Takeaway: Wet soil can look like it is calling the rain, but scientists have to separate that signal from seasons and lingering storms before they can trust the story.

Rainfall forecasts, drought planning, wildfire risk, and climate models all depend on a deceptively simple question: does wet soil help make tomorrow’s rain, or are we just seeing the leftovers of yesterday’s storm? This episode follows a paper that acts like a scientific caution sign. Tuttle and Salvucci show that the land-atmosphere connection is real and important, but it is easy to fool ourselves if we do not separate true feedback from seasonal cycles, multi-day storms, and statistical bias. We unpack Granger causality as a practical test of whether one thing improves prediction of another, then translate the paper’s three big traps into everyday examples: summer patterns, storm systems that linger, and the problem of soil moisture being both cause and consequence. Along the way, we talk about why this matters for models, satellite data, weather risk, and the way scientists build confidence without overclaiming. Citation: Tuttle, S. E. and G. D. Salvucci (2017), Confounding factors in determining causal soil moisture-precipitation feedback, Water Resources Research, 53, 5531–5544, doi:10.1002/2016WR019869. This episode uses AI-generated voices for the hosts.

Full citation: Tuttle, S. E. and G. D. Salvucci (2017), Confounding factors in determining causal soil moisture-precipitation feedback, Water Resour. Res., 53, 5531–5544, doi:10.1002/2016WR019869.

Takeaway: The ground can nudge tomorrow’s rain, but in this study wet soil helped rain chances in much of the dry West and often worked the other way in the humid East.

Rain does not only fall from the sky; it also leaves a memory in the ground. That matters for drought forecasts, flood risk, farming decisions, and how climate models represent the daily conversation between land and atmosphere. In this episode of Waterlines, we explore a study that asked a deceptively simple question: after it rains and the soil changes, can that soil moisture make the next rain more—or less—likely?

Using satellite soil-moisture observations, rain-gauge-based precipitation data, and a careful statistical approach designed to separate cause from coincidence, Samuel Tuttle and Guido Salvucci found a striking U.S. pattern. In much of the drier West, wetter-than-usual soils tended to raise the odds of next-day rain, while drier soils lowered them. In much of the more humid East, the feedback often flipped: drier soils were linked to higher rain probability, and wetter soils to lower probability. Across the contiguous United States, significant feedback appeared over about 38% of the land area, with soil moisture changing rainfall probabilities by a median factor of 13% where the signal was significant.

We unpack why this is not as simple as “wet ground makes rain.” Soil moisture changes how sunlight is split between evaporation and heating the air, shaping humidity, temperature, and the growth of the lower atmosphere. But storms also have their own momentum, seasons have their rhythms, and yesterday’s rain naturally leaves today’s soil wet. The episode walks through how the researchers tried to avoid mistaking weather persistence for land feedback, what the east-west contrast may reveal about arid and humid climates, and why the Great Plains result was especially interesting.

Citation: Tuttle, S., and Salvucci, G. “Empirical evidence of contrasting soil moisture–precipitation feedbacks across the United States.” Science 352(6287), 825–828 (2016). https://doi.org/10.1126/science.aaa7185

Disclosure: This Waterlines episode uses AI-generated voices for the hosts.

Full citation: Tuttle, S., and Salvucci, G. “Empirical evidence of contrasting soil moisture–precipitation feedbacks across the United States.” Science 352(6287), 825–828 (2016). DOI: 10.1126/science.aaa7185

Takeaway: Satellites do not replace people on the ground, but they can show the shape of missing or overflowing water when gauges are too few, too late, or too far apart.

Floods can arrive overnight; droughts can creep in for months before anyone agrees what to call them. This episode matters because communities, farmers, emergency managers, insurers, and water planners all need the same thing: a clearer picture of where water is, where it is missing, and how fast conditions are changing. We explore how satellites help fill the gaps between river gauges, rain stations, soil probes, and field reports—especially in places where ground data are sparse or disasters unfold across huge regions.

Using the edited scientific volume Remote Sensing of Hydrological Extremes as our guide, we unpack how different satellites “see” water: optical sensors that spot dark floodwater, microwave instruments that can work through clouds, radar that detects flooded forests, gravity missions that weigh changes in underground and surface water storage, and vegetation signals that reveal drought stress in crops and ecosystems. We visit examples from the Magdalena River in Colombia, snowmelt flooding in the Red River of the North, NASA global flood maps, Congo floodplain hydraulics, Southeast Asia flood impacts, and drought monitoring in Brazil, China, and the western United States.

We also keep the science honest: satellite maps can miss water under trees, confuse cloud shadows with floods, struggle with coarse pixels, or need careful calibration with models and field data. But when used thoughtfully, they can turn scattered clues into practical warning and recovery information.

Citation: Lakshmi, V. (ed.). Remote Sensing of Hydrological Extremes. Springer Remote Sensing/Photogrammetry. Springer International Publishing Switzerland, 2017. https://doi.org/10.1007/978-3-319-43744-6

Disclosure: This Waterlines episode package is designed for production with AI-generated voices.

Takeaway: A satellite soil-moisture map is more trustworthy when rainy days and wet ground line up in the pattern real soils naturally make.Flood warnings, drought outlooks, farm decisions, and climate models all depend on a deceptively simple question: how wet is the ground? Satellites can scan huge areas that no field crew could visit every day, but their soil-moisture maps are hard to check because a satellite pixel covers many kilometers while a ground sensor samples one small spot. This episode explores a clever workaround: instead of asking whether a satellite matches scattered soil probes, the researchers ask whether its wet-and-dry pattern makes sense when compared with large-scale rainfall.We unpack how soil moisture sits at the busy meeting place between rain, runoff, evaporation, plant water use, and heat. Then we follow the paper’s main idea: over time, real soils leave a recognizable fingerprint. Dry ground tends to be linked with lower recent rainfall, wet ground with higher rainfall, and the curve between them has a distinctive S-like shape because drainage, runoff, and evaporation respond differently as soils fill. If a satellite product scrambles that relationship, it may be carrying more error. The authors used a statistic called mutual information, explained here as a way to measure how much two messy datasets “know” about each other, to compare three AMSR-E satellite soil-moisture products across the contiguous United States from 2002 to 2011.The result is not one universal winner. The University of Montana product carried the most useful rainfall-linked information across about 50 percent of the region, the VUA-NASA product across about 47 percent, and the NASA product across about 3 percent. The better-performing product also depended on landscape: flatter, less vegetated areas tended to favor the University of Montana product, while more rugged or more vegetated areas often favored VUA-NASA. The study also found winter and frozen-ground behavior that needs care, reminding us that satellite data are powerful but not magic.Citation: Tuttle, S. E., & Salvucci, G. D. (2014). A new approach for validating satellite estimates of soil moisture using large-scale precipitation: Comparing AMSR-E products. Remote Sensing of Environment, 142, 207–222. https://doi.org/10.1016/j.rse.2013.12.002Disclosure: This Waterlines episode package is written for public-science listening and uses AI-generated voices.

Takeaway: A snowpack can look still, but its microwave glow changes when it flips between frozen storage and liquid water.

Spring runoff, flood risk, irrigation water, forest timing, and even winter wildlife can all hinge on a quiet moment: when snow first starts turning to liquid water, then freezing again overnight. This episode follows a study that improves how satellites spot those freeze-thaw pulses from space. Instead of treating every day-night change in microwave signal as melt, the researchers ask a practical question: how much of that signal is just the snow and land getting warmer, and how much is actually water changing phase?

We unpack how passive microwave satellites “see” snow, why wet snow suddenly looks brighter in microwave measurements, and why air temperature can help separate ordinary warming from real melt and refreeze. The conversation travels from a flat farm-country satellite pixel in the Northern Great Plains to Colorado’s Senator Beck Basin, where snow surface temperature and energy-balance instruments help test whether the satellite detections make physical sense. Along the way, we talk about why this matters for spring streamflow forecasts, snowpack models, climate records, and the everyday water supply that begins as mountain snow.

Citation: Tuttle, S. E., & Jacobs, J. M. (2019). Enhanced identification of snow melt and refreeze events from passive microwave brightness temperature using air temperature. Water Resources Research, 55, 3248–3265. https://doi.org/10.1029/2018WR023995

Disclosure: This Waterlines episode uses AI-generated voices for the hosts.

Takeaway: In the glacial aquifer, water that rushes through thin, carbonate-poor sediment tends to stay acidic, while water that lingers through carbonate-rich layers often turns alkaline enough to loosen arsenic.

A glass of well water can look perfectly clear and still carry clues about the ground it traveled through for years, decades, or longer. In this episode, we explore why groundwater pH is more than a number from chemistry class: it can shape whether arsenic stays stuck to aquifer sediments, whether manganese moves into water, and whether water is more likely to corrode plumbing. The study follows the huge glacial aquifer system across the northern United States, a water source for about 30 million people, and asks how machine learning can help map pH in a place too large and geologically tangled to model grain by grain.

We unpack how researchers used pH measurements from 18,386 wells, information about soils and sediments, and estimates of groundwater age and flowpath length to predict where water is more acidic or more alkaline. Along the way, we talk about glaciers as messy aquifer builders, carbonate minerals as natural antacids, and why this kind of map is useful for planning monitoring—but not a substitute for testing an individual well.

Citation: Stackelberg, P.E., Belitz, K., Brown, C.J., Erickson, M.L., Elliott, S.M., Kauffman, L.J., Ransom, K.M., and Reddy, J.E. (2021). Machine Learning Predictions of pH in the Glacial Aquifer System, Northern USA. Groundwater, 59(3), 352–368. https://doi.org/10.1111/gwat.13063

Disclosure: This Waterlines episode uses AI-generated voices for the host conversation.

Takeaway: At this Bay of Fundy beach, the tide pushes seawater through the sand and gravel like a giant rinse cycle, sending it back to the coast in less than a day.Some coasts don’t just meet the ocean at the surface; they exchange water underground, tide after tide. This episode visits Advocate Beach in Nova Scotia, on the Bay of Fundy, where some of the world’s largest tides push seawater into a steep sand-and-gravel beach and pull it back out through the seabed. That hidden circulation can move nutrients and chemicals into coastal waters, shape shoreline ecosystems, and change how scientists think about beaches in a warming, changing coastal world.We unpack how researchers measured submarine groundwater discharge using seepage meters, radon and radium tracers, and water-level physics; why this mega-tidal beach showed very high discharge rates; and why most of the water was not fresh groundwater from land, but seawater rapidly recycled through the beach in less than a day.Citation: Craddock, R. D., Mohammed, A. A., Tamborski, J. J., & Kurylyk, B. L. (2024). Submarine Groundwater Discharge at a Mega-Tidal Beach. Hydrological Processes, 38, e15319. https://doi.org/10.1002/hyp.15319Disclosure: This Waterlines episode package is written from the cited paper and is designed for production with AI-generated voices.

Takeaway: As Arctic permafrost thaws, the coast can gain hidden freshwater plumbing, but a rising sea can press back and slow that flow.

Arctic coasts are not just losing ice at the surface; the ground beneath them is changing the routes water takes to the ocean. That matters because groundwater can carry carbon, nutrients, salt, metals, and other dissolved material into coastal ecosystems that people and wildlife depend on. In this episode, we unpack a modeling study that asks a surprisingly practical question: as the Arctic warms and seas rise, will more groundwater leak toward the coast, or less?

The paper shows that the answer depends on a tug-of-war. Warming can thaw new pathways through permafrost, letting freshwater move toward the shore. Sea-level rise can push back by flattening the downhill pressure difference that drives groundwater seaward. In some scenarios, coastal groundwater discharge increases; in others, it drops. We translate that into everyday terms: frozen ground acting like a basement wall, thawed zones acting like hidden pipes, and the ocean acting like a rising water tank pressing against the land.

Citation: Guimond, Julia A., Aaron A. Mohammed, Michelle A. Walvoord, Victor F. Bense, and Barret L. Kurylyk. “Sea-level rise and warming mediate coastal groundwater discharge in the Arctic.” Environmental Research Letters 17, no. 4 (2022): 045027. https://doi.org/10.1088/1748-9326/ac6085.

Disclosure: This Waterlines episode package is written for public-science storytelling and uses AI-generated voices for the hosts.

Takeaway: Frozen soil is not a sealed lid; old roots and cracks can let snowmelt sneak down until that water refreezes and plugs the way.

Winter runoff shapes spring floods, wetland refilling, groundwater supplies, and the movement of nutrients from fields into streams and aquifers. But frozen ground is not as simple as a hard lid on the landscape. This episode follows a modeling study from the Canadian Prairies that asks a deceptively everyday question: when snow melts on frozen soil, does the water run away, soak in, or recharge groundwater below? The answer depends on tiny shortcuts in soil—old root holes, worm channels, and cracks—that can let meltwater move quickly downward, until that same water refreezes and plugs the path for the next melt.

We unpack how researchers built a hillslope model with two kinds of soil space: the ordinary fine pore network, and larger “macropores” that act like back roads for water. Their simulations show why one midwinter thaw may mostly infiltrate, while later melt events can produce more runoff after earlier water freezes inside the shortcuts. We also talk about why some groundwater recharge can begin before the whole ground thaws, what this means for prairie pothole wetlands and farm landscapes, and why warming winters may change both flood timing and recharge in ways that are not always obvious.

Paper featured: Mohammed, A. A., Cey, E. E., Hayashi, M., & Callaghan, M. V. (2021). Simulating preferential flow and snowmelt partitioning in seasonally frozen hillslopes. Hydrological Processes, 35(8), e14277. https://doi.org/10.1002/hyp.14277

This Waterlines episode uses AI-generated voices for the hosts, with human-directed scripting and review for clarity and scientific accuracy.

Takeaway: Frozen soil is not just a hard lid; if open pores stay clear, snowmelt can rush downward, but if those same pores ice up, the water is pushed across the surface.

Spring melt can look simple from above: snow disappears, puddles form, streams rise. But just below our boots, frozen soil may either block meltwater like a lid or let it dive through hidden cracks and wormhole-like pores. This episode explores why that split matters for floods, groundwater recharge, farm nutrients, road salt, and water planning in cold regions as winters become more variable. We unpack a modeling study that gives frozen soil two “personalities”: the dense soil matrix and fast-flowing macropores. The researchers tested whether a physically based model could reproduce rapid infiltration and drainage through frozen structured soils, and how freezing inside those pathways can later plug them and turn meltwater into runoff. Along the way, we talk about what models can reveal, what they simplify, and why field observations are still essential. Citation: Mohammed, A. A., Cey, E. E., Hayashi, M., Callaghan, M. V., Park, Y.-J., Miller, K. L., & Frey, S. K. (2021). Dual-permeability modeling of preferential flow and snowmelt partitioning in frozen soils. Vadose Zone Journal, 20, e20101. https://doi.org/10.1002/vzj2.20101. This episode uses AI-generated voices.