Crowdsource Disaster Response

.@CHNickerson and @SaraAnnWylie say wetlands are more than a home to wildlife — they are a “vital protector” during hurricanes that deserve federal investment.

“These are people’s homes on the line. These are people’s livelihoods on the line.”×720/6ZtMtcIfw3vqLdKv.mp4?tag=14

Find out more about these causes here:

Originally tweeted by Zerlina on Peacock (@ZerlinaShow) on 08/31/2021.

Cartoscope–Crowdsource analysis of Images from Environmental Disasters

In the aim of building connections between community science and regulatory action, I have collaborated with Northeastern computer science professor Seth Cooper to develop a website for crowdsourcing analysis of images from both rapid and protracted environmental disasters. This project further develops one of Public Lab’s open source software tools, MapMill for crowdsorting of image sets. MapMill was adapted and used by FEMA in response to Hurricane Sandy to organize a crowd-sorting effort that yielded thousands of images. This was the first case of a federal agency rapidly picking up open source software and using crowdsourced input to inform damage assessment. As crowdsourcing was beyond the original mission of Public Lab, Cooper and I took on the tool for further development. We received a Tier 1 Northeastern research award in 2016 to develop this renamed project, Cartoscope. We received an NSF grant (National Science Foundation under Grant No. 1816426) to further develop the project.


Land Pollution Lookout

In 2022 in collaboration with Unique Mappers in Nigeria and Healthy Gulf, Cartscope released Land Pollution Lookout to support Unique Mappers in identifying damage from oils spills in the Niger Delta. SciStarter supported the release for April’s Global Citizen Science month. 

Land Loss Lookout

After the Gulf Restoration Network rebranded to become Healthy Gulf we began a project tracking land loss in the Gulf of Mexico. For this collaboration we teamed up with SciStarter and Caroline Nickerson, who represented the project. This project began by looking at 2016 near infrared satellite data. Over 1000 people have participated. The project is currently in it’s second iteration looking at 2008 satellite data to compare between the time periods.

See coverage of the project:

Ida Damage Lookout

Following Hurricane Ida in 2020 we worked with Healthy Gulf to release a tracker for assessing damage to industrial facilities following the hurricane and to identify oil spills.

Algal Bloom Tracking

We worked with the Cleveland Museum of Natural History and Rafat Ansari’s Citizen Science Algal Bloom Tracking site to use this website to crowd-identify potential algal blooms. This is an example of the kind of slow disaster that is likely to become much more common in the context of Climate Change.

This slideshow requires JavaScript.

Oil Sheen Tracking following Hurricane Harvey

Hurricane Harvey devastated the Gulf Coast of Texas, the Houston Metro Area, and the Western Gulf Coast of Louisiana. Petrochemical facilities all along the coast were in the direct line of Harvey’s path. Many of these facilities released a host of different chemicals into the floodwaters. Help the Gulf Restoration Network‘s efforts in identifying oil sheens in aerial images of the area.


Seth Cooper
Sophie Spatharioti (lead developer)

Kutub Gandhi (lead developer)
Rafat Ansari
Lorna Dsilva,
Vasanti Mahajan,
Rick Menasce,
Shubhi Mittal,
Kisalaya Prasad,
Nitin Shetty,
Becca Govoni

2022 Cartoscope website redesign:

Archana Apte (design), Asha Padmashetti (implementation), Drishti Sabhaya (implementation)

Land Pollution Lookout Outreach:

Emi Suyama, Emily Philpott, Kate Clemenz




Project Publications

Under review:

Sofia Eleni Spatharioti, Eliza Boetsch, Scott Eustis, Kutub Gandhi, Naomi Yoder, Matt Rota, Archana Apte, Seth Cooper, Sara Wylie. (2021) “An Effective Online Platform for Crowd Classification of Coastal Wetland Loss”. Conservation Science and Practice. Submitted Dec 2021.



Spatharioti, S. E., Wylie, S., & Cooper, S. (2021, August). Exploring Q-Learning for Adaptive Difficulty in a Tile-based Image Labeling Game. In 2021 IEEE Conference on Games (CoG) (pp. 1-8). IEEE.

Gandhi, K.*, Miller, J.* A., Spatharioti, S. E., Apte, A., Fatehi, B., Wylie, S., & Cooper, S. (2021, August). A Comparison of Augmented Reality and Browser Versions of a Citizen Science Game. In The 16th International Conference on the Foundations of Digital Games (FDG) 2021 (pp. 1-8).

Sofia Eleni Spatharioti*, Borna Fatehi, Melanie Smith*, Avery Rosenbloom*, Josh Aaron Miller*, Magy Seif El-Nasr, Sara Wylie, and Seth Cooper. 2020. Tile-o-Scope AR: An Augmented Reality Tabletop Image Labeling Game Toolkit. In International Conference on the Foundations of Digital Games (FDG ’20). Association for Computing Machinery, New York, NY, USA, Article 85, 1–4. DOI:

Sofia Eleni Spatharioti*, Sara Wylie, Seth Cooper. 2019.Using Q-Learning for Sequencing Level Difficulties in a Citizen Science Matching Game”. Extended Abstracts of the 2019 Annual Symposium on Computer-Human Interaction in Play (CHI PLAY 2019) pp. 679-686.

Spatharioti, Sofia Eleni*, Sara Wylie and Seth Cooper. 2018. “Does Flight Path Context Matter? Impact on Worker Performance in Crowdsourced Aerial Imagery Analysis.” Presented in Proceedings of the Information Systems for Crisis Response and Management (ISCRAM) 2018 Conference, Rochester, NY, USA.

A Required Work Payment Scheme for Crowdsourced Disaster Response: Worker Performance and Motivations Sofia Eleni Spatharioti, Rebecca Govoni, Jennifer S. Carrera, Sara Wylie, Seth Cooper Proceedings of the 14th International Conference on Information Systems for Crisis Response And Management, (2017)

On Variety, Complexity, and Engagement in Crowdsourced Disaster Response Tasks Sofia Eleni Spatharioti, Seth Cooper Proceedings of the 14th International Conference on Information Systems for Crisis Response And Management, (2017) Best Student Paper Award Nomination