Sally Thompson

Our lab group studies ecohydrology - a fairly new discipline that addresses the interaction of ecosystems and the water cycle. Life (especially plants, but also animals such as beavers – and of course people!) change how water moves and is stored on the Earth's surface.  The availability of water also helps control what plants and animals can live in a given place. The resulting  feedbacks between life and the water cycle can generating both important outcomes (e.g. water availability for people, or sustainability of land and water-based ecosystems), but also intriguing and surprising outcomes (e.g. plants growing in regular patterns through space).  We study ecohydrology on scales ranging from individual plant leaves, to whole plants, ecosystems, landscapes and river basins, and also engage with strongly human-dominated systems.  

The research portfolio in our group changes regularly.  Some projects that were current in summer 2017 are outlined below.


Eel River Critical Zone Observatory

The Eel River at the ERCZO
Elder Creek is a tributary to the Eel River,
and a site of intensive research at the ERCZO

The Eel River Critical Zone Observatory (ERCZO) is an intensively monitored field site focusing on the critical zone (Earth's living skin, spanning the point where fresh bedrock is reached in the subsurface, to the top of the boundary layer in the atmosphere).  ERCZO research follows watershed currencies (water, solutes, gases, biota, sediment, energy, and momentum) through the subsurface physical environment and microbial ecosystems in the critical zone into the terrestrial ecosystem, up into the atmosphere, and out through drainage networks to the coastal ocean.

Soil, forest, and riverine ecosystems interact with these currencies, mediating the delivery of nutrients to the sea. CZO investigations are co-evolving with models that mechanistically links the critical zone to atmospheric processes, watershed routing, ecosystems dynamics, stream flow, and coastal processes in order to investigate fundamental questions and to provide a modeling tool for management issues.  

Learn more about the ERCZO.

Published papers from our work at ERCZO include:

Current work includes developing and testing models of flow dependence on CZ structure, drivers of the stream thermal regime in the Eel River, the importance of preferential flow for water budget partitioning, testing unmanned aerial systems and their value for observing variations in channel temperatures, and the significance of cold water tributaries in creating habitat for fish in the warmer main channel of the Eel River.

How does restoring a natural fire regime influence forest structure and hydrology in California's Sierra Nevada?

Fire-sculpted landscape in the Illilouette Creek Basin, Yosemite National Park
Forty years of a restored fire regime have reshaped the landscape
in Yosemite National Park's Illilouette Creek Basin.  
Here, new wetland plants are growing in the remains of a dense forest.

Mountain watersheds supply most water used in the western United States, and these ecosystems are, fundamentally, fire prone. Warming climates threaten water supply from these watersheds in several ways: snow-packs are likely to decrease and melt earlier in the year, while the forests that cover watersheds are likely to experience more frequent and severe fires which can damage infrastructure and diminish water quality.

Management approaches that can simultaneously reduce fire risk while protecting water supplies are sorely needed. One such approach is managed wildfire, which allows natural fires to burn without intervention (if safe to do so) - in contrast to the current approach of fire suppression, which has been practiced for almost 100 years, and results in fires being extinguished on detection. Fire suppression has changed forests in ways that have likely both increased fire risk and reduced water supply.

This research project takes advantage of a unique pair of watersheds in the Sierra Nevada where managed wildfire implemented since the 1970s. We are determining how vegetation cover, snow and soil moisture behavior, and water yields have responded to the changed fire regime. The results are synthesized in a modeling framework that can be applied to evaluate the potential consequences of similar shifts in fire regime elsewhere in the Sierra Nevada. 

Published papers from this project include:

  • Boisramé, G.F., Kelly, N., Stephens, S. and Thompson S.E.: Vegetation Change During 40 Years of Repeated Managed Wildfires in the Sierra Nevada, California, Forest Ecology and Management, accepted.
  • Boisramé, G., Thompson, S.E., Collins, B., and Stephens, S., Managed Wildfire Effects on Forest Resilience and Water in the Sierra Nevada, Ecosystems, 2016. doi:10.1007/s10021-016-0048-1.

... with more on the way!

Using UASs to support in-situ water column characterization and sampling

UAS thermal sampling system
Thermal sampling
system on the UAS


Unmanned aerial systems (UAS), offer the potential for extensive spatial coverage and sampling of water systems, even in hard-to-access areas.  We are working with an innovative aerial thermal sensing platform that uses a temperature sensor to profile temperature variations vertically and laterally in water columns, comparing its performance against in-situ temperature arrays, and assessing how the specifications of the platform match up against various modeled scenarios of thermal variation (e.g. those induced by hyporheic flow in rivers) in order to inform sampling campaign design with the UAS.  Next steps will include an assessment of how the temperature sensor and IR cameras can work together to facilitate observations.

Papers from this project include:



Hydrologic reconstruction and prediction in data-scarce regions

A dry shallow well in the Arkavathy Basin
Surface wells such as this one in the
Arkavathy Basin, South India, have been
dry since the 1970s.  
Photo credit: Gopal Penny

Complexity and non-stationarity can obscure the mechanisms that drive hydrologic change - a big problem in data-scarce areas where change is occurring fast, and evidence-based management is urgently needed.  Hydrologic reconstruction seeks to address this challenge by quantifying historical processes beyond the period of observed data records. Hydrologic prediction in these areas demands the use of low-complexity models that are process based (allowing non-stationarity in drivers, at least, to be handled, without having high data demands).  

We have proposed that reconstruction in data scarce regions requires using the method of of multiple working hypotheses, developing new kinds of datasets, and understanding process change, as a precursor to model or theory based analyses.  We have also worked extensively to extend the use of parsimonious probabilistic models for predictions in these regions, particularly by adapting their use to satellite derived data, seasonally-dry systems, predictions in ungauged basins, and quantifying the economic value of information that can be supplied to the models.

Our study systems have included extensive work in the Arkavathy Basin in South India, in collaboration with the ACCUWa project at the Ashoka Trust for Research in Ecology and the Environment (ATREE), and in the context of exploring barriers to rural electrification via microhydropower in Nepal.

Papers from these projects include:

Fog impacts on water balance - Pilarcitos Creek Watershed and beyond.

Time lapse cameras show distinct fog patterns in the 
Pilarcitos Creek Watershed, leading to different strategies for
interpolating observations of throughfall and fog drip.


In summer-dry environments, fog can provide a critical supplementary source of water to ecosystems and potentially surface water bodies.  Although the importance of such "non-meteoric" precipitation is increasingly recognized, it remains challenging to quantify its impact on large scales.  We have been working in Pilarcitos Creek, a small water supply catchment for the San Francisco Public Utility Commission (SFPUC) to understand how summer fog might influence streamflow and the watershed behavior.  The watershed is influenced by fog traveling along distinct pathways (North-South versus West-East) that differ with the prevailing wind direction.  Using time-lapse photography and satellite images, we can determine which mode of fog input occurs, and use this as a template for distinct interpolation strategies to upscale measurements of fog intensity, through fall and suppression of transpiration.  We have used similar methods to assess leaf-scale impacts of non-meteoric water at Blue Oak Ranch Reserve in collaboration with Cynthia Gerlain-Safaldi and Kelly Caylor by linking mechanistic models with observations of leaf wetness.

Papers from these projects are upcoming.

The Endless Summer - how does water stress impact plants and their communities?

Pepperwood Preserve in the Mayacamas Mountains in Sonoma County offers
an exciting location to explore vegetation x lithology x water interactions.

California is recovering from five years of extreme drought. What has the impact of this "window into the future" - a future where drought may be more frequent, severe and hotter - taught us about the potential impacts of extremes on vegetation communities?  We have been following the health of a series of trees across a drought gradient to observe the immediate impacts of drought and recovery from drought in oak savannas.  While climatic drought is a first-order control on plant responses, at any given location, we would often find a huge range in vegetation health - suggesting that individual plant traits, their locations on different lithologies or variable access to water sources - were ultimately controlling plant health and drought responses.  We are now exploring this hypothesis in more depth at Pepperwood Preserve in Sonoma County, by making detailed measurements of plant traits, plant health, and hydrological processes across a set of species and lithological substrates.  

Papers from these projects include:

... with more on the way!

Connectivity and its impacts on plant water availability in drylands

Lehavim LTER in Israel will be a major
study site for this project.

Degradation of drylands, or desertification, impacts 40% of Earth’s land surface, 40% of the global population, and is responsible for up to 25% of global greenhouse gas emissions. Many dryland ecosystems are sustained by runoff-runon processes that result in the transport of resources, particularly water, from bare to vegetated sites, determining the ultimate availability of water to plants. Such transport is dependent on storm and site characteristics, making its behavior temporally dynamic and spatially complex. In this context, predicting the likely vulnerability of desert landscapes to degradation or the quality of proposed restoration activities from readily observed surface features of drylands would be helpful. This project attempts to advance on previous attempts to generalize such predictions using models of plant water availability or runoff volumes exported from drylands, by testing models of source-sink transport of water during storm events against tracer data, and developing realistic representations of spatial vegetation patterning.  We exploit isotopic tracer experiments to reveal patterns of connectivity and water availability to vegetation on a well-characterized dryland hillslope at the Lehavim LTER in Israel. The patterns will be used to test the predictions of runoff-runon models by adapting these models to predict Lagrangian transport of water. Model experiments using the tested model will draw on the emerging field of multi-point statistics to enable realistic realizations of landscape surface features, allowing uncertainty in predicted water availability, degradation risk or restoration quality to be characterized.

Papers related to this project include: