The Eastern North Pacific is a species rich area. A total of 30 marine mammals are known to occur in Oregon and Washington waters. However, the seasonal abundance and distribution of marine mammals in Oregon’s near shore waters is not well understood. The goals of my project are to use passive acoustic monitoring, visual line transects and oceanographic data collection in Newport, Oregon’s near shore waters to [A] study the temporal distribution, spatiotemporal scales of occurrence and movement patterns of marine mammals; [B] study physical, chemical and lower-trophic-level ecological drivers of these occurrence patterns, producing a quantitative model of occurrence; I am going to specifically concentrate on harbor porpoises as an indicator species. Harbor porpoises are of elevated concern because of their high sensitivity to anthropogenic noise, such as wave energy converters.
Since October of 2013, I have been using two methodologies with high spatial and temporal resolution – combined passive-acoustic and visual surveys – to effectively monitor porpoises in Newport, Oregon’s near shore waters. In addition, physical, chemical, and biological oceanographic data has been collected in-situ with during the duration of these survey methodologies. At this point, I have survey data from about 18 months of transect surveys. Data for this analysis were collected from multiple surveys at the Pacific Marine Energy Center (PMEC) North (NETS) and South (SETS) Energy Test Sites and the Newport Hydrographic Line (NH).Both Nets and Sets are near shore (within 5 miles), The NH line extends west from Newport for 25 miles, and all three are subject to up and downwelling events ubiquitous to the Oregon Coast.
There has previously been a lack of mammal distribution data in the area and therefore no reference to the data set, I have collected and created. It is expected that I will find trends with near shore distributions (NETS and SETS sights) that change seasonally with upwelling and downwelling events. Similar distribution changes are expected along our offshore NH line.
Using established statistical analysis including environmental variable mapping and spatial interpolation methods, I am hoping to quantify correlations between harbor porpoise movement patters and distributions with biophysical variables. I would also like to identify and quantify harbor porpoise “hang-outs” with habitat- association models.
During the course, I would like to first map my sighting data over a bathymetry layer in ArcGIS to see a visual representation of my collected data so far. I would like to separate my sightings by harbor porpoise and other, to immediately hope to see a trend (without statistics) of harbor porpoises being more distributed near shore than off shore since they are known to congregate on shelf breaks in less than 200 meters of water. Next, I would like to spatial interpolation methods to bin my oceanographic data to create a “fluid” picture of Newport, Oregon Oceanography. Finally, I will look for trends (and use spatial statistics) to determine what biological, physical, chemical, and lower-trophic levels drive the occurrence of porpoises. Determining what factors affect distributions can be incorporated into a habitat model to help predict when and where harbor porpoises may be in the future.
The analysis of this data set will provide needed information on harbor porpoise occurrence and behavior and an understanding of the physical and biological factors leading to these occurrences to serve as a baseline measure of mammal hot spots. Results from this study will generate data and information that can be used to answer key regulatory and impact – mitigation questions for renewable energy siting and permitting.
As far as my experience goes to do all of this, I have taken introductory classes on the statistical package R and ArcGIS, which included a couple of model builder lessons, but have never worked with python. This is really my first chance to break down my excel spreadsheet of data and upgrade into a visual representation and understanding of distributions and movements of harbor porpoises across space and time. I am looking forward to advancing my skills, struggling through the 5 stages of grief, and working with my classmates. Here’s to learning!
Amanda,
I picked your Spatial Problem because I worked as the Program Manager for Marine Mammal Monitoring at the Port of Anchorage Expansion Project. One of the species we would observe regularly was harbor porpoises. Over a couple years I noticed a pattern where the beluga whales and porpoises would move toward the pile driving (loud sheet piles being pounded into sub-strait) while active and away once it stopped. This occurred many times since as they were sighted within 300m the operation would be halted, then they would move outside the area and operations could begin again. I wondered if they were curious, even when it might be damaging to their biology or communications.
The idea I had for your distribution study is that if Newport has any areas of louder underwater acoustics that may create a draw or buffered area. Also if the photo is of you doing observations that should be noted. I am exited to see your results!
Amber
Hi Amber,
It is interesting that you saw the opposite behavior of marine mammals in relation to pile driving than we would expect. There is strong evidence of harbor porpoise avoiding renewable energy construction sites, and pile driving in particular. I am curious if the sound of pile driving was causing the animal to surface, resulting in more visual detentions. The Newport harbor is an area of high anthropogenic activity prior to any renewable energy installation, mainly fishing, resulting in an increased ship strikes and entanglement for harbor porpoise.
I have no explanation for why you saw the opposite of what I would expect as a reaction to harbor porpoise.
I’ll keep you updated. Cheers!
Amanda
Hi Amanda,
Nifty project. I bet it’s pretty exciting to observe such magnificent animals and get to know their behavior.
A few questions I have center around selecting variables for your model. Do you have any ideas on how you will select variables? Will you use a multivariate regression model? I imagine that the important variables for habitat selection vary throughout the seasons and with changes in ocean circulation, etc… How will you account for non-stationarity in your model? Will you have different models depending on the time of year?
Tullia