By Courtney Hann (NOAA Fisheries, West Coast Sustainable Fisheries Division)
Thinking back, as Leigh’s first M.Sc. student for the GEMM Lab, I wonder what poignant insight could have prepared me for my future endeavors. And having faced years of perseverance and dedication in the face of professional unknowns, perhaps the answer is none at all; fore maybe it was the many unknown challenges met that led me to where I am today.
I graduated in December of 2015, with my Masters in Marine Resource Management, and stamped completion of my research with the GEMM Lab. While my research focused on marine mammals, my broader love for the Earth’s oceans and lands guided my determination to help keep our planet’s precious ecosystem resources wild and free. So when I landed a position in terrestrial ecology after graduating, I chose to embrace the challenging decision of jumping away from theoretical research and moving back towards applied research. Consequently, I fell in love with botany, moth identification, birding, and explored the unknowns of a whole new world of conservation biology in Scotland with the Royal Society for the Protection of Birds. Not only was this work incredibly fun, interesting, and spontaneous, it offered me an opportunity to take my knowledge of developing research projects and apply it to nature reserve management. Every survey I completed and dataset I analyzed provided information required to determine the next land management steps for maximizing the conservation of rare and diverse species. From the GEMM Lab, I brought skills on: how to work through what, at times, seemed like an impassible barrier, complete tasks efficiently under a tight deadline, juggle multiple activities and obligations, and still make time to ponder the importance of seeing the bigger picture, while having fun learning new things.
For me, the long game of seeing the bigger picture has always been key. And at the end of the day, I remained steadfast in answering the questioned I posed myself: Why do all of this work if not to make a truly positive impact? With that in mind, and with an expiring visa, I moved back to the West Coast of the U.S. and landed a contracting position with NOAA Fisheries. Where I met my second female mentor, Heidi Taylor, who inspired me beyond words and introduced me to the amazing world of fisheries management. All the while, I kept working my second part-time job with the West Coast Regional Planning Body (now called the West Coast Ocean Alliance, WCOA). Working two jobs allowed me to not only accelerate my learning capacity through more opportunities, but also allowed me to extend the reach of growing a positive impact. For example, I learned about coordinating region-wide ocean management, facilitation of diverse groups, and working with tribes, states, and federal agencies while working for the WCOA. While there were moments that I struggled with overworking and fatigue, my training in graduate school to persevere really kicked in. Driven by the desire to attain a permanent position that complimented my talents and determination to provide sustained help for our Earth’s ecosystems, I worked for what sometimes felt endlessly to reach my goal. Getting there was tough, but well worth it!
One of the most challenging aspects for me was finishing my last publication for the GEMM Lab. I was no longer motivated by the research, since my career path had taken a different turn, and I was already burnt out form working overtime every week. Therefore, if it was not for Leigh’s encouraging words, the promise I made to her to complete the publication, and my other co-author’s invitation to submit a paper for a particular journal, then I likely would have thrown in the towel. I had to re-do the analysis several times, had the paper rejected once, and then ended up re-writing and re-structuring the entire paper for the final publication. In total, it took me two and half years and 100s of hours to complete this paper after graduating. Of course, there was no funding, so I felt a bit like an ongoing graduate student until the paper was finally accepted and the work complete. But the final acceptance of the paper was so sweet, and after years of uncertain challenges, a heavy weight had finally been lifted. So perhaps, if there is one piece of advice I would say to young graduate students, it is to get your work published before you graduate! I had one paper and one book chapter published before I graduated, and that made my life much easier. While I am proud for finishing the final third publication, I would have much preferred to have just taken one extra semester and finished that publication while in school. But regardless, it was completed. And in a catharsis moment, maybe the challenge of completing it taught me the determination I needed to persevere through difficult situations.
With that publication out of the way, I was able to focus more time on my career. While I no longer use R on a daily basis and do not miss the hours of searching for that one pesky bug, I do analyze, critique, and use scientific literature everyday. Moreover, the critical thinking, creative, and collaborative skills I honed in the GEMM Lab, have been and will be useful for the rest of my life. Those hours of working through complicated statistical analyses and results in Leigh’s office pay off everyday. Reading outside of work, volunteering and working second jobs, all of this I learned from graduate school. Carrying this motivation, hard work, determination, and perseverance on past graduate school was undeniably what led me to where I am today. I have landed my dream job, working for NOAA Fisheries Sustainable Fisheries Division on salmon management and policy, in my dream location, the Pacific Northwest. My work now ties directly into ongoing management and policy that shapes our oceans, conservation efforts, and fisheries management. I am grateful for all the people who have supported me along the way, with this blog post focusing on the GEMM Lab and Leigh Torres as my advisor. I hope to be a mentor and guide for others along their path, as so many have helped me along mine. Good luck to any grad student reading this now! But more than luck, carry passion and determination forward because that is what will propel you onward on your own path. Thank you GEMM Lab, it is now time for me to enjoy my new job.
Solène Derville, Entropie Lab, French National Institute for Sustainable Development (IRD – UMR Entropie), Nouméa, New Caledonia
Ph.D. student under the co-supervision of Dr. Leigh Torres
Species Distribution Models (SDM), also referred to as ecological niche models, may be defined as “a model that relates species distribution data (occurrence or abundance at known locations) with information on the environmental and/or spatial characteristics of those locations” (Elith & Leathwick, 2009). In the last couple decades, SDMs have become an indispensable part of the ecologists’ and conservationists’ toolbox. What scientist has not dreamed of being able to summarize a species’ environmental requirements and predict where and when it will occur, all in one tiny statistical model? It sounds like magic… but the short acronym “SDM” is the pretty front window of an intricate and gigantic research field that may extend way beyond the skills of a typical ecologist (even so for a graduate student like myself).
As part of my PhD thesis about the spatial ecology of humpback whales in New Caledonia, South Pacific, I was planning on producing a model to predict their distribution in the region and help spatial planning within the Natural Park of the Coral Sea. An innocent and seemingly perfectly feasible plan for a second year PhD student. To conduct this task, I had at my disposal more than 1,000 sightings recorded during dedicated surveys at sea conducted over 14 years. These numbers seem quite sufficient, considering the rarity of cetaceans and the technical challenges of studying them at sea. And there was more! The NGO Opération Cétacés also recorded over 600 sightings reported by the general public in the same time period and deployed more than 40 satellite tracking tags to follow individual whale movements. In a field where it is so hard to acquire data, it felt like I had to use it all, though I was not sure how to combine all these types of data, with their respective biases, scales and assumptions.
One important thing about SDM to remember: it is like a cracker section in a US grocery shop, there is sooooo much choice! As I reviewed the possibilities and tested various modeling approaches on my data I realized that this study might be a good opportunity to contribute to the SDM field, by conducting a comparison of various algorithms using cetacean occurrence data from multiple sources. The results of this work was just published in Diversity and Distributions:
Derville S, Torres LG, Iovan C, Garrigue C. (2018) Finding the right fit: Comparative cetacean distribution models using multiple data sources and statistical approaches. Divers Distrib. 2018;00:1–17. https://doi. org/10.1111/ddi.12782
If you are a new-comer to the SDM world, and specifically its application to the marine environment, I hope you find this interesting. If you are a seasoned SDM user, I would be very grateful to read your thoughts in the comment section! Feel free to disagree!
So what is the take-home message from this work?
There is no such thing as a “best model”; it all depends on what you want your model to be good at (the descriptive vs predictive dichotomy), and what criteria you use to define the quality of your models.
The predictive vs descriptive goal of the model: This is a tricky choice to make, yet it should be clearly identified upfront. Most times, I feel like we want our models to be decently good at both tasks… It is a risky approach to blindly follow the predictions of a complex model without questioning the meaning of the ecological relationships it fitted. On the other hand, conservation applications of models often require the production of predicted maps of species’ probability of presence or habitat suitability.
The criteria for model selection: How could we imagine that the complexity of animal behavior could be summarized in a single metric, such as the famous Akaike Information criterion (AIC) or the Area under the ROC Curve (AUC)? My study, and that of others (e.g. Elith & Graham H., 2009), emphasize the importance of looking at multiple aspects of model outputs: raw performance through various evaluation metrics (e.g. see AUCdiff; (Warren & Seifert, 2010), contribution of the variables to the model, shape of the fitted relationships through Partial Dependence Plots (PDP, Friedman, 2001), and maps of predicted habitat suitability and associated error. Spread all these lines of evidence in front of you, summarize all the metrics, add a touch of critical ecological thinking to decide on the best approach for your modeling question, and Abracadabra! You end up a bit lost in a pile of folders… But at least you assessed the quality of your work from every angle!
Cetacean SDMs often serve a conservation goal. Hence, their capacity to predict to areas / times that were not recorded in the data (which is often scarce) is paramount. This extrapolation performance may be restricted when the model relationships are overfitted, which is when you made your model fit the data so closely that you are unknowingly modeling noise rather than a real trend. Using cross-validation is a good method to prevent overfitting from happening (for a thorough review: Roberts et al., 2017). Also, my study underlines that certain algorithms inherently have a tendency to overfit. We found that Generalized Additive Models and MAXENT provided a valuable complexity trade-off to promote the best predictive performance, while minimizing overfitting. In the case of GAMs, I would like to point out the excellent documentation that exist on their use (Wood, 2017), and specifically their application to cetacean spatial ecology (Mannocci, Roberts, Miller, & Halpin, 2017; Miller, Burt, Rexstad, & Thomas, 2013; Redfern et al., 2017).
Citizen science is a promising tool to describe cetacean habitat. Indeed, we found that models of habitat suitability based on citizen science largely converged with those based on our research surveys. The main issue encountered when modeling this type of data is the absence of “effort”. Basically, we know where people observed whales, but we do not know where they haven’t… or at least not with the accuracy obtained from research survey data. However, with some information about our citizen scientists and a little deduction, there is actually a lot you can infer about opportunistic data. For instance, in New Caledonia most of the sightings were reported by professional whale-watching operators or by the general public during fishing/diving/boating day trips. Hence, citizen scientists rarely stray far from harbors and spend most of their time in the sheltered waters of the New Caledonian lagoon. This reasoning provides the sort of information that we integrated in our modeling approach to account for spatial sampling bias of citizen science data and improve the model’s predictive performance.
Many more technical aspects of SDM are brushed over in this paper (for detailed and annotated R codes of the modeling approaches, see supplementary information of our paper). There are a few that are not central to the paper, but that I think are worth sharing:
Collinearity of predictors: Have you ever found that the significance of your predictors completely changed every time you removed a variable? I have progressively come to discover how unstable a model can be because of predictor collinearity (and the uneasy feeling that comes with it …). My new motto is to ALWAYS check cross-correlation between my predictors, and do it THOROUGHLY. A few aspects that may make a big difference in the estimation of collinearity patterns are to: (1) calculate Pearson vs Spearman coefficients, (2) check correlations between the values recorded at the presence points vs over the whole study area, and (3) assess the correlations between raw environmental variables vs between transformed variables (log-transformed, etc). Though selecting variables with Pearson coefficients < 0.7 is usually a good rule (Dormann et al., 2013), I would worry of anything above 0.5, or at least keep it in mind during model interpretation.
Cross-validation: If removing 10% of my dataset greatly impacts the model results, I feel like cross-validation is critical. The concept is based on a simple assumption, if I had sampled a given population/phenomenon/system slightly differently, would I have come to the same conclusion? Cross-validation comes in many different methods, but the basic concept is to run the same model several times (number of times may depend on the size of your data set, hierarchical structure of your data, computation power of your computer, etc.) over different chunks of your data. Model performance metrics (e.g., AUC) and outputs (e.g., partial dependence plots) are than summarized on the many runs, using mean/median and standard deviation/quantiles. It is up to you how to pick these chunks, but before doing this at random I highly recommend reading Roberts et al. (2017).
The evil of the R2: I am probably not the first student to feel like what I have learned in my statistical classes at school is in practice, at best, not very useful, and at worst, dangerously misleading. Of course, I do understand that we must start somewhere, and that learning the basics of inferential statistics is a necessary step to, one day, be able to answer your one research questions. Yet, I feel like I have been carrying the “weight of the R2” for far too long before actually realizing that this metric of model performance (R2 among others) is simply not enough to trust my results. You might think that your model is robust because among the 1000 alternative models you tested, it is the one with the “best” performance (deviance explained, AIC, you name it), but the model with the best R2 will not always be the most ecologically meaningful one, or the most practical for spatial management perspectives. Overfitting is like a sword of Damocles hanging over you every time you create a statistical model All together, I sometimes trust my supervisor’s expertise and my own judgment more than an R2.
A few good websites/presentations that have helped me through my SDM journey:
Dormann, C. F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carré, G., … Lautenbach, S. (2013). Collinearity: A review of methods to deal with it and a simulation study evaluating their performance. Ecography, 36(1), 027–046. https://doi.org/10.1111/j.1600-0587.2012.07348.x
Elith, J., & Graham H., C. (2009). Do they? How do they? WHY do they differ? On ﬁnding reasons for differing performances of species distribution models . Ecography, 32(Table 1), 66–77. https://doi.org/10.1111/j.1600-0587.2008.05505.x
Elith, J., & Leathwick, J. R. (2009). Species Distribution Models: Ecological Explanation and Prediction Across Space and Time. Annual Review of Ecology, Evolution, and Systematics, 40(1), 677–697. https://doi.org/10.1146/annurev.ecolsys.110308.120159
Friedman, J. H. (2001). Greedy Function Approximation: A gradient boosting machine. The Annals of Statistics, 29(5), 1189–1232. Retrieved from http://www.jstor.org/stable/2699986
Mannocci, L., Roberts, J. J., Miller, D. L., & Halpin, P. N. (2017). Extrapolating cetacean densities to quantitatively assess human impacts on populations in the high seas. Conservation Biology, 31(3), 601–614. https://doi.org/10.1111/cobi.12856.This
Miller, D. L., Burt, M. L., Rexstad, E. A., & Thomas, L. (2013). Spatial models for distance sampling data: Recent developments and future directions. Methods in Ecology and Evolution, 4(11), 1001–1010. https://doi.org/10.1111/2041-210X.12105
Redfern, J. V., Moore, T. J., Fiedler, P. C., de Vos, A., Brownell, R. L., Forney, K. A., … Ballance, L. T. (2017). Predicting cetacean distributions in data-poor marine ecosystems. Diversity and Distributions, 23(4), 394–408. https://doi.org/10.1111/ddi.12537
Roberts, D. R., Bahn, V., Ciuti, S., Boyce, M. S., Elith, J., Guillera-Arroita, G., … Dormann, C. F. (2017). Cross-validation strategies for data with temporal, spatial, hierarchical or phylogenetic structure. Ecography, 0, 1–17. https://doi.org/10.1111/ecog.02881
Warren, D. L., & Seifert, S. N. (2010). Ecological niche modeling in Maxent: the importance of model complexity and the performance of model selection criteria. Ecological Applications, 21(2), 335–342. https://doi.org/10.1890/10-1171.1
Wood, S. N. (2017). Generalized additive models: an introduction with R (second edi). CRC press.
Solène Derville, Entropie Lab, French National Institute for Sustainable Development (IRD – UMR Entropie), Nouméa, New Caledonia
Ph.D. student under the co-supervision of Dr. Leigh Torres
Drone technology has illustrated itself as particularly useful to the study of cetacean in the GEMM Lab (see previous post by Dawn and Leila) and in the marine mammal research community in general. The last Conference on the Biology of Marine Mammals in Halifax staged several talks and posters describing the great potential of drones for observing animal behaviors, collecting blow samples, estimating the size and health of animals, or estimating densities. The GEMM Lab has been conducting leading research in this field, from capturing exceptional footages of lunge feeding blue whales in New Zealand, to measuring gray whale health on the Oregon coast.
Using drones in New Caledonia
In September 2017 I participated in a scientific cruise undertaken by Opération Cétacés /IRD to study New Caledonian humpback whales, and we were lucky to be joined by Nicolas Job, a professional diver, photographer and drone pilot. It was one of those last minute decisions: one of our crewmates canceled the week before the survey and we thought “who could we bring on instead?”. We barely knew the man but figured it would be good to get a few humpback whale drone images… We invited him to join us on the research expedition only a few days before the trip but this is not the kind of opportunity that a photographer would pass on!
Far from trying to acquire scientific data in the way the GEMM Lab does with blue whales and gray whales, we were only hoping to take “pretty pictures”… we were not disappointed.
Once we got past a few unexpected issues (YES you need to wear gloves to protect your fingers when trying to catch a flying drone (Fig 1), and NO frigate birds will not attack drones as long as they don’t smell like fish), Nicolas managed to fly the drone above our small research boat and capture footage of several humpback whale groups, including mothers with calf and competitive groups.
As I said, no groundbreaking science here, but this experience convinced me that drones can bring a new perspective to the way we observe and interpret animal behavior. As a known statistics/R-lover in the lab, I often get so excited by the intricacies of data analysis that I forget I am studying these giant, elegant, agile, and intelligent sea creatures (Fig 2). And the video clips that Nicolas put together just reminded me of that.
The clear waters of the Natural Park of the Coral Sea allowed us to see whales as far as 30 meters deep in some areas! This perspective turned our usual surface observations into 3D. We could see escorts guarding maternal females and preventing other males from approaching by producing bubble trails. Escorts also extended their pectoral fins on either side of their body, a behavior supposed to make them look more imposing in the presence of a challenger. Competitive groups were also very impressive from above. During the breeding season, competitive groups form when several males aggregate around a female and compete for it. These groups typically travel at high speed and are characterized by active surface behaviors such as tail slaps, head lunges, and bubble trails.
Drones and seamounts
Since the discovery of humpback whale offshore breeding areas in Antigonia seamount in 2007 and Orne bank in 2016, a lot of research has been conducted to better understand habitat preferences, distributions and connectivity in oceanic waters of New Caledonia (see previous post). Surveys have always been strongly multidisciplinary, including boat-based observation, biopsy sampling, and photo-identification, satellite tracking, in situ oceanographic measurements and acoustics. Will drones soon become an essential component of this toolbox?
One potential application I could imagine for my personal research questions would be to use aerial photogrammetry to measure the size of newborn calves. Indeed, we have found that offshore seamounts are used by a relatively great number of mothers with calf (Derville, Torres & Garrigue, In Press JMAMM). This finding is counter intuitive to the paradigm that maternal females prefer sheltered, shallow and coastal waters as shown in many breeding grounds around the world. Yet, we believe unsheltered oceanic areas might become more attractive to maternal females as the calf grows bigger and more robust to harsh sea states and encounters with competitive adult males. Drone photogrammetry of calves could likely help us confirm this hypothesis.
But for now, I will leave the science behind for a bit and let you enjoy the sheer beauty of this footage!
Film directed by Nicolas Job (Heos Marine) with images collected during the MARACAS3 survey (Marine Mammals of the Coral Sea: IRD/ UMR Entropie/Opération Cétacés/ Gouv.nc/ WWF/ Ministère de la Transition écologique et solidaire).
By: Solène Derville, Entropie Lab, Institute of Research for Development, Nouméa, New Caledonia (Ph.D. student under the co-supervision of Dr. Leigh Torres)
Once again the austral winter is ending, and with it ends the field season for the scientific team studying humpback whales in New Caledonia. Through my PhD, I have become as migratory as my study species so this is also the time for me to fly back to Oregon for an intense 3 months of data analysis at the GEMM Lab. But before packing, it is time for a sum-up!
In 2014, the government of New Caledonia has declared all waters of the Economic Exclusive Zone to be part of a giant marine protected area: the Natural Park of the Coral Sea. These waters are seasonally visited by a small and endangered population of humpback whales whose habitat use patterns are poorly known. Indeed, the park spans more than 1.3 million km2 and its most remote and pristine areas therefore remained pretty much unexplored in terms of cetacean presence… until recently.
In 2016, the project WHERE “Humpback Whale Habitat Exploration to improve spatial management in the natural park of the CoRal Sea” was launch by my PhD supervisor, Dr. Garrigue, and I, to conduct surveys in remote reefs, seamounts and shallow banks surrounding New Caledonia mainland. The aim of the project is to increase our understanding of habitat use and movements of humpback whales in breeding grounds over a large spatial scale and predict priority conservation areas for the park.
This season, three specific areas were targeted for survey during the MARACAS expeditions (Marine Mammals of the Coral Sea):
– Chesterfield and Bellona reefs that surround two huge 30- to 60m-deep plateaus and are located halfway between New Caledonia and Australia (Fig. 4). Considered as part of the most pristine reefs in the Coral Sea, these areas were actually identified as one of the main hotspots targeted by the 19th century commercial whaling of humpback whales in the South Pacific (Oremus and Garrigue 2014). Last year’s surveys revealed that humpback whales still visit the area, but the abundance of the population and its connection to the neighboring breeding grounds of New Caledonia and Australia is yet to establish.
– Walpole Island and Orne bank are part of the shallow areas East of the mainland of New Caledonia (Fig. 4), where several previously tagged whales were found to spend a significant amount of time. This area was explored by our survey team for the first time last year, revealing an unexpected density of humpback whales displaying signs of breeding (male songs, competitive groups) and nursing activity (females with their newborn calf).
Antigonia seamount, an offshore breeding site located South of the mainland (Fig. 4) and known for its amazingly dense congregations of humpback whales. The seamount rises from the abyssal seabed to a depth of 60 m, with no surfacing island or reef to shelter either the whales or the scientists from rough seas.
During our three cruises, we spent 37 days at-sea while a second team continued monitoring the South Lagoon breeding ground. Working with two teams at the same time, one covering the offshore breeding areas and the other monitoring the coastal long-term study site of the South Lagoon, allowed us to assess large scale movements of humpback whales within the breeding season using photo-ID matches. This piece of information is particularly important to managers, in order to efficiently protect whales both within their breeding spots, and the potential corridors between them.
So how would you study whales over such a large scale?
Well first, find a ship. A LARGE ship. It takes more than 48 hours to reach the Chesterfield reefs. The vessel needs to carry enough gas necessary to survey such an extensive region, plus the space for a dinghy big enough to conduct satellite tagging of whales. All of this could not have been possible without the Amborella, the New Caledonian governement’s vessel, and the Alis, a French oceanographic research vessel.
Second, a team needs to be multidisciplinary. Surveying remote waters is logistically challenging and financially costly, so we had to make it worth our time. This season, we combined 1) photo-identification and biopsy samplings to estimate population connectivity, 2) acoustic monitoring using moored hydrophone (one of which recorded in Antigonia for more than two months, Fig. 5), 3) transect lines to record encounter rates of humpback whales, 4) in situ oceanographic measurements, and finally 5) satellite tracking of whales using the recent SPLASH10 tags (Wildlife Computers) capable of recording dive depths in addition to geographic positions (Fig. 6).
Satellite tracks and photo-identification have already revealed some interesting results in terms of connectivity within the park and with neighboring wintering grounds.
Preliminary matching of the caudal fluke pictures captured this season and in 2016 with existing catalogues showed that the same individuals may be resighted in different regions of the Park. For instance, some of the individuals photographed in Chesterfield – Bellona, had been observed around New Caledonia mainland in previous years! This match strengthens our hypothesis of a connection between Chesterfield reef complex and New Caledonia.
Yet, because the study of whale behavior is never straightforward, one tagged whale also indicated a potential connection between Chesterfield-Bellona and Australia East coast (Fig. 6). This is the first time a humpback whale is tracked moving between New Caledonia and East Australia within a breeding season. Previous matches of fluke catalogues had shown a few exchanges between these two areas but these comparisons did not include Chesterfield. Is it possible that the Chesterfield-Bellona coral reef complex form a connecting platform between Australia and New Caledonia? The matching of our photos with those captured by our Australian colleagues who collected data at the Great Barrier Reef in 2016 and 2017 should help answer this question…
While humpback whales often appear like one of the most well documented cetacean species, it seems that there is yet a lot to discover about them!
These expeditions would not have been possible without the financial and technical support of the French Institute of Research for Development, the New Caledonian government, the French Ministère de la Transition Ecologique et Solidaire, and the World Wide Fund for Nature. And of course, many thanks to the Alis and Amborella crews, and to our great fieldwork teammates: Jennifer Allen, Claire Bonneville, Hugo Bourgogne, Guillaume Chero, Rémi Dodémont, Claire Garrigue, Nicolas Job, Romain Le Gendre, Marc Oremus, Véronique Pérard, Leena Riekkola, and Mike Williamson.
Guest Writer: Alyssa Gomez, GEMM Lab summer intern, University of Idaho, Doris Duke Conservation Scholar
Upon my arrival in Newport, OR, the sand greeted my toes, the sun my skin, and the ocean my heart. I’m an Idahoan and have yearned for the ocean my whole life, only getting glimpses of it here and there while on vacation. I have savored these memories, but for the summer of 2017, I no longer need to rely on the past. I’m only a hop, skip, and a jump away from tides and salty air until August 5th. Despite how distracting the scenery here may be, there is a lot of work to be done, as I am interning in the GEMM Lab, under the supervision of Dr. Leigh Torres, in collaboration with Craig Hayslip (Whale Telemetry Group) and Kaety Jacobson (Oregon Sea Grant).
In the short time I am here, my goal is to find out how probable it is for a gray whale (Eschrichtius robustus) to be injured to the point of scarring, and what is causing this scarring. In order to do this, I’m analyzing thousands of photos of gray whales capture in Oregon waters, which span from 2012-2016. In these thousands of images, I am identifying both anthropogenic (i.e., from fishing gear or a vessel propeller) and natural (i.e., killer whale teeth rake marks) scarring, with most focus on the anthropogenic scars. This project is collaborative, not only in terms of the data we are looking at, but in terms of who will be looking at the data. Once I’ve compiled all of the scarred whale photos, we hope to have fishermen asses the photos as well, in order to identify causes of the scars. If they believe the scars are from entanglements in fishing gear, we will ask for their opinion on the type of fishing gear that caused the scar. Hopefully, with this type of collaboration, we will be able to better understand the complex relationship between fisheries and gray whales.
While whale entanglement events are rare, Dungeness crab fishing gear is often involved. Dungeness crab is a very important fishery for this region, both economically and culturally, with a large commercial fleet and many recreational fishers. Dungeness crab pots are stationary on the sea floor, often placed in near shore waters and left out for many days in between drop off and pickup, and sometimes even abandoned altogether. Because gray whales, specifically the Pacific Coast Feeding Group of gray whales, feed in the same habitat as many Oregon commercial and recreational crab gear, they sometimes get entangled in the lines. Recently, there has been a great deal of discussion on this entanglement issue and how to maintain fishery profits while reducing entanglements. A working group of scientists, crab fishers, and gear experts met in Portland in March of 2017 to discuss this issue. Dr. Leigh Torres was in attendance, and thus, my project was born. Our goal is to identify the body regions most often involved, describe gear types if possible, and quantify healing rates of scars. We are hoping that this information will fill in some knowledge gaps and help us come up with effective solutions to this entanglement issue.
This seems like a big undertaking for me, as I’ve never been exposed to marine science, let alone marine mammals and all of the analysis programs, protocols, etc., that I am now using daily. There is certainly a learning curve; however, I have exactly the support and the freedom needed in order to prosper and learn in the GEMM Lab. Leigh, Florence, Dawn, Leila, and some honorary guests of this lab have been exceptionally welcoming and inviting, not to mention all others here at Hatfield. Each day is filled with countless new opportunities, such as dock walks, necropsies, field work, meetings, and seminars. Although I haven’t been here long, I already know that this lab is a real GEMM. I’m excited for all that is yet to come.
Solène Derville, Entropie Lab, Institute of Research for Development, Nouméa, New Caledonia (Ph.D. student under the co-supervision of Dr. Leigh Torres)
My flight back to New Caledonia gives me time to think and process all that I have experienced in the last few days. From April 4th to 6th, I had the great opportunity to attend the “Whales in a Changing Ocean” conference held in Nuku’alofa, in the Kingdom of Tonga. This conference organized by SPREP (Secretariat of the Pacific Regional Environment Programme) as part of the “Protect Pacific Whales – Ocean Voyagers” campaign brought together members of the Pacific Island governments, whale-watching operators, NGOs, IGOS and scientists.
As a relatively novice PhD student studying humpback whale spatial ecology in New Caledonia this was my first experience attending a conservation and management focused conference. To be completely honest, when I was asked to attend the conference as part of the work I am conducting on the effect of environmental changes on humpbacks of the South Pacific breeding grounds, I gladly accepted the offer, as an opportunity for me to learn more about the political mechanisms underlying international conservation plans. However, I was a little sceptical as to what tangible outputs could come out of such event. How would the science be integrated into this rather political event? How many delegations would be able to make it? Would they manage to agree on strong objectives regarding the conservation of cetaceans in the region?
On the first day of the conference, we sat through several hours of formal opening ceremony and comments from the governmental delegations that had travelled to Tonga from all over the Pacific: Samoa, Papua New Guinea, Tuvalu, Niue, French Polynesia, New Zealand, Australia, the Cook Islands, Palau, Fiji and many others. These comments mainly consisted of a succession of (well deserved) acknowledgments to the Tongan government for hosting the event and the enumeration of the endless list of threats faced by cetaceans in the region. Despite the tedious nature of this inevitable display of etiquette, I was impressed by the sight of all these governmental and non-governmental delegations sitting around the same table to discuss the future of whales. I was surprised to hear a note of emotion in several of the speeches that day. I clearly had not realized how valuable whales are to the Pacific islanders. Valuable economically of course, as whale watching is one of the most important drivers of tourism to several of these islands, most of all to Tonga. But also importantly, whales and dolphins bear a strong cultural value to the people of the Pacific. Many of the attendees shared stories and legends about whales, and I quickly realized that these animals were indeed a “cultural heritage” that people were eager to protect and preserve.
The next two days of the conference were built around a series of plenaries and workshops surrounding 3 themes: sustainable whale-watching, scientific research and emerging threats. While I was initially a bit lost and did not quite understand where all of this was going, I progressively saw several recommendations and objectives emerging from the discussions. By the end of the conference, I realized how much had been accomplished in only three days and that these achievements were more than just words. Four main outcomes resulted from this conference:
The commitment to adopt and sign a Pacific Island Year of the Whale Declaration by 11 nations/territories of the region (out of 21), namely: Australia, the Cook Islands, Fiji, New Caledonia, New Zealand, Palau, Papua New Guinea, Samoa, Tonga, Tokelau and Tuvalu. Not all of the governmental representatives were able to sign but it is likely that some will join later.
The agreement to a voluntary commitment to “Protect, conserve and restore whale populations in the Pacific islands”, which will be presented at the UN Oceans conference in June 2017.
A technical and scientific input from international working groups to help establish the next SPREP Whale and Dolphin Action Plan for 2018-2023
Tonga’s announcement of a whale sanctuary in their waters.
Whether these declarations of intentions and recommendations will actually lead to tangible actions in the short term, I could not tell. But I am glad I got the opportunity to witness the very first regional conference on whales in the Pacific Islands, and the celebration of these beautiful creatures and their place in Pacific cultures.
By Dawn Barlow, MSc Student, Department of Fisheries and Wildlife, Oregon State University
The year is rapidly coming to a close, and what a busy year it has been in the Geospatial Ecology of Marine Megafauna Lab! In 2016, our members have traveled to six continents for work (all seven if we can carry Rachael’s South African conference over from the end of 2015…), led field seasons in polar, temperate, and tropical waters, presented at international conferences, processed and analyzed data, and published results. Now winter finds us holed up in our offices in Newport, and various projects are ramping up and winding down. With all of the recent turmoil 2016 has brought, it is a nice to reflect on the good work that was accomplished over the last 12 months. In writing this, I am reminded of how grateful I am to work with this talented group of people!
The year started with a flurry of field activity from our southern hemisphere projects! Erin spent her second season on the Antarctic peninsula, where she contributed to the Palmer Station Long Term Ecological Research Project.
The New Zealand blue whale project launched a comprehensive field effort in January and February, and it was a fruitful season to say the least. The team deployed hydrophones, collected tissue biopsy and fecal samples, and observed whales feeding, racing and nursing. The data collected by the blue whale team is currently being analyzed to aid in conservation efforts of these endangered animals living in the constant presence of the oil and gas industry.
Midway atoll is home to one of the largest albatross colony in the world, and Rachael visited during the winter breeding season. In addition to deploying tracking devices to study flight heights and potential conflict with wind energy development, she became acutely aware of the hazards facing these birds, including egg predation by mice and the consumption of plastic debris.
Early summertime brought red-legged kittiwakes to the remote Pribilof Islands in Alaska to nest, and Rachael met them there to study their physiology and behavior.
As the weather warmed for us in the northern hemisphere, Solene spent the austral winter with the humpback whales on their breeding grounds in New Caledonia. Her team traveled to the Chesterfield Reefs, where they collected tissue biopsy samples and photo-IDs, and recorded the whale’s songs. But Solene studies far more than just these whales! She is thoroughly examining every piece of environmental, physical, and oceanographic data she can get her hands on in an effort to build a thorough model of humpback whale distribution and habitat use.
Summertime came to Oregon, and the gray whales returned to these coastal waters. Leigh, Leila, and Todd launched into fieldwork on the gray whale stress physiology project. The poop-scooping, drone-flying team has gotten a fair bit of press recently, follow this link to listen to more!
And while Leigh, Leila, and Todd followed the grays from the water, Florence and her team watched them from shore in Port Orford, tracking their movement and behavior. In an effort to gain a better understanding of the foraging ecology of these whales, Florence and crew also sampled their mysid prey from a trusty research kayak.
With the influx of gray whales came an influx of new and visiting GEMM Lab members, as Florence’s team of interns joined for the summer season. I was lucky enough to join this group as the lab’s newest graduate student!
Our members have presented their work to audiences far and wide. This summer Leigh, Amanda, and Florence attended the International Marine Conservation Congress, and Amanda was awarded runner-up for the best student presentation award! Erin traveled to Malaysia for the Scientific Convention on Antarctic Research, and Rachael and Leigh presented at the International Albatross and Petrel Conference in Barcelona. With assistance from Florence and Amanda, Leigh led an offshore expedition on OSU’s research vessel R/V Oceanus to teach high school students and teachers about the marine environment.
Wintertime in Newport has us tucked away indoors with our computers, cranking through analyses and writing, and dreaming about boats, islands, seabirds, and whales… Solene visited from the South Pacific this fall, and graced us with her presence and her coding expertise. It is a wonderful thing to have labmates to share ideas, frustrations, and accomplishments with.
As the year comes to a close, we have two newly-minted Masters of Science! Congratulations to Amanda and Erin on successfully defending their theses, and stay tuned for their upcoming publications!
We are looking forward to what 2017 brings for this team of marine megafauna enthusiasts. Happy holidays from the GEMM Lab!
Solène Derville, Entropie Lab, Institute of Research for Development, Nouméa, New Caledonia (Ph.D. student under the co-supervision of Dr. Leigh Torres)
If you are a follower of our blog, you may have noticed that bioinformatics and statistics hold a very important role in the everyday life of the GEMM Lab. As good-old field observations remain essential to the study of animal behaviour and ecosystems, the ecology field has greatly benefited from advances in information technologies. In fact, data analysis is now a discipline in itself, as innovative solutions must continuously be developed to cope with the challenges of ever increasing dataset size and complexity.
So how does a poor biology student find her/his way in this digital and mathematical world? Most ecology departments will provide classes to learn the basics of statistical modelling and data analysis, but there is only so much you can learn through formal education. In practice, we ultimately always run into a problem, an exception that we have never heard of, and we have to figure it out on our own. As my initial training was in fundamental biology, self-teaching of other disciplines (statistics and bioinformatics) has taken a lot of my time as a Master’s student and now as a PhD student. This has made me feel lonely and a bit lost at times when I run into challenges that always seemed too big for me. But in the end, there is nothing more rewarding then solving problems by yourself after long hours of mind-scrambling.
Oh, sorry, did I say by myself? Nothing could be more wrong and more true at the same time! Because the place where I find all the answers to my questions, is in fact born from the contribution of thousands of scientists, which, despite not actually knowing each other, all work together to develop innovative solutions to modern world scientific challenges. The internet scientific network has been my best colleague over these past years and here I would like to share my enthusiasm for some of its best features that have helped me in my research.
If you look at my Firefox toolbar you will find two types of websites: let’s call them the “practical” and the “reflectional”.
The practical websites:
These are the websites I consult if I have a specific and practical question. Many forums exist where people exchange their experiences solving a great variety of problems. But sometimes conversations get lost in never-ending exchanges of opinions, some of which are not always scientifically well-founded. On the contrary, the StackExchange platform launched in 2009 has a strict policy on how questions should be asked (as precise and focused as possible) and should be answered (in an objective, opinion-free way). This makes it a very powerful tool to find quick and practical solutions to your everyday problems. This platform includes 136 different websites, each dedicated to a different topic. In my field, I mostly use: CrossValidated for statistical issues (e.g., Why does including latitude and longitude in a GAM account for spatial autocorrelation?) and StackOverflow for programming (e.g., plotting pie graphs on map in ggplot).
The latter will usually provide you with codes in the programming language of your choice (R, python, java, sql, etc.). Interestingly, even with more queries regarding Python to StackOverflow in 2015, R was the fastest-growing language between 2013 and 2015 on this same platform. If you haven’t decided on the language you want to “speak” yet, check out this fun infographic. But always remember that these tools keep evolving…
Coming back to serious matters, it is important to recognize that you need the right key-word to access this gold-mine of website knowledge and sharing. The accuracy of your search answer will only be proportional to the quality of your question. In R for instance, if you keep googling “table” instead of “dataframe”, “list” instead of “vector”, or “size” instead of “dimensions”, you will likely get quickly drowned in the google-limbo. One way to be more efficient at your search strategy is to make sure you know your basics. Most of the programming languages used in ecology (e.g., R, Python, Matlab) share a similar vocabulary and structure, but before you start to run all sorts of crazy statistical analysis it is important to know what types of objects you are working with and how you want to format them. In R, I have found Hadley Wickham’s book, Advanced R, particularly useful to understand what happens back-stage.
Another good reference in the spatial ecology field is ZevRoss “Technical Tidbits From Spatial Analysis & Data Science”. This website is a particularly up-to-date blog for data processing and visualization in R.
More generally, I regularly check R-bloggers or simply the Comprehensive R Archive Network. A note on the latter: I know it doesn’t look pretty and the reference manuals for R packages are rather intimidating but it is still the number one reference to check when encountering a problem with a given function. Some authors make a special effort to write more user-friendly tutorials to their packages. Check for those by looking at the CRAN page of a given package, in the “downloads” section, “vignettes” subsection (e.g., for the adehabitatLT package vignette).
The reflectional websites:
The web is also an amazing media to reflect on our scientific practices, learn about current ecological theories, and acquire general knowledge across disciplines. In the scientific network, many blogs and forums exist where scientists can converse and debate ideas without the pressure of publication requirements. As a student trying to find my way in the great world of statistical modelling, I find these discussions and blogposts most useful to put my methodological choices in perspective and progressively build myself an opinion (still rather vague I’ll admit). Some of my most recent findings are: Dynamic Ecology Multa novit vulpes and From the bottom of the heap, the musings of a geographer. I am sure each of you has your own “rock star of the web”, so please share your favorite sites with us in the comments below.
Science not longer needs to wait for publication to be shared between peers and with the general public. The web offers us a new space to communicate, not only on that small part of our work that led to positive results, but also our negative results, frustrations and failures, which can at times be as informative and useful to the scientific community than our successes. So, wherever you stand, tell us about your ideas, and tell us about the challenges you have encountered, where you failed and where you succeeded. Because, this is what ecology is all about. Sharing knowledge across borders and cultures to understand the planet we live on and together take better care of it.
Solène Derville, Entropie Lab, Institute of Research for Development, Nouméa, New Caledonia (Ph.D. student under the co-supervision of Dr. Leigh Torres)
The deep ocean is awe-inspiring: vast, mysterious, and complex… I can find many adjectives to describe it, yet the immensity of it prevents me from picturing it in my mind. Landscapes are easy to imagine because we see them all the time, but their hidden ocean counterparts of seascapes with several kilometer-high seamounts and abyssal trenches are hard to visualize.
When I started a PhD on the spatial ecology of humpback whales, a species typically known for its coastal distributions, I never imagined my research would lead me to seamounts. Lesson of the day: you never know where research will lead you… So here is how it happened.
About twenty years ago when my supervisor, Dr Claire Garrigue, started working on humpback whales in New Caledonia, she was told by fishermen that humpbacks were often observed in prime fishing locations, about 170 km south of the mainland. After a little more investigation into this claim, it was discovered that these fishing spots corresponded with two seafloor topographic features: the Antigonia seamount and Torch Bank (Fig. 1), These features rise from the seafloor to depths of 30 m and 60 m respectively and are surrounded by waters about 1500 m deep. This led Dr. Garrigue to implement an ARGOS-satellite tagging program to follow the movements of humpbacks leaving the South Lagoon (one of the main breeding area in New Caledonia, Fig. 1). Sure enough, most of the tagged whales (61%) visited the Antigonia seamount (Fig. 2; Garrigue et al. 2015).
Seamounts are defined as “undersea mountains rising at least 100m from the ocean seafloor” (Staudigel et al. 2010). Most of them have a volcanic origin and the majority of them are located in the Pacific Ocean (Wessel 2001). But what is the link between these structures and marine life? The physical and biological mechanisms by which seamounts attract marine wildlife are diverse (for a review see: Pitcher et al. 2008). In a nutshell, topography of the ocean floor influences water circulation and isolated seabed features such as seamounts affect vertical mixing and create turbulences, consequently resulting in higher productivity.
For instance, have you ever heard of internal waves? Contrary to the surface waves people play in at the beach, internal waves propagate in three dimensions within the water column and can reach heights superior to a 100m! When these waves encounter steep topography, they break, similar to what a “normal” wave would do when reaching shore. This creates complex turbulence, which in turn may attract megafauna such as cetaceans (see com. by Hans van Haren).
The importance of seamounts for cetaceans is often referenced in the literature, however, few studies have tried to quantify this preference (one of which was recently published by our labmate Courtney Hann, see Hann et al. 2016 for details). So what importance do these seamounts serve for humpback whales in New Caledonia? Are they breeding grounds, do they serve as a navigation cue, a resting area, or even a foraging spot (the latter being the less likely hypothesis given that humpback whales have never been observed feeding in tropical waters)?
To answer this question, an expedition to Antigonia was organized in 2008 and about 40 groups of whales were observed in only 7 days! The density of this aggregation, the high occurrence of groups with calves and the consistent singing of males suggested that this area may be associated with breeding or calving behavior. Several other missions followed, confirming the importance of this offshore habitat for humpbacks.
Looking through all this data I was struck by two things: 1) whales were densely aggregated on top of these seamounts but were rarely found in the surrounding area (Fig. 3), and 2) other seamounts with similar characteristics are only a few kilometers from Antigonia, but seem to be rarely visited by tagged whales.
What is so special about these seamounts? Why would energetically depleted females with calves choose to aggregate in these off-shore, densely occupied and unsheltered waters?
I will spend the next two months at the GEMM lab in Newport, OR, trying to answer these questions using ocean models developed by New Caledonian local research teams (at IRD and Ifremer). I will be comparing maps of local currents and topography of several seabed features located south of the New Caledonia main island. The oceanographic model used for this study will allow me to analyze a great number of environmental variables (temperature, salinity, vertical mixing, vorticity etc.) through the water column (one layer every 10m, from 0 to 500m deep) and at a very fine spatio-temporal scale (1km and 1day, even 1 hour at specific discrete locations) to better understand humpback whale habitat preferences.
Looking forward to uncovering the mysteries of seamounts and sharing the results in December!
Garrigue C, Clapham PJ, Geyer Y, Kennedy AS, Zerbini AN (2015) Satellite tracking reveals novel migratory patterns and the importance of seamounts for endangered South Pacific Humpback Whales. R Soc Open Sci
Hann CH, Smith TD, Torres LG (2016) A sperm whale’s perspective: The importance of seasonality and seamount depth. Mar Mammal Sci:1–12
Solène Derville, EnTroPie Lab, Institute of Research for Development, Nouméa, New Caledonia (Ph.D. student under the co-supervision of Dr. Leigh Torres)
After one month and a half in the field, I am now comfortably sitting at my desk in the Institute of Research for Development (IRD) in Nouméa and I am finally finding the time to look back on my first marine mammal field experience.
The New Caledonian South Lagoon is certainly not the worst place on earth to study whales. While some people spend hours trying to spot extremely rare and shy species living in freezing cold polar waters, I have to endure a 25°C temperature, turquoise waters and a study species desperate for attention (series of a dozen breaches are not uncommon). As with all field work, there were ups and downs but following humpback whales during the 2015 breeding season was by far the most exhilarating field experience I’ve ever had.
During the austral winter, humpback whales are thought to travel and stay in different areas of the New Caledonian Economic Exclusive Zone. Using satellite telemetry, several seamounts (e.g. Antigonia), banks (e.g. Torche bank) and shallow areas have been shown to play an important role for breeding and migrating humpback whales (Garrigue et al. In Press). However, as much as we would like to study whales in these areas, offshore field missions are logistically and financially hard to conduct. This is why most of the data on humpback whales in New Caledonian waters have been collected in coastal waters, and more specifically in the South Lagoon. Opération Cétacés, a local NGO, has been studying whales in this area for about two decades and I was lucky to participate in this year’s field season with their experienced team.
The usual day in Prony (the village that we live in during the whale season) usually starts early. We get up at about 5:30, and start by engulfing a bowl of porridge (nicknamed “globi” and considered as a highly exotic dish). By 6:30 everyone is standing in our rigid-hulled inflatable boat, listening to the weather forecast on the radio. After a 15 minute trip across the bay of Prony, two people disembark and climb to a land-based lookout, the N’Doua Cape, where they will spend the day trying to spot humpback whales and guiding the boat towards their location via VHF radio communication. The vessel-based team slowly approaches the whale groups to do photo-identification (using the unique marks on the ventral surface of the tail flukes), biopsy collection, and behavioral activity monitoring. The particular coastal geography of this study area (see previous post: Crossing Latitudes) allows us to uniquely combine land-based and boat-based surveying. These methods increase our encounter rate and allow us to collect more individual-based data. Yet, compared to a standardized boat-based surveys, our survey effort is much more complex to estimate and account for in a spatial distribution model.
This season, the number of whale encounters was particularly high. We spent 31 days at sea and observed a total of 99 groups. Using photo-identification, we documented 113 different individuals, some of which were first observed more than 15 years ago! Biopsy samples were collected from 139 different individuals and we managed to record 4h of songs performed by six different whales. Given that the size of the New Caledonian population is currently thought to be less than 1000 individuals, our sampling is not too bad!
Another great achievement of this season was the tagging of two adult humpback whales with ARGOS satellite-tracking devices. It was a thrilling experience to be part of this procedure and witness the level of concentration and experience required to place a tag on a whale. Our two individuals, one a presumed male and the other a female with calf, were respectively baptized Lutèce (the name Romans gave to Paris) and Ovalie (an old fashioned way to call rugby in France). Their tags transmitted for 15 and 20 days respectively, which was not long enough to follow their migration south towards Antarctica. Yet, both whales spent time on seamounts that are known to play an important role for humpback whales in the region. We were very interested in Ovalie’s track (map given below), as she travelled along the Loyalty ridge, a seafloor structure of great interest to us. We suspect that whales could be using this ridge as a navigational aid and/or using shallow areas (seamounts and banks) along the ridge as resting or breeding habitats. The amount of humpback whales present in this area and the eventual role played by oceanic features along the Loyalty ridge will be the subject of my future research.
But now that we have all this data, let’s get back to work! As much as I love being in the field, there comes a time when you have to sit in front of your computer and try to make sense of all this information you collected.
And that is where my collaboration with the GEMM Lab comes in! I am looking forward to visiting Newport once again in December and to start shedding a light on the ‘How’s and ‘Why’s of New Caledonian humpback whales’ space use.
Garrigue, C., Clapham, P. J., Geyer, Y., Kennedy, A. S., & Zerbini, A. N. (In Press). Satellite tracking reveals novel migratory patterns and the importance of seamounts for endangered South Pacific Humpback Whales. Royal Society Open Science.