pipes that were capped with wire meshes to prevent debris from clogging the pipe. It was then assumed the diver became stuck on the pipe or on one of the wire caps.

The lake was dredged to no avail. On January 6, 2012 the search was called off permanently, as no body had been found.

A year and a half later, on July 6, 2013, a local cadaver dog foundation searched again around the pipe inlet. As before, the dogs became alarmed in this vicinity. A drop camera (video camera connected to a monitor on a boat) was lowered, and an image that appeared to be a human face was spotted. The following week, the Mayor of Amstelveen saw the video images and asked the Dutch Royal Navy to revisit the site and inspect the new findings.

A few days later the Navy arrived on the scene and searched again with their AUV sonar. As before, the bathymetric Surfer map and side scan sonar mosaic helped the Navy program their AUV and prepare individuals for another dive. The AUV data didn’t initially show the body, but later that day, the missing diver was found by the Navy and his body was successfully recovered.

An autopsy was unable to reveal the cause of death. Instead, the man’s dive computer, used to track his diving depth intervals, was used to piece together the final minutes of his life. The data showed the man had ascended to the surface of the lake then descended back to the bottom. Five minutes after he went missing, his computer showed no movement of any kind. The man most likely died of heart failure or a stroke. This hypothesis was supported by the fact his oxygen tank was half-full and his diving gear was completely intact.

The case of the missing diver became one of the largest missing person cases in Holland. Thanks to the Royal Navy, police and fire departments, detective agencies, and many volunteer groups and individuals, the diver was finally laid to rest.

Henk De Vries now works for RemoSens BV (short for remote sensing) where he and his team continue to utilize Golden Software products to create a wide variety of maps and models.

To read more about this case, check out the featured article in GeoConnexion International Magazine:

1968 Hits 3 Comments Read More

Subscribe To Our Blog

Most Popular

This week many users experienced a GsDraw error (1): GenericError. This error began occurring follow...
This week's new feature series is a tutorial on how to use the new coordinate system dialogs in MapV...
Over the years, one of the most common questions asked is “How can I get my contour map out of Surfe...

Exceeding expectations

Go to top