San Francisco, CA

CTrees scientists are heading to San Francisco for AGU26, December 7-11, 2026 at the Moscone Center. The conference is the world’s largest annual gathering of Earth and space scientists.

In eight presentations at the meeting, our team will showcase advances in science and technology for monitoring forests, including new research on methods to model wildfire spread, detect individual trees in the Sierra Nevada, estimate biomass in mangroves globally (more details below).

CTrees scientists at AGU26 will include:

  • Sassan Saatchi, Ph.D., CEO and chief scientist
  • Zhihua Liu, Ph.D., research scientist
  • Adugna Mullissa, Ph.D., senior research scientist and radar science lead
  • Fabien Wagner, Ph.D., senior research scientist and tree-level data science lead
  • Bryan Shaddy, Ph.D., affiliated research scientist
  • Dafeng Zhang, Ph.D., research scientist

Interested in meeting up at AGU, or collaborating with our science team? Please reach out and we will get back to you as soon as possible.

Following are our presentations, listed by day, with links to the full abstracts and presentation details on the AGU26 conference website.

MONDAY, DECEMBER 7, 2026

G13A-01 | NISAR Reveals Landscape-Scale Forest Carbon Variations (Invited)

Lead author: Sassan Saatchi, Ph.D., CEO and chief scientist

Time & location: 2:15-2:25 p.m. | 158 - South (MC)

AGU abstract

Public data from the NASA–ISRO Synthetic Aperture Radar (NISAR) mission first became available this year. Applying 3-6 months of recent imagery, Sassan Saatchi, CTrees’ CEO and chief scientist, will present aboveground biomass estimates for boreal, temperate, tropical wet, and tropical dry forests, as well as results for rapid disturbance detection and forest recovery monitoring.

TUESDAY, DECEMBER 8, 2026

B22D-01 | Observation‑Driven Emission Factors: A Novel Approach Linking Burn Intensity and Biomass Change to Improve Global Wildfire Carbon Emissions (Invited)

Lead author: Sassan Saatchi, Ph.D., CEO and chief scientist

Time & location: 10:35-10:50 a.m. | 3007 - West (MC)

AGU abstract

Climate change, heat waves, and drought are changing fuel conditions and fire behavior worldwide, making it harder for scientists to estimate the resulting carbon emissions. Sassan Saatchi will present a new approach to produce localized, dynamic emission factors by combining observations of burn area and fire intensity with multi‑temporal lidar and radar measurements.

B24F-04 | L and C-Band SAR Based Analysis of Forest Loss and their Drivers in the Pan-Tropics

Lead author: Adugna Mullissa, Ph.D., senior research scientist and radar science lead

Time & location: 4:50-5:00 p.m. | MC, 3008 - West

AGU abstract

Consistent monitoring of deforestation and degradation in tropical forests remains challenging due to persistent cloud cover that limits the effectiveness of traditional optical sensors. Adugna Mullissa, senior research scientist and radar science lead, will present a monitoring approach that applies L- and C-band synthetic aperture radar to penetrate clouds, and an advanced machine learning framework that attributes forest loss to deforestation and degradation. The research analyzes where forest loss is concentrated in the tropics, how that footprint has shifted over the past five years, and how drivers vary by region.

NH22B-04 | Data-Driven Probabilistic Wildfire Spread Prediction Using Conditional Flow Matching

Lead author: Bryan Shaddy, Ph.D., affiliated research scientist

Time & location: 11:00-11:10 a.m. | 307-308 - South (MC)

AGU abstract

Current models of wildfire behavior often use imperfect, semi-empirical formulas for rate of spread to predict how wildfires spread. Bryan Shaddy, affiliated research scientist, will present a deep generative model that brings together advanced remote sensing data with vegetation data and recently observed wildfires in California to better predict the often complex and chaotic nature of wildfire growth.

WEDNESDAY, DECEMBER 9, 2026

B32A-08 | Advances in Active Remote Sensing of Forest Structure and Carbon (Invited)

Lead author: Sassan Saatchi, Ph.D., CEO and chief scientist

Time & location: 11:40-11:50 a.m. | 3008 - West (MC)

AGU abstract

Advances in remote sensing are transforming scientists’ ability to observe forest structure and quantify aboveground biomass across scales relevant to carbon science and management. With examples from tropical and temperate forests, Sassan Saatchi will present on methods that integrate data from multiple sensors to reduce uncertainty in biomass estimates and improve detection of change. The advances will produce better data for carbon accounting, ecosystem modeling, and forest management at regional and global levels.

GC32A-01 | Individual Tree Delineation of All Large Trees across the Sierra Nevada, California, and their Survival Following the 2020–2021 Fire Seasons

Lead author: Fabien Wagner, Ph.D., senior research scientist and tree-level data science lead

Time & location: 10:30-10:45 a.m. | MC, 2008 - West

AGU abstract

Large trees make an outsize contribution to ecosystems’ biomass storage, habitat structure, and overall functioning. However, their distribution and health dynamics remain poorly quantified at regional scale. Fabien Wagner, senior research scientist and tree-level data lead, presents a deep learning model and canopy height model to delineate individual large trees across the Sierra Nevada Floristic Province. The map of more than 6.5 million trees with crown areas greater than or equal to 100m2 demonstrates an approach to monitor the health of large trees at regional scale.

B33D-05 | Tracking Daily Carbon Emissions from Individual Western United States Wildfires

Lead author: Zhihua Liu, Ph.D., research scientist

Time & location: 3:00-3:10 p.m. | 3007 - West (MC)

AGU abstract

Wildfire emissions inventories are commonly produced on a monthly basis by vegetation type, limiting their ability to quantify the emissions impact of individual fires. CTrees research scientist Zhihua Liu will present a study of daily CO2 emission from 519 recent wildfires in the western U.S. The results demonstrate that wildfires emit CO2 in brief, intense pulses, and that fuel moisture has more influence on emissions than vegetation type. In a changing climate, daily observation of emissions can help improve models for smoke spread, atmospheric chemistry, and carbon accounting.

THURSDAY, DECEMBER 10, 2026

B43D-05 | Large Capacity of Global Mangrove Forests for Carbon Storage and Sequestration

Lead author: Dafeng Zhang, Ph.D., research scientist

Time & location: 2:55-3:05 p.m. | MC, 3010 - West

AGU abstract

Mangroves are among Earth’s most carbon‑dense ecosystems, yet widespread deforestation and degradation have undermined this capacity. To date, mangrove carbon stocks and change have lacked a detailed, high‑resolution assessment. Dafeng Zhang, affiliated research scientist, will present a study quantifying the world’s mangrove biomass at 25-meter resolution for the year 2020, with carbon gains and losses reconstructed for 1990-2020. The study finds the mangrove carbon sink is increasingly dependent on remaining undisturbed stands, whose capacity is weakening. As countries seek to protect and restore their mangrove ecosystems as part of climate policies, the dataset offers a benchmark for monitoring, reporting, and verification.