Skip to content
Planet 45

Cool Earths: Discovering exoplanets beyond the snow-line with gravitational microlensing

Principal Investigators

  • Prof. Dr. Joachim Wambsganss
    Ruprecht-Karls-Universität Heidelberg, Heidelberg


The prolific discoveries of planets orbiting distant stars over the past two decades have had a radical effect on our understanding of planetary systems. Current formation theories postulate that proto-planetary cores form in dusty accretion discs surrounding the host star. These proto-planets co-evolve with the disk and may undergo orbital decay due to torque asymmetries in the surrounding disk material.
The snow-line defines the distance from a star beyond which the disk temperature drops low enough that liquid water turns to ice. Theory predicts that beyond the snow-line, the formation of ice grains allows planetary embryos to develop sufficiently massive solid cores and gradually grow by accreting material from the surrounding gaseous disk, transforming them into gas giants. Should the available disk material be depleted during this process, planetary growth stops, leaving behind smaller bodies. Simulations suggest that such bodies should exist in abundance, yet they lie beyond the sensitivity limits of most planet-finding techniques.
Besides the tantalizing possibility of discovering cold Earths, discovering these planets is critical for improving our understanding of the physical processes that drive planetary formation. The gravitational microlensing technique provides a unique window at and beyond the snow-line, allowing us to probe their distribution.
This project represents an international effort to detect distant exo-planets beyond the snow-line via robotic observations of microlensing events towards the Galactic Bulge. One of its central aspects is the extensive use of advanced target selection and prioritization systems that optimize observations over the global robotic telescope network of the Las Cumbres Observatory.
Through intensive monitoring of the light curves of microlensing events it is possible to identify the tell-tale signs of cold planets in 1-10 AU orbits around late-type stars, with a detection sensitivity down to about one Earth-mass. To achieve this, our team has pioneered the development of adaptive robotic scheduling of microlensing targets by employing a prioritization algorithm that continuously updates the list of observable microlensing targets, giving higher priority to those more likely to reveal a planetary signal. The algorithm calculates the optimal sampling frequency for each microlensing in order to maximize the planet detection probability.
Microlensing is the fastest and most cost-efficient way to probe and understand the population of cold exoplanets as it relies on observations that are directly related to the mass of the planetary system and not its brightness. It can therefore detect planets around extremely faint or unseen stars and is uniquely sensitive to planetary systems at distances of several kilo-parsec, allowing us to get an unbiased glimpse of the true Galactic population of planets.


Yiannis Taspras, M. Hundertmark, R. Street, E. Bachelet, K. Horne, M. Dominik, V. Bozza, A. Cassan, Shude Mao, E. Euteneuer


Planetary Microlensing and its Challenges
– by Esther Euteneuer

“The search for extrasolar planets is a necessary part of ongoing efforts to improve our understanding of the physical processes that drive planet formation and a first step in the search for signs of life on planets other than the Earth.”

A planet’s gravity can perturb the light rays of a more distant source star on their way towards us, temporarily changing the source star’s brightness. By detecting and analyzing these brightness changes we are able to find and characterize extrasolar planets. This method is called planetary microlensing. 

To determine the characteristics of a planet detected with microlensing, we construct several model light curves, each representing a certain set of planet properties. These models are then compared to the data to find the model that best represents them.

After spending my bachelor’s thesis on the transit method of planet detection, I wanted to really challenge myself in my master’s thesis. So I started to study the detection concepts, fitting algorithms and modelling degeneracies involved in planetary microlensing.

Knowing it would be a challenge, I was still surprised to learn that it can take several years to fully analyze a single microlensing light curve. Luckily, my research group had already begun  the analysis of a microlensing event, giving me the opportunity to generate and test my own models and contribute important work in the final  publication. 

Comparing a large number of models to the data can be very time consuming. At the same time, the amount of data gathered by modern telescopes is increasing due to continuously improving telescope hardware and survey strategies. This imbalance between the quantity of available data and the efficiency of current analysis techniques calls for further development and optimization of existing techniques.
The NASA WFIRST space mission is expected to discover thousands of microlensing planets. In order to prepare for the difficult job of identifying planetary signals correctly, they have initiated a series of data challenges. Artificial planetary signals have been generated and injected in the test data. This data set is then shared with the community and the challenge is to identify the planets and extract their parameters. The software development encouraged by this challenge provides us with tools to automatically and efficiently classify and analyze microlensing events. Participating in the first challenge in 2018, not only did I get the chance to work with an international research team and visit New York as part of the 23rd International Microlensing Conference, but I was able to directly contribute to the field’s forefront of research. 

News Item: The ROME/REA team are preparing for the first data release (DR1) of the ROME/REA project which contains photometric measurements in three observing bands (SDSS-g’,r’,i’) over three consecutive years for approximately 4 million stars in the Galactic bulge.

Invited Guests

E. Bachelet

To top