RPP18: Extracting Information from Archived Pictures for a Patch Dynamics Study in the Asturian Rocky Shore
Supervised by José Luis Acuña, Universidad de Oviedo
The PERCEBES Project is part of the European BiodivERsA network and is coordinated by José Luis Acuña at Universidad de Oviedo (Spain). Other Partners include Universidad de Vigo (Spain), Universidade de Évora (Portugal), Universidade de Aveiro (Portugal), ENSTA Bretagne (France) and Station Biologique de Roscoff of Sorbonne Université (France). [1]
The project investigates how harvesting the stalked barnacle (Pollicipes pollicipes) affects their stands in terms of productivity, connectivity and biodiversity.
For this purpose a Human Exclusion Experiment is run in several locations in Alentejo (Portugal), Atlantic Islands (Spain), Western Asturias (Spain) and South Brittany (France). Plots are marked with metal frames and either covered with a cage (= exclusion) or not (=openly accessible) and deliberately harvested by the scientists or not.
Data is collected for each plot in form of photographs, taken monthly over a 2 year period.
This information can be used to adapt spatial management strategies of P. pollicipes.
A place, where spatial management of stalked barnacle stocks has been around for more than 20 years, is the Asturian coast of Spain. Fishers, scientists and government agencies share their knwoledge and work together to grow and harvest the local delicacy in the most sustainable way possible. [3]
One important factor in maintaining a healthy barnacle stock is knowing, how their larvae are dispersed and under which conditions they grow best. The master student Berenike Dietrichs with the work group under José Luis Acuña at Universidad de Oviedo, is trying to help with that. Her goal is to create a model that can calculate where new stands of P. pollicipes will form, simply by looking at a rock's surface.
There are a number of factors involved in the successful attachment of larvae to the exposed intertidal rocky shore and their development into healthy, harvastable stocks. The main questions considered in this analysis are the following two:
Preferred Surface
P. pollicipes larvae are free floating and attach to a solid surface before growing into their sessile adult form. Usually the cypris stage will find an adult barnacle and attach right next to it. This is how a patch grows bigger.
It is however also possible for the Cypris zu attach elsewhere and form an entirely new patch.
But how does it decide, where to settle? Is it drawn to certain surfaces? Certain intertidal species?
By recording the surface characteristics of spots with new patches, we hope to find an answer to these questions.
Picture [4] (l.t.r: large rock features, cracks in rock, Mytilus edulis, Chthalaus sp., other species (e.g. Lithophyllum tortuosum and L. incrustans, Ralfsia verrucosa, Patella spp., Corallina elongata, Actinia equina, and many more)
Visibility
Now that we have information about possible surfaces, we want to know if it is possible to predict, where a new patch is most likely to form. This kind of information can be helpful with spatial managment of the stalked barnacle harvest.
This is easier said than done. Why, you ask?
The cypris larva is microscopic. It is not possible to see, when exactly it has attached to the surface, in our pictures.
We only know, when the adult barnacle was first visible. As we do not know how long it takes the barnacle to grow to a certain size, we have to assume, that the attachment could have happend anywhere within the previous year.
Therefore we analyse the surface characteristics for every picture that was take 3, 6, 9, 12, or 15 month before the adult barnacle was first visible. [4]
Editing of Pictures for Comparability
Pictures were taken at different angles. The photos had to be edited to overlap perfectly, to compare cover percentages and see differences in patch sizes. One photo was chosen as the basis and all others were Georeferenced to it in QGIS by selecting approx. 60 features that are clearly visible in both pictures.
A common quadrat (red) was determined, which is within each layer's metal frame.
Identification and Selection of P. pollicipes patches
3. Barnacle patches within the quadrat were then outlined as polygons with QGIS in each layer. Development and new growth of barnacle patches is now clearly visible.
Picture shows earliest patch area in red and latest in violet
4. Barnacles that appeared outside of patches that existed before are identified and marked with a 2x2 cm square (blue).
5. Control squares of 2x2 cm (green) are appointed randomly and analysed alongside the new patches.
Identification of Surface Characteristics
6. Different relevant factors are recorded for each new patch and control square. This includes information on the date of first appearance, location of the experimental plot, the experimental conditions (harvest/ no harves; cage/ no cage), the previous area calculations, a relative cover of established barnacle patches within the common quadrat and the presence of rock features (cracks, etc.).
7. All species and rock features visible within the 2x2 cm squares at 3, 6, 9, 12 and 15 month before first patch appearance are identified and their cover percentages roughly estimated.
How many structures can you identify? Answer in Sources [5]
8. Part of this data is used to create a mixed linear model in R. The rest is used to test the model. It is then possible to see, which of the models (3,6,9,12 or 15 month data) is the most accurate.
[1] https://www.unioviedo.es/percebes/
[2] Picture taken by Teresa Cruz, University Évora
[3] Rivera, Antonella, Gelcich S., García-Flórez L., Acuña J.L. (2017), Heterogenous management and conservation perceptions within the goosneck barnacle co-management system in Asturias (N.Spain), Marine Policy, Vol. 81, 229-235, DOI: 10.1016/j.marpol.2017.03.020
[4] Illustrations in Section "Larval Recruitment": created by Laura S. Belter
[5] The answer is: Patella spp. (top left), Ralfsia verrucosa (black overgrowth), Chthalamus spp. (round structures across whole picture) and bare rock in between.
The Headline background displays the goosneck barnacle species Lepas anatifera. The picture was take by Laura S. Belter in Áviles, Spain.
Team work!
For more information also visit the websites of IMBRSea students:
Wyndel Sañoza, Lara Brodie and Karlis Heimrats.
Thank you to José Luis Acuña, Berenike Dietrichs, Pelayo Rico Fernandez, Yago Iván Baragaño and Ignacio Paulin Montes for 6 weeks of great team work and fun meetings.
A big Thank You also to the IMBRSea coordination office for organising the remote professional practices in such a short time and working tirelessly to ensure that our education continues, as close to normal as possible.
