Our current project: Rhizaria in upwelling regions

UVP
UVP
UVP
Puzzle Mode / Sorting Mode

Puzzle Mode

What are we currently working on?

At the moment, we are especially interested in rhizaria, single-celled zooplankton, which are often able to conduct photosynthesis. Many Rhizaria are very fragile and difficult to investigate, but especially these fragile groups can make up a substantial part of the zooplankton biomass in the ocean. In the upwelling area off California, 81% of the zooplankton is made up by Rhizaria! We want to have a look at further upwelling areas, in order to identify the distribution of Rhizaria in those regions and to understand their possible role in the carbon transport to the deep sea.

Where did we get our data?

Our new dataset was recorded during RV Meteor cruise M148 and covers a transect from Brazil to Africa. During this transect, full depth profiles to almost 5000 m depth were conducted. Furthermore, a lot of images were collected at the African coast. In the map on the left you can see the current research area, as well as the completed regions. Each dot represents one profile. Below, you can find out where exactly the transect profiles where recorded and you can have a look at the progress of our current project.

What can you do?

We need 10 identifications for each image in order to properly evaluate the assignment of this plankton organism. From the 16 images that you evaluate, 8 are randomly picked and shown together with 8 further test images. This means that we need around 11000 identification rounds to complete this project. If everybody sorts 1000 images, we are quickly done within a few weeks.

Click here to start the puzzle mode!

Our progress

Bild

Project Overview Here you can find out where exactly our data now stem from.

Our progress

Bild
UVP
We already finished 73%! Thank you for your help!
Bild

Busy-players-statistics: This graphic shows how many people already sorted more than 5000 images and you see their progress. The colour of each bar represents the success (see scale on the right).

Bild

Statistics: This graphic shows how many people already played, and how many images everyone sorted. The colour of each bar represents the success (see scale on the right). Here we only present results until 5000 sorted images.

Bild

Review: Here you can see how many images were sorted on each day since our project started in January 2017.

Bild

Success stats: Here you can see how successful people are at the game. A total success of more than 90% is really good, but 80% is also very helpful and every good round of sorting (i.e. > 85%) is valuable for us. If you do not manage to regularly obtain more than 80%, feel free to write us an email. We would be happy to help you with further explanations.

Sorting Mode

Why did we develop the sorting mode?

We developed the sorting mode in order to be able to show you an even greater variety of images and to get a prediction from you, which category these images could belong to. While you could only confirm the prediction of our neural network in the puzzle mode, you can choose from many different categories in the sorting mode.

Where did we get our data?

We are now working on a dataset from RV Meteor cruise M136 from Peru, where we were able to collect 99 UVP profiles.

What can you do?

You can also help us by playing our sorting mode. We show you a set of images from three main categories which you can place in a fitting category.

Click here to start the sorting mode!

In order to create an account you have to log in to the puzzle mode. When you have sorted 1000 images there, you unlock a link to the sorting mode so that you can play both modes.

Click here to start the puzzle mode!

Our progress

Bild

Project Overview Here you can see where our data stems from. Each point is a profile.

Bild

Statistics: This graphic shows how many people already played and how many images everyone has sorted. The colour of each bar represents the success (see scale on the right from 0-100 percent). On the left you see those players, who have already sorted more than 5000 images, while you see those players who have sorted up to 5000 images on the right.

Bild

Review: Here you can see how many images have been sorted each day since the start of the project.