HOW DOES AVAILABILITY OF SURFACE AND SUBSURFACE WATER VARY UNDER DIFFERENT CONDITIONS?
WHAT ARE THE DRIVERS OF STREAM INTERMITTENCY?
The presence of surface water is dictated by how much water arrives at a point from upstream and how much capacity the subsurface has to transport that flow. Learning more about the spatial variability in these drivers requires observing surface water presence and its variability in time. We have developed machine learning models to detect surface water presence in satellite imagery, and we are testing this models at new sites and developing guidelines for how to train and develop models to achieve reliable datasets. We are using these and existing datasets to develop a more comprehensive understanding of surface water presence variability and its drivers.
​
HOW DO ROOT ZONE PROCESSES IMPACT RUNOFF GENERATION?
In most places in the world, water infiltrates into the ground when it rains. That water then moves through soils and underlying weathered bedrock before reaching the groundwater table from which water flows to streams. The root zone extends through a significant portion of the subsurface so that plants can capture and use water as it moves along its course. In seasonally dry environments, the root zone can be many meters deep so that plants have a large impact on hydrological functioning of hillslopes. It is very difficult to observe these dynamics since they are deep underground. Deep observation wells and other state-of-the-art instrumentation setups allow us to monitor the spatiotemporal dynamics of water accumulation and use deep underground. These findings have implications for biogeochemical processes, runoff generation, plant response to changes in climate, and quantifying recharge.
​
HOW DO LANDSCAPE FACTORS IMPACT WATER AVAILABILITY?
Both natural and human factors drive where water is stored in landscapes and whether surface water is present. In desert landscapes, there is very little surface water present, so subsurface storage drives ecosystem dynamics across most of the landscape. To take advantage of this, earthworks have been constructed across the southwestern United States. However, the long-term impacts on ecosystem functioning are poorly understood. In wetter regions, there is substantial variability in stream types from slow-moving shallow streams to streams that respond rapidly to rainfall with high spikes and drops in flow. These differences are controlled by subsurface properties but also may relate to ecosystems through factors such as the presence of large wood in channels.
​​
Read about earthwork detection and impacts here and here.
Read about the impact of wood on channel morphology here.
HOW DO OVERLAND FLOW MECHANISMS FUNCTION?
In landscapes with a shallow subsurface confining layer, the shallow storage is quickly filled, and excess rainfall runs over the surface.
Water age and flow pathways should be related, but the specifics of these relationships are still unknown. We combine field observations of runoff generation at the Dry Creek catchment in Northern California with StorAge Selection (SAS) age models to explore the relationship between streamwater age and runoff pathways in a saturation overland flow-dominated landscape.
​
See a short presentation on this topic here or a lecture here.
HOW CAN WE DETECT AND MANAGE STREAMFLOW DEPLETION FROM GROUNDWATER PUMPING?
BALANCING THE NEEDS OF AGRICULTURE AND AQUATIC ORGANISMS IN WISCONSIN
Wisconsin is home to abundant surface and groundwater resources that support both human needs and diverse aquatic ecosystems. However, the relationship between groundwater resources and stream ecology remains poorly quantified at the regional scale, particularly with respect to the relative importance of shifting precipitation patterns and increased groundwater reliance across the state. Given variable data availability at the state-wide scale, we explore the utility of different approaches to quantify hydrological alteration in a management context and explore how quantitative estimates of the effects of groundwater withdrawals on streams can be interpreted in terms of aquatic ecosystem health.
​
Read the publications here and here.
​
* artwork by Anna Klein, supported through the Flow microgrant*
HOW CAN WE MEASURE THE IMPACTS OF GROUNDWATER PUMPING ON STREAMS?
Groundwater pumping leads to reductions of flow in nearby streams. While well understood, it remains extremely difficult to detect because streamflow may vary based on climatic conditions as well as human impacts. We used models to explore how streamflow depletion impacts hydrographs and stream temperature. Streamflow depletion leads to longer, lower low flows and higher summer temperatures. Smaller streams are most vulnerable to impacts.
​
Read the publications here and here.
HOW RELIABLE ARE MANAGEMENT TOOLS UNDER CURRENT AND FUTURE CONDITIONS?
WHERE AND WHEN DOES THE RATIONAL METHOD WORK?
The Rational Method, despite or because of its simplicity, is one of the most widely recommended tools for estimating the peak flow of a given return period. The apparent simplicity of the Rational Method masks a number of complicated assumptions. We used analytical and numerical modeling of idealized hillslopes to identify the impacts of hillslope shape and soil properties on reliability of the Rational Method. We found that the timescale typically used in the Rational Method (the time of concentration) is not always a good estimate for the critical duration, the theoretically appropriate timescale. We developed new Arc Hydro tools to assist managers in identifying if the Rational Method can be used at their sites.
​
Read about underlying model development and testing here.
Read the publications here, here, and here.
Read a Conversation article about this work here.
Hear us talk about our new tools in an ESRI webinar.
Read more about our new tools here.
​
ROOT-ZONE STORAGE DEFICITS DRIVE CHANGES IN RUNOFF FOLLOWING DROUGHTS
More frequent droughts and increasing temperatures imposed by climate change threaten snowpacks, which sustain mountain water resources globally. Following a recent drought in California, the traditionally used model for snowmelt runoff failed. Here, we present a model that reveals the essential role of root-zone storage dynamics in snowmelt runoff. Through transpiration, montane forests generate water storage deficits in the soils and weathered bedrock that comprise the root zone. These deficits must be replenished by rain and snowmelt before significant runoff generation can occur. Overprediction of 2021 post-drought runoff in California can be primarily attributed to unprecedented root-zone storage deficit magnitudes. Adding a measure of deficit reduced 2021 streamflow prediction error from 100% to 12%.
​
Read the publication here.
Read a Conversation article about this work here.
Hear me talk about this work here.
This work covered in The New York Times here.
​
Image source: NASA Worldview • Graphic by Pai/Bay Area News Group