Joining our lab at University of Arizona


Fall 2021

The Tellman Lab is hiring 5 (undergrad or grad) research assistants for Fall 2021!

The Tellman Lab has four NASA-funded projects using machine learning on satellite imagery. This semester we will be building two models; one to identify clandestine roads and airstrips in Central America’s protected areas and another to detect floods in urban areas around the world. We are looking to bring on research assistants for labeling for machine learning 10 hours per week (paid $17/hour) and python programming (10-20 hours per week, flexible, paid $19/hour) for the fall semester (starting August 27th) to support our work. In this job, you will work closely with Dr. Beth Tellman, post-docs Dr. Rohit Mukherjee and Dr. Jonathan Giezendanner, and PhD student Hannah Friedrich to generate validation data sets and help run deep learning models.

Machine Learning Research Assistant Position
Good machine learning models require good training data labels from high resolution imagery. We will train you to generate validation and training data by identifying (and hand labeling, on a computer screen) infrastructure and flood water from a commercial satellite called Planetscope. This job can be done either in our lab on our computers, or remotely on your own laptop. Students who wish to get involved in resulting publications from this work will be given the opportunity to co-author papers with our lab. We will train you on all skills needed for this job, but candidates with experience in GIS, remote sensing, and upperclassmen and women (seniors and juniors) will be given priority. Graduate students are also welcome to apply.

Python Programming Research Assistant Position
You will help write and run python scripts to curate large datasets of labeled images. Later in the semester, you will help run code to train deep learning networks on these datasets. We will train you to process satellite imagery using python libraries (rasterio) and geospatial software (QGIS), and write code in Fastai (or PyTorch) to run the models on our server and the university HPC (high performance computing). Experience coding in python is required for this position. Coursework or experience with remote sensing, geospatial data, or machine learning (especially deep learning) is a plus.

Interested candidates (University of Arizona students only) should email,, AND with the subject line “Machine Learning Research Assistant” or "Python Programming Research Assistant" a one paragraph expression of interest, and a CV that lists relevant experience or coursework. For the python position, please send example python code you have written. These positions will be open until filled, and will start reviewing applications on Wednesday, August 18th. This is a good opportunity to learn more about satellite imagery, machine learning, deforestation, and flooding.

Fall 2021 or Winter 2022

The lab is currently seeking one postdoctoral scientist for two new NASA funded projects (described below) to start in Winter 2022 (Jan 3 preferred, but earlier Fall or Summer start dates are possible). See this link to apply:

"Understanding flood risk in human altered landscapes from cities to farms: inferences from satellites and machine learning."

This project leverages inundation observations to understand extreme flood risk in human-modified landscapes, specifically in urban and agricultural areas. We will use commercial and public sensors to i) map maximum inundated area for extreme flood events, ii) quantify flood risk at watershed scales, and iii) estimate of flood return periods at household scales (30m resolution). These objectives will be reached by using deep learning approaches to map floods and damage, and estiamte return periods with Bayesian Hierarchical Models. We will use optical and radar satellites, including MODIS, Sentinel-1, Planetscope, and the HLS (Harmonized Landsat Sentinel-2 dataset), comparing satellite-based flood maps and return period predictions for Bangladesh to those produced by the Bangladesh Flood Forecasting and Warning Centre (from a physically-based model). We will discuss applications for insurance and climate risk financing with collaborators in Bangladesh.

"High resolution imagery to train and validate deep learning models of inundation extent for multiple satellite sensors"

Our dataset, called FloodPlanet, will include 25 major flood events, labeled with co-located Planet, Maxar, HLS (Harmonized Sentinel-2 and Landsat), and Sentinel-1 data to promote cross-sensor fusion across public and commercial sensor and deep learning modeling efforts in the remote sensing and computer vision fields. We will develop our own deep learning models for high resolution urban flood detection with this dataset, and release the the high quality benchmark dataset we will produce in this proposal.

The post doc will work closely with Dr. Tellman and collaborators Dr. Kobus Banard in the Computer Science department at Arizona, Dr. Upmanu Lall at Columbia, NASA scientists Dr. Molthan and Dr. Gurung, and Bangladeshi collaborator Dr. Saiful Islam, as well as staff at the Bangaldesh Flood Forecasting and Warning Center. We plan to visit Bangaldesh during the project. Opportunities to collaborate across other NASA projects at our lab (for megacity flood mapping and infrastructure mapping with deep learning in Central America) will be welcomed and encouraged! Post docs in our lab will be given 20% time to dedicate to personal projects, publishing their dissertations, and applying for grants to further their career.

Human Environment Geography

I am a human-environment geographer seeking to understand, map, and mitigate the causes and consequences of global environmental change with and for marginalized populations. I engage in a wide array of disciplines and methods from land system science, to hydrology, to the social sciences. I will be Assistant Professor of Geography at the University of Arizona in August 2021 and am recruiting PhD students and postdocs that are fully funded! Send your CV, a piece of writing you are proud of, and an expression of interest to my email. PhD and Master's applicants should have either expertise in or a willingness to learn satellite image analysis, programming, and quantitative skills. Applicants must also demonstrate a commitment to building a more socially just world. We are devoted to building a diverse, inclusive, and anti-racist lab. Women and all expressions of gender identity, Black, Indigenous, and people of Color, LGBTQ people, and individuals with disabilities are especially encouraged to apply. Funding support for application fees is available for students from low and middle income countries- please indicate this in an email to me. PhD applications will be accepted for Fall 2022 in January 2022.


Arizona is beautiful, affordable, and diverse socially, geologically, culturally, and ecologically, with plenty of opportunities for outdoor recreation, activism, and volunteering. Tucson has a pioneering climate justice policy and a great local food scene!

Cochise climbing area, Dragoon mountains.

Dropping water on trails near Ajo, Az with No More Deaths