Meet the 2022 Leading Women in ML4EO
These 14 women are pushing the boundaries of science using machine learning for Earth observation worldwide.
Happy International Women’s Day!
We are excited to introduce this year’s cohort of women who contribute to society’s betterment using machine learning for Earth observation (ML4EO).
The role of ML4EO in helping to create a society that is balanced socially, ecologically, and environmentally is paramount. But the technologies we humans develop are not without bias. They mirror our thinking and behaviors. The gender divide of those driving innovative technologies is subsequently of concern. For this reason, we celebrate the women at the forefront of ML4EO, those who are shedding light on our patterns and helping us make data-driven decisions. It is our way to elevate the work of women in Technology.
It is the second year we are hosting this annual award (check last year’s awardees). This year, we invited 45 thought leaders in the community, including last year’s winners, to submit nominations for 2022’s Leading Women in ML4EO. Nominators offered details about nominees’ contributions and the technological solutions they provide to the socio-economic and environmental problems we face in this world. Nominators also expanded on the achievements and leadership roles of the nominees.
Radiant Earth received 24 nominations and selected the final awardees based on the recommendations submitted by Nominators, the nominees’ research contributions, and their proven record of success that are recognized nationally, regionally, or internationally. We also researched each nominee in addition to the recommendations provided, assessing other impacts not listed.
We want to thank all the nominators for their excellent recommendations!
Featured below are the 2022 Leading Women in ML4EO. Congratulations to the finalists!
Subscribe to our Twitter list to follow the work of the Leading Women in ML4EO in 2021 and 2022.
Alexandra Tyukavina, University of Maryland (Russia)
Dr. Alexandra (Sasha) Tyukavina is an Assistant Research Professor at the University of Maryland’s Global Land Analysis and Discovery (GLAD) lab. As a part of the team producing global land cover and change maps, Sasha is responsible for map validation and co-leads the CEOS Land Product Validation Working group’s Land Cover focus area. Her research attributes drivers of forest loss in the tropics and globally using both machine learning and sample-based methods. Sasha’s most recent study maps forest loss due to fire vs. other drivers of forest loss globally at a 30m resolution. She is also a lead investigator of the NASA-funded study to quantify drivers of forest loss globally using high-resolution PlanetScope data. Sasha’s work is highly cited and now contributes to the new protocols from the Committee on Earth Observations on validation and area estimation using best practices. Follow Sasha’s research on Google Scholar.
Amy Pickens, University of Maryland (United States)
Dr. Amy Pickens is a Postdoctoral Associate in the Global Land Analysis and Discovery (GLAD) lab at the University of Maryland. Amy’s research spans many types of large-scale land cover and land-use change with a particular focus on surface water dynamics and near-real-time forest disturbance monitoring. She has mapped global surface water dynamics from 1999 to 2021 and employed a sample-based approach for reporting global areas that nations can use to support the Sustainable Development Goals. She also developed an operational near-real-time forest loss alert system for the Amazon basin using Sentinel-2 data and, together with this, maintains the pan-tropical GLAD Landsat-based alert system. Forest rangers, indigenous communities, and local governments use these alert systems to reduce deforestation through intervention and enforcement. Explore the Google Earth Engine Apps for some of her collaborations and follow Amy’s work on Google Scholar.
Ane Alencar, Amazon Environmental Research Institute (Brazil)
Dr. Ane Alencar is the Science Director at the Brazilian Instituto de Pesquisa Ambiental da Amazônia (Institute for Amazonian Environmental Research — IPAM) and a leading expert on Amazonian fire science. She holds a Master degree in Remote Sensing and Geographic Information System from Boston University and a Ph.D. in Forest Resources and Conservation at the University of Florida. For the past 26 years, she has been working and coordinating a research team dedicated to understanding the dynamics of fire and forest degradation and its relation with land use and climate change in the Amazon and Cerrado. Her ground-breaking discovery and mapping of fires in the Amazon led to the MapBiomas Alert platform that maps all fire scars in Brazil from 1985 until 2021 and secondary findings on the links between deforestation logging and fires. Ane coordinates the Cerrado and Brazil burned area mapping within the Mapbiomas initiative and the Land Use Change Sector of the Greenhouse Gas Emissions Estimation System (SEEG) initiative and integrates, as one of the lead authors, the Scientific Panel for the Amazon. Google Scholar and LinkedIn.
Beth Tellman, University of Arizona and Cloud to Street (United States)
Dr. Beth Tellman is a human-environment geographer whose research addresses the causes and consequences of global environmental change on people, focusing on access to water, flood risk, and land-use change. She engages in multiple disciplines and methods to “socialize the pixel” or understand the social processes behind environmental change captured in satellite image pixels and leverage satellite data to improve human well-being. Beth is an Assistant Professor at the University of Arizona in the School of Geography, Development, and Environment. She is co-founder and Chief Science Officer at Cloud to Street, a public benefit corporation that leverages remote sensing data to build flood monitoring and mapping systems for low- and middle-income countries and insurance companies. Explore her Github page to learn more about her work. Beth received her Ph.D. in Geography (2019) from Arizona State University. She has held fellowships with Echoing Green, Fulbright, and the National Science Foundation.
Caitlin Adams, FrontierSI (Australia)
Dr. Caitlin Adams is a Senior Data Scientist at FrontierSI, focusing on enabling better use of remote sensing data in both the public and private sectors. Key to this is Caitlin’s passion for educating and supporting others. In her work with the Digital Earth Africa and Digital Earth Australia programs, Caitlin consistently develops and delivers high-quality documentation, examples, and live workshops, enabling people to better leverage freely available satellite imagery. Caitlin sees machine learning as an important technique for automating the development of informative data products on large physical scales. She has used deep learning to map tree canopy coverage in Australia and supervised and unsupervised classification to develop a strategy for collecting crop-type ground truth data in Zambia. Follow Caitlin’s work on LinkedIn and Twitter.
Elizabeth Duffy, UP24 (United States)
Ms. Elizabeth Duffy is a Geospatial Analyst and Product Marketing professional at UP42 with 7+ years of combined education and professional experience implementing and promoting environmental/geospatial tech in the US, EU, and Asian markets. Her masters thesis research topics include DEM-based karstification analysis, subsidence analysis with persistent scatterer interferometrics, and DL and ML techniques for urban morphology analysis. Elizabeth works with OS and commercial geospatial software, all types of EO data from multispectral to SAR, and coding languages like Python and SQL. She is currently focusing on incentive alignment with solution providers to help influence and promote geospatial tech adoption.
Elizabeth’s greatest joy is learning and applying cutting-edge tech to solve real-world problems, especially environmental sustainability and resiliency. During her undergraduate studies at UNCW, this passion led her to be the inaugural recipient of the eponymous award: the “Carolyn Elizabeth Duffy Student Sustainability Award.” Follow her work on Twitter and Linkedin.
Fang Yuan, Digital Earth Africa (Australia)
Dr. Fang Yuan is the Director of Operations for the Digital Earth Africa establishment team in Geoscience Australia. Fang has a Ph.D. in Physics from the University of Michigan and has a wide range of expertise across optical, high-energy, and radar remote sensing, image analysis, automation, and big data. As an astrophysicist, Fang studied cosmic explosions and discovered some of the most luminous supernovae known to date. As an Earth observation scientist, Fang has led the development of satellite imagery-based products for natural resources mapping, disaster management, and land cover change monitoring for Australia and Africa. In Digital Earth Africa, Fang works with a diverse multidisciplinary team to provide free and open data and services to help address sustainable development challenges. Follow Fang on Twitter for updates on her work.
Ge Peng, the University of Alabama in Huntsville (United States)
Dr. Ge Peng is a Senior Principal Research Scientist at the University of Alabama in Huntsville (UAH). In the last decade, Dr. Peng has led cutting-edge research and developed community practices/standards on assessing the quality of individual Earth observation datasets. She is a strong advocate for consistently documenting and globally sharing interoperable dataset quality information, essential to AI/ML applications. She is currently leading a team of UAH/IMPACT domain experts to develop a machine learning training dataset strategy and action plan for NASA’s Earth Science Data System Program. This plan aims to ensure and enhance the quality, discovery, access, and (re)use of NASA ML training datasets and to work with international research data communities to make ML training datasets and models FAIR. Aside from her regular duties, Dr. Peng serves as a chief editor of the Earth System Science Data journal and chairs the WMO Expert Team on Information Management. Follow Peng’s work via her LinkedIn Profile.
Gladys Mosomtai, United Nations Economic (Kenya)
Ms. Gladys Mosomtai is a passionate user of geospatial technology for solving global environmental challenges, especially in Africa. She is a Research Fellow at the United Nations Economic Commission for Africa and currently finalizing her Ph.D. at the University of KwaZulu-Natal in South Africa and the International Centre of Insect Physiology and Ecology in Kenya. Her doctoral research explored Earth observation datasets to understand the factors influencing coffee pest populations at landscape and plot levels while integrating machine learning algorithms. Gladys has co-authored several scientific publications in peer-reviewed journals, and in 2018, she received the L’Oréal-UNESCO For Women in Science Sub-Sahara Africa Fellowship. She is also the Next Einstein Forum Ambassador for Kenya, aiming to promote science in Africa. Gladys hopes to inspire young women to join STEM and is fully committed to breaking the glass ceiling in her career while dropping the ladder for others to climb. Keep track of her work on Twitter and LinkedIn.
Katie Dagon, The US National Center for Atmospheric Research (United States)
Dr. Katie Dagon is a Project Scientist at the National Center for Atmospheric Research (NCAR) in Boulder, Colorado, working in the Climate and Global Dynamics Laboratory. Her research focuses on modeling the impacts of climate change on land-atmosphere interactions and climate variability. She applies machine learning approaches to climate data to detect extreme weather events, understand climate predictions, and quantify uncertainty in model projections of climate change. At NCAR, she co-leads the Earth System Data Science initiative to promote deeper collaboration centered on analysis workflows, large datasets, and open development. Katie is active in science communication and public engagement and supports diversity and inclusion efforts across the Earth sciences. She obtained her Ph.D. in Earth and Planetary Sciences from Harvard University in 2017 and her B.S. in Mathematics-Physics from Brown University in 2010. Katie lists all professional activities on her website. You can also keep track of her work on Twitter and LinkedIn.
Phoebe Oduor, Regional Centre for Mapping of Resources for Development (Kenya)
Ms. Phoebe Oduor works at the Regional Centre for Mapping of Resources for Development (RCMRD), where she supports two functions: Thematic Lead for Land Use Land Cover and GHG Inventories for the SERVIR Eastern and Southern Africa which is a joint initiative for USAID and NASA, and the Point of Contact for the AfriGEO Secretariat, a regional initiative for the Group on Earth Observations (GEO). Phoebe utilizes Earth observation data and machine learning algorithms and tools to address development challenges in 10 African countries (Kenya, Uganda, Rwanda, Malawi, Namibia, Zambia, Botswana, Tanzania, Lesotho, and Ethiopia). Some of these data have been used to support monitoring of land use activities and track forest activities for measurement and reporting, particularly in determining existing baselines and quantifying existing natural capital. She was shortlisted for the 50 Rising stars 2022 in Geospatial World Media to recognize these efforts. Follow her work on Twitter and LinkedIn.
Sherrie Wang, University of California, Berkeley (United States)
Dr. Sherrie Wang is a Ciriacy-Wantrup Postdoctoral Fellow at UC Berkeley, where she uses machine learning and remote sensing to map the world’s progress toward sustainable development. She is particularly passionate about creating maps for regions where ground data are scarce and development is fastest, such as South Asia and Sub-Saharan Africa. She develops novel machine learning algorithms suitable for remote sensing data and label-scarce settings to create such maps. Sherrie’s work to date has focused on agriculture and natural hazards; for example, her latest research includes using satellite imagery and deep learning to delineate crop fields and crop types at high resolution across India for the first time. By creating geospatial datasets, Sherrie aims to evaluate policies and technologies to improve outcomes for farmers while minimizing environmental degradation. Sherrie is also committed to increasing the representation of women in data science and helped organize the first WiDS datathons. Explore Sherrie’s Github page to learn more about her research.
Tanuja Shrestha, International Center for Integrated Mountain Development (Nepal)
Ms. Tanuja Shrestha works as a Geospatial Associate in Data Science at the International Centre for Integrated Mountain Development (ICIMOD). At ICIMOD, she works with the SERVIR-HKH Initiative and is currently researching deep learning methods and applications of computer vision in geospatial data. Using freely available Sentinel 2 satellite imagery, OpenStreetMap data, TensorFlow — a deep learning framework, and the Google Earth Engine, Tanuja developed a model to predict the extent of urbanization in Nepal. Currently, she is applying her model to generate maps for built-up areas for different years to monitor urbanization in the country. Tanuja has also used image classification and object detection frameworks to detect animal species in camera trap imagery. Her research interests are artificial intelligence, machine learning, deep learning, computer vision, remote sensing, geographical information system, data visualization and analysis, and geospatial big data. In 2017, she completed a master’s degree in high performance and scientific computing from Swansea University as a Chevening Scholar. Tanuja also holds a master’s degree in environmental science and natural resources from Kathmandu University in Nepal. Follow her on Twitter and Github.
Tzu-Hsin Karen Chen, Yale University (Taiwan)
Dr. Tzu-Hsin Karen Chen is a Donnelley Postdoctoral Associate at the Yale University’s School of the Environment. Her research interest is at the intersection of machine learning, urban environments, and health inequality. During her Ph.D. in the Big Data Centre of Environment and Health at Aarhus University (2017–2020), she worked on an interdisciplinary project on the relationship between dynamic urban form and mental health inequality. This work was complemented by a research exchange at the German Aerospace Center (DLR) in 2019, where she and collaborators developed the first deep learning method predicting 3-D urban structure information at 30m resolution across time. In 2020, she was a visiting researcher and a lecturer at the University of Copenhagen where she co-taught Remote Sensing in Land System Science. At Yale, she focuses on open data science for urbanization studies in the Global South, including sustainable development in the Himalayas and Sub-Saharan Africa. Keep track of Tzu-Hsin research on LinkedIn.
