Fish Monitoring and Counting Systems
Wild Fish Monitoring
Environmental Assessments
Marine Research
Ecological Restoration
Fish Telemetry
Fish Surgery
Species at Risk
Fisheries Studies
Research Photos
News and Media
Colleagues and Links
Rental Equipment

BRAVO Systems Blog/Update



April 24 2015  The live streaming feed at Thornbury (Node 10) is now back up and running. Fish Counts for the weekend will be uploaded Monday. Fish movement is expected to be low based on cool water temperatures.


April 22 2015  The underwater fish monitoring and counting system at Thornbury (Node 10) is currently experiencing technical difficulties. Our team is working hard to get it back up and running as quickly as possible. However, fish counts during this period will be underestimated until the problem is rectified.    


December 31 2014  Passage rates in 2014 for Rainbow Trout and Chinook Salmon were considerably lower at the Thornbury Fishway (Beaver River, Ontario) and the Denny's Dam Fishway (Saugeen River, Ontario), due to multiple confounding factors. These include below average temperatures and poor visibility resulting from increased precipitation, turbulence caused by woody debris, and mechanical malfunction (i.e. non-functioning self-cleaning mechanism). However, an actual decrease in the passage rate of Salmonids is also supported by trap data conducted by the fishway management team.


This year we have deployed six new underwater fish monitoring and counting systems. Node 12  was deployed in Scully Creek, Terrace, British Columbia, to observe migrating Pink Salmon, Sockeye Salmon and Cutthroat Trout (Node 12 Archive). Node 13, was deployed at Lake Eugenia, Ontario to monitor Sportfish and Carp populations. Significant observations reveal huge schools of adult Carp and hundreds of skinny juvenile Yellow Perch, which may be a result of the lack of food due to abundant Zebra Mussel populations (Node 13 Archive). Nodes 14-17 are deployed in the Menominee River, to monitor migrating Lake Sturgeon through the Menominee Dam in Michigan (Node 14 & 15) and the Parkmill Dam in Wisconsin (Node 16 & 17). These locations are also retrofitted with custom made PIT Tracking Antennas to track the movement of PIT tagged Lake Sturgeon. 


During the field season, the state of the Grand River was improved. Algal growth and vegetation was not as prolific as seen in previous years. This may be due to colder than average temperatures and delayed ice melt.  


September 25 2013  We have upgraded and significantly improved live streaming for node 10 and node 11 and we can support a potentially unlimited viewing audience.  Lets test how many machines we can serve to before we see all that buffering nonsense again!  We have also bridged some of the cross-platform incompatibility issues that we have faced for a long time and we can now support Ipod/Iphone and other Apple products.  We are working on Android operating systems and hopefully we will be able to provide live streaming on these machines soon.  All other nodes will be upgraded when full field testing is complete at N10 and N11.


Fish counts at Thornbury and Southampton (Ontario) continue to be a challenge as we are still working on our automated algorithms.  However, fish passage of Salmonids continue to be observed. The nature-like fishway at Thornbury is more effective at passing Salmonids than the modified pool-weir-orifice type at Southampton. 


Node 1 is currently being upgraded, but due to flood conditions in the Grand River the project has been delayed. The condition of Grand River has improved; algal growth has been noted to be  reduced compared to previous year. This may be due increased precipitation and cold winter temperatures.


July 2 2013  In 2013, we have executed a fully automated tracking project with 40 radio-tagged White Suckers in the Welland River, conducted a fish spawning survey at a diamond mine in Attiwapiskat, Ontario, with underwater cameras and standard sampling techniques, and worked with Simbiota for an assessment of fish passage structures in Brazil. 


On June 22nd, Smallmouth Bass were observed guarding their nest in the Grand River during opening of the angling season near Deer Ridge/Doon.  In addition, Walleye were angled for the first time by our sampling crew in the central Grand near Deer Ridge.  We also caught and examined White Sturgeon in the Columbia River at Astoria in Oregon/Washington.


April 24 2013  Broadband underwater video is now being filtered and re-broadcast through a CDN network to support a larger viewing audience for Node 10 and Node 11. We would like to thank Scott Kidwell and HuronMedia for assisting us with this process.


Increased precipitation have resulted in poor visibility for our underwater fish monitoring and counting systems.  Fish passage have been difficult to accurately gauge at Node 11 as hydraulic conditions in the fishway have been compromised by debris and high water velocity. 


April 4 2013  The underwater fish monitoring and counting systems for Node 10 and Node 11 have been deployed and are fully operational.  We have reduced the streaming bit-rate to accommodate multiple simultaneous viewers for our live streaming. We are beginning online experimentation with a CDN re-broadcast system that will facilitate streaming to a much wider audience than preciously possible.  Our goal is to provide full broad-band streaming from all of our BRAVO nodes within the next few months. In addition, our pilot system (Node 1) in the Grand River and Node 3 at the Queens University Biological Station will be de-commissioned and upgraded to a self-cleaning system.


March 15 2013  The Node 10 system at Thornbury is now deployed and secured with  U- Bolts to prevent movement of camera as seen in 2012. Fish passage count in 2011 and 2012 showed remarkable similarity, despite the fact that fish migration began in mid March in 2012 versus April in 2011. This year we continue to record fish passage, as well as, passive Lamprey (Silver Lamprey Sea Lamprey) movement through the Thornbury fishway as they attach to migrating Salmonids. 


October 10 2012 - Thornury and Denny's Dam Fishways
Power was disrupted for mainline repairs at Node 11 and we are now working to restore the satellite link.  We hope to have the system back online by tomorrow.  We would really like to adjust the Node 10 camera that was knocked out of place by large Salmon, but we will wait until the run subsides.  We are still able to accurately process fish passage data, but the image quality
for online streaming and archives is compromised. 


These fishway counter nodes are still works in progress.  It is not an easy task to build algorithms and processes to accurately count, identify and size-class fish using video cameras and image processing - but we are pretty good at what we do given the challenges we face.  The BRAVO system is being refined on a daily basis and data are being accurately uploaded as they are processed, but sometimes there may be a delay of several hours to several days.  We are currently limited by system resources and funding.  In addition, there are some misconceptions about why we have built and deployed this network in the first place.  It is not designed to let anglers know where and when is the best time or place to go fishing.  Just because fish pass our counting systems in the fishways does not exclude the possibility that fish fall back over the dams, hold up in head-ponds or other deep pools and many other factors that affect anglers and tourists interpretation of our data relative to their pursuit of seeing or catching fish.  Our information from this network should be considered preliminary (as usually noted) until we are able to process all the relevant data that our system produces.  It is all stored onsite, and is effectively transferred to our servers automatically under most circumstances.  The full complement of data from any one site may take several days to verify and accurately process, so please do not plan your day around our counts, because they are subject to change (increase).  Live feeds are good though, so if you see fish live, we are counting them and there are fish going through the structures that are being monitored in real-time.


Please take note - there are important and exciting system upgrades that we expect to announce in early 2013.



September 28 2012 - Denny's Dam Fishway


OK - so we finally got something good going at Denny's.  With new LED lighting, different camera placement, fish guidance with the funnel, reflective background and increased water clarity the system is finally starting to function as intended.  Improvements will continue to be made until we have achieved complete system performance.


September 25 2012 - Fishway Counts

Sometimes the data files sent from Thornbury and Denny's dam in Southampton to our server are corrupt and we are missing data here in Kitchener, although it is all saved onsite (as with all of our nodes at each location).  The counts and
species distributions will be updated after we retrieve the raw data files
from the computers onsite and by the end of the year all of the data that
we have collected from this site will be analyzed and posted online.  We
are working to resolve the corrupt file problem and it appears it is
related to the internet connection and the popularity of the feed that is
chewing up all of the bandwidth onsite.  We have several plans in the
works to help solve this and we are certain that at least one of them will
work.  We also have problems at Denny's dam that we are addressing and
will make efforts to rectify on Thursday of this week (related to water
clarity and bandwidth issues with our satellite link). 

All features of are currently functional (except water temps at Thornbury-hardware issue onsite) and php issues related to our hosting shift have been rectified.


August 30 2012 - Website upgrade


We have shifted our website to a different hosting company and in the process some features of our BRAVO network are temporarily unavailable, such as sensor data streaming and some images require modification.  We will have the website fully functional and operational within a week.


August 27 2012 - Chinook Salmon Migration at Node 10


By identifying individual fish using marks such as lamprey scars, lampreys, fin abrasion, unusual coloration, deformities, pigmentation patterns, etc, we have found that Chinook Salmon can spend multiple days inside the Thornbury fishway. We hope to start conducting telemetry experiments to find out exactly how long these fish are inside the fishway or whether they leave and return at a later date. 


March 16 2012 - Monitoring at Node 10


Node 10 is back online and counting of fish usage has resumed in time for the (unusually early) upstream migration of rainbow trout. Fish counts using our algorithms, combined with trapping at the fishway are being posted as they become available and can be viewed here.


January 18 2012 - Unusual Weather Patterns


When was the last time there was no ice on the river in the middle of winter?    I wonder how many people are thinking the same thing.  There are tulips poking out of the ground and it isn't even groundhog day yet.  Here in Southern Ontario, the air temperature is flipping back and forth across the freezing mark like Lake Huron Chinook salmon vacillate in front of our fishway cameras.  It is going to be an interesting spring and another interesting summer indeed - just some thoughts about the current state of our crazy, apparently La Nina induced, but ever so unusual winter of 2011/2012.


January 9 2012 - State of the Grand River

We have never seen the Grand River in such poor condition since we began working on it 17 years ago. There is so much construction, including two new high traffic bridges, a poorly planned and placed pedestrian bridge (Walter Bean trail bridge in Doon) that will require people to hike across a City of Kitchener owned golf course (serious public safety and liability issues are sure to occur and it is a matter of time before people are injured by golf balls), and a forced main sewage line that has been installed with great difficulty under Schneider's Creek at its confluence with the Grand River.  The river is choked full of sediment and mud.  Our Node 1 camera (online since 2005) is embedded in mud and has been for several months.  This is the second year in a row that we were unable to  document successful reproduction by black redhorse (species at risk), and sedimentation and deposition of silt has surely negatively affected endangered mollusks such as the wavy-rayed lamp mussel.  This is also the first year that the river has not been frozen over upstream from the Mannheim weir in January.  Our predictions for improved water quality and impacts on fisheries in the Grand River are dire to say the very least over the next several years.


December 21 2011 - BRAVO system update


Node 10 in the Thornbury Fishway (Beaver River) has been turned off for the winter.  The fishway was dewatered on December 6.  Monthly fish counts broken down by species and the 2011 data summary of fish counts in relation to water temperature and river level have been posted on Node 10 links.  This system as well as Node 11 we be back online around Mid March 2012.


Node 1 is a non self-cleaning system, and maintenance has been hindered by elevated water levels due to flooding in the Grand River since November.  When river levels recede, the lens will be cleaned and image quality improved.


December 10 2011 - System information updates


Node 6 in the Rock River, Wisconsin is back online after persistent electrical problems at the dam.


Fish count summaries and species composition are now online and temperature data for Node 10 have been standardized to noon local time throughout the dataset for the year. The camera has been removed from the fishway to repair the lens cleaning system that was bent by crazy salmon earlier this fall.  It will be re-installed in March 2012 when the fishway is re-opened.


The Grand river continues to be flooded and we are unable to access and clean important submerged equipment (e.g., Node 1 camera and sensors) until flow levels subside.


Node 2 full duplex PIT system remains online and fully functional after continuous operation for 26 months and counting.


November 10 2011 - System information updates


Node 11 at Denny's dam on the Saugeen River, Ontario, has been installed and tested; however we will not be streaming or analyzing data from this site until rainbow trout migration begins in March 2012.

Node 10 was repaired but the camera and sensors will be removed for the winter.  Node 9 is being upgraded and will be online as soon as conditions permit.


October 17 2011 - Self-cleaning system at Node 10 damaged (temporarily) by large fish

We will repair the self cleaning mechanism at Node 10 that was damaged by a group of large Chinook salmon in the fishway.  At the moment, only half the lens is being cleaned.  Power is out in Thornbury today, so live streaming will resume once local electrical service is restored.


October 14 2011 - Live feed buffering and bandwidth issues related to overwhelming popularity of live underwater video feeds...

On some occasions we have a lot of internet traffic at our monitoring nodes that interferes with the live stream.  The stream continues to buffer until enough data is loaded to show images.  We have had several weeks when we received around 11K hits per day at one node.  Our systems are using internet connections to send data to our servers for processing.  The public feed is a bit of a bonus, and the primary function of our fishway monitoring systems is to provide fish count and fish activity data for research and management.  If you continue to have buffering issues it may be related to user bandwidth limitations, but remember that the systems are doing their job in the background.

October 13 2011 - Fishway monitoring and fish counting systems


Our fishway monitoring and automated fish counting systems (Node 10) are not designed to indicate to anglers how many fish may be available to be caught upstream (as some of you think based on your emails).  The systems are designed to indicate levels of fish activity and timing/seasonality during periods of fish migration in the fishways.  Since the systems are at the fishway exits (upstream end) we are able to enumerate fish that leave the fishway.  Most fish pass by the system one time and are easily identified, size-classed and counted (e.g., rainbow trout).  However, some fish such as Chinook salmon move up and down within the fishway and at the fishway exit repeatedly.  Our systems detect individual identifying marks such as lamprey scars, lampreys, fin abrasion, unusual coloration, deformities, pigmentation patterns, etc so we are usually able to count fish that move back and forth across our detector and camera only once.  This is not always possible and to maintain accuracy with our counts, we provide a high and low estimate and give a range of values.  This variation at Thornbury in particular, is related to the behaviour of Chinook salmon in the fishway, and not our systems' ability to accurately count fish.

View older entries



        Please contact us with inquiries related to this blog

1998-2016+ Biotactic Inc. All rights reserved. T6