I am a post-doctoral researcher at MPIA in the ERC Origins group of Prof. Dr. Henning.
My research is focused on the detection and characterization of extrasolar planets using high-contrast imaging.
• Finding faint planet signals in noise-dominated data
I have developed a method to detect exoplanets in high-contrast imaging data, called TRAP, reminiscent of the transit lightcurve detection method.
The method presents a conceptual shift from traditional approaches to direct imaging, in which the stellar signal is removed to unveil the faint planet signal by modeling the stellar halo (stellar PSF / speckle pattern) in the image-domain.
In this new method, we do not look at individual images to model the speckles, but instead we look at the time-series of individual pixels and remove the temporal trends confounding the planet signal. This is possible because in direct imaging we use a technique called angular differential imaging (ADI), in which the data is taken in such a way that the sky rotates. Due to the changing relative position of the planet with respect to the star, the planet signal moves over the detector pixels and each detector pixel sees a (positive) "transit-like" signal, that we can predict and model, because the sky rotation is known.
Why is this approach beneficial? A big problem in the traditional image based approach is that the planet does not move much if it is very close to star even if we have 90 degrees of sky rotation. That means we cannot construct a speckle model because the planet never fully moves away from the pixels we want to determine the noise model from. It becomes very hard to distinguish stellar signal from the signal of the planet itself. In the new approach, we do not have this problem, because the temporal trends in the data are determined from non-local pixels. For this reason, we can use all of the data without excluding any frames to construct a noise model regardless of how close the planet is to the star. We can track the changes in the systematic noise in real-time improving performance significantly.
A quick overview of the method is presented in this poster presented at the Exoplanets 3 Online conference 2020.
• Obtaining reliable spectral information I have been working on adapting the SCExAO/CHARIS instrument pipeline for extracting microspectra from integral field spectrographs to work for the SPHERE-IFS instrument. The code is open-source and written in Python (see codes).
• Learning about the atmospheres of planets I have led the characterization work on the relatively cold, methane-rich planet 51 Eridani b using the VLT/SPHERE planet imaging instrument, and contributed the atmospheric analysis work to several papers on recently discovered planets (PDS 70b, Müller et al. 2018; HIP 65426b, Cheetham et al. 2019). I see myself at the interface between the data and models, finding ways to understand both the systematics of an instrument (which we need to model to even detect planets in the first place, see algorithms like TRAP described above), and the planetary atmosphere model we need to interpret the signal. We cannot fully understand one without the other. I try to find holistic approaches for modeling both simultaneously, such that the results are statistically justifiable and we produce knowledge of these systems that we can trust. As we detect fainter and fainter planets, we are permanently at the cutting edge of what we can do technologically, as well as pushing the boundaries of our knowledge of planets. I hope to contribute to both.
• Samland, M.; Brandt, T. et al,
"Adaptation of the CHARIS pipeline for SPHERE-IFS: Revisiting the spectrum of 51 Eridani b", in prep.
• Samland, M.; Bouwman, J.; Hogg, D. W.; Brandner, W.; Henning, T.; Janson, M., “TRAP: a temporal systematics model for improved direct detection of exoplanets at small angular separations”, 2021, A&A...646A..24S.
• Samland, M.; Molliere, P.; Bonnefoy, M.; Maire, A. -L.; Cantalloube, F.; Cheetham, A. C.; Mesa, D.; Gratton, R.; Biller, B. A.; Wahhaj, Z.; Bouwman, J.; Brandner, W.; Melnick, D.; Carson, J.; Janson, M.; Henning, T.; Homeier, D.; Mordasini, C.; Langlois, M.; Quanz, S. P.; van Boekel, R.; Zurlo, A.; Schlieder, J. E.; Avenhaus, H.; Beuzit, J. -L.; Boccaletti, A.; Bonavita, M.; Chauvin, G.; Claudi, R.; Cudel, M.; Desidera, S.; Feldt, M.; Fusco, T.; Galicher, R.; Kopytova, T. G.; Lagrange, A. -M.; Le Coroller, H.; Martinez, P.; Moeller-Nilsson, O.; Mouillet, D.; Mugnier, L. M.; Perrot, C.; Sevin, A.; Sissa, E.; Vigan, A.; Weber, L., “Spectral and atmospheric characterization of 51 Eridani b using VLT/SPHERE”, 2017, A&A...603A..57S.
I am often asked how I learned to read and speak Japanese fluently on my own. I wrote a guide quite some time ago to explain the methodology I followed. I am a big fan of self-study for language learning, as you can go at your own pace and do not rely on your teacher correctly balancing the progression speed of classes based on an average of the participants (who may not all share your passion and interest). If you are interested in learning the language give it shot! Remember that learning a language is not a sprint, but a marathon and therefore having fun rather than "discipline" is the way to go. Five new words a day will sum up to almost 2,000 words a year. Don't be discouraged when you're not fluent after a couple of months, but if you keep at it, you should be fluent after a couple of year (if not, you're probably doing it wrong). Also, everyone who claims that you can learn Japanese in 2 weeks is trying to sell you something. :p
The resources I used are freely available online and are linked in the guide. Good luck and have fun!