A hilly landscape in the Reynolds Creek Critical Zone Observatory, with charred soil in the foreground.
In August 2015, the Soda Fire burned more than 68 square kilometers of land in the Reynolds Creek Experimental Watershed and Critical Zone Observatory (CZO) in Idaho. Credit: Kathleen Lohse

We live in the Anthropocene, a time when the intensity and frequency of disturbances to Earth’s ecosystems are increasing as humanity’s footprint grows and the climate changes. The impacts of some disturbances, like land use changes for agriculture, can persist over decades to centuries and thus represent persistent, or press, disturbances. Some, like floods and fires, occur naturally as short-term pulses. Regardless of duration, all of these events can strongly affect Earth’s critical zone, the thin layer of vegetation, soil, rock, and water in which nearly all the planet’s terrestrial life lives.

The physical structure and chemical composition of the critical zone constrain its responses to extreme events and climate and land use change and are thus key to predicting these responses and to sustaining the ecosystem services from which society benefits [Field et al., 2015]. Emerging research in critical zone science has shown that many Anthropocene disturbances are accelerating the lateral movement of topsoil and associated soil organic matter across landscapes and that some disturbances penetrate more deeply and persist longer than previously appreciated. These and related discoveries, developed from the National Science Foundation’s Critical Zone Observatories (CZOs) and critical zone science more broadly, have profound implications for soil fertility and agriculture, natural climate regulation, ecosystem productivity, and ecosystem resiliency to anthropogenic influences.

Restructuring Through Redistribution

Around the world, extreme pulse events like wildfires are happening more often and with greater intensity as a result of climate change and other human-mediated influences, such as the introduction of invasive grasses, historic fire suppression, woody encroachment, and power lines and houses built in wildlands. These processes are accelerating rates of sediment and organic matter erosion, thus removing soil nutrients and carbon. In many regions, soil is being lost to erosion more rapidly than it is replenished by natural soil formation, which occurs over millennia.

Smoke rises from a line of wildfire burning the landscape in the distance.
The Soda Fire sweeps through the Reynolds Creek CZO in 2015. Wildfires are an example of a pulse disturbance. Credit: Hugo Sindelar

All five CZOs in the western United States have experienced wildfires within the past decade. For example, the Catalina-Jemez CZO in the Coronado National Forest in Arizona and Valles Caldera National Preserve in New Mexico (a CZO partner organization) experienced two human-caused fires, in 2011 and 2013, respectively. In a recently burned catchment at the New Mexico site, erosion rates were 3 orders of magnitude higher than baseline rates, as documented by measurements of suspended sediment loads in streams, repeat airborne and terrestrial lidar data, and cosmogenic isotopes that help estimate soil denudation rates [Orem and Pelletier, 2016]. Indeed, pulsed erosion following fire represented more than 90% of the denudation at this site over long (thousand- to million-year) timescales.

In the Sierra Nevada of California, following the 2013 Rim Fire, the highest rates of postfire erosion and preferential enrichment of pyrogenic carbon in eroded sediments were measured in the most severely burned areas.

These enhanced erosion rates were comparable to findings at the Reynolds Creek CZO in Idaho, where researchers measured losses of sediment and particulate organic carbon (POC) following a wildfire in 2015. Losses in the year after the fire were equivalent to roughly 20 years of export at baseline levels, or 1–2 orders of magnitude higher than 25-year averages established from sediment records and hindcasted POC. These studies, and research from other CZOs documenting floods and hurricanes, highlight the value of long-term monitoring to quantify and understand the magnitude of responses to pulsed disturbances. The work also reveals a tight coupling of geomorphic landscape features and biological processes in regulating these losses.

In addition to demonstrating accelerated lateral movement of soil, researchers have shown how the redistribution of carbon across a landscape by pulse disturbances can restructure the critical zone itself. At the Catalina–Jemez CZO, for example, soil horizons and watersheds that exhibited different spatial distributions of legacy pyrogenic carbon (PyC; soot or black carbon from earlier fires) were homogenized during the Thompson Ridge Fire in 2013 [Galanter et al., 2018]. Near the Southern Sierra CZO in the Sierra Nevada of California, following the 2013 Rim Fire, the highest rates of postfire erosion and preferential enrichment of PyC in eroded sediments were measured in the most severely burned areas. The interactive effect of fire and erosion led to lateral transport of large amounts of soil, carbon, and PyC from soil, particularly from high-severity burn areas, illustrating how postfire erosion can be an important control on carbon dynamics in soil systems [Abney and Berhe, 2018]. Finally, researchers at the Reynolds Creek CZO identified that preferential loading of fine sediments into swale channels by wind-driven, or aeolian, processes in the immediate postfire period, followed by flushing of these sediments via runoff during snowmelt, can be a large source of stream sediment transport [Vega et al., 2020].

Two researchers carrying equipment hike through a charred hilly landscape
Researchers hike to the Macks Creek subcatchment in the Reynolds Creek CZO to instrument the area with soil respiration chambers in October 2015, 2 months after the Soda Fire. Credit: Kathleen Lohse

These findings highlight the role of aeolian and hydrologic erosion processes in redistributing carbon and restructuring the critical zone after disturbances. Movement of soil organic matter to depositional environments dictates carbon balance and nutrient dynamics across landscapes. Studies in the Southern Sierra CZO show that climatically mediated transport of soil by water and wind can significantly affect plant nutrient redistribution within and into soils [Stacy et al., 2015; Aciego et al., 2017]. Further, modeling of soil carbon migration across landscapes and associated carbon fluxes to and from the atmosphere demonstrates that the depth to which carbon is buried can determine whether eroded soil organic carbon is mineralized to carbon dioxide or retained in the soil system, and whether eroded soil organic matter can promote ecosystem productivity downslope [Billings et al., 2019].

Anthropogenic press disturbances also restructure the critical zone. Long-term enhanced erosion due to agriculture can mobilize and redeposit surface soil from uplands to bottomlands, which alters carbon source and sink locations. By coupling models with experimental and observational data, scientists at the Intensively Managed Landscapes CZO, colocated in Illinois, Iowa, and Minnesota, have characterized the lateral and vertical redistribution of soil and carbon by industrial agricultural practices. At upland erosional sites in these landscapes, for example, the rate of soil organic carbon decomposition appears to be slower than the gain from plant residues, generating net sinks for atmospheric carbon dioxide, whereas lowland depositional sites appear to serve as net carbon sources [Yan et al., 2018]. Meanwhile, tile drains, which remove excess water from the subsurface, cause vertical transport and further loss of dissolved organic carbon concurrent with water efflux into nearby waterways.

Preconditioning and Resilience

Disturbances can leave legacies in critical zone processes over geological timescales. At the Susquehanna Shale Hills CZO, critical zone responses to Anthropocene disturbances vary across different types of sedimentary bedrock in the region and are preconditioned by periglacial (glacier-adjacent) processes that were active during cold periods over much of the past 2.6 million years (i.e., the Pleistocene). Indeed, Anthropocene land use is highly correlated with local geology [Li et al., 2018]. For example, agriculture is never present where sandstone bedrock dominates in the region, but it is the dominant land use over calcareous substrates like limestone. Some areas underlain by noncalcareous shales, meanwhile, have cultivation histories, but agricultural productivity in those areas was so poor that most farms were abandoned in the 20th century and forests now dominate.

Satellite image of forested areas and farmland in Pennsylvania
This satellite image shows agriculture as a repeated press disturbance in the vicinity of the Susquehanna Shale Hills CZO in Pennsylvania. Forested areas are present where sandstone bedrock dominates, whereas agriculture dominates on calcareous bedrock. Credit: Google Earth

Geologic legacies appear to shape ecosystem responses to Anthropocene disturbances and, at least in some cases, seem to have enhanced the resiliency of these systems to Anthropocene effects.

On all substrates, periglacial solifluction—the downslope movement of thawed permafrost—transported sediment from hillslopes to valley floors [Del Vecchio et al., 2018], but the depth and composition of valley floor fill and its proximity to anthropogenic activities are distinct for each rock type. Overall, geologic legacies appear to shape ecosystem responses to Anthropocene disturbances. And at least in some cases, periglacial soil movement seems to have enhanced the resiliency of these systems to Anthropocene effects, such that despite roughly 200 years of cultivation, ecosystem productivity has not dramatically declined.

Pleistocene glacial-interglacial cycles also can have legacy effects on the vertical and lateral structure of the critical zone. At the Intensively Managed Landscapes CZO, aeolian activity toward the end of the last glacial cycle covered much of the Midwestern landscape with sediment layers of varying thickness, and ecosystem transitions resulting from simultaneous climate change culminated in prairie and wetland landscapes with organic-rich fertile soils that support modern industrial agriculture [Kumar et al., 2018]. Today the resilience and continuing usefulness for farming of this naturally fertile land are maintained through continuous human intervention. Extensive deployment of tile drains amid these low-permeability and largely flat landscapes allows for early spring deployment of agricultural machinery, for example. Together with substantial nutrient inputs from fertilizers, such interventions result in high crop yields.

The Deep Reach of the Anthropocene

After European settlers instituted row crop agriculture in North America, many more regions experienced soil erosion that far outpaced soil production, and many forested landscapes were transformed from closed-canopy, hardwood forests into fragmented patchworks of land uses dominated by agriculture. Research at CZOs has probed deep into the soil to understand the natural formation of soil profiles [Bacon et al., 2012]; the nutrient and carbon dynamics produced through interactions among plants, microbes, and minerals; and the depths to which human-induced signals of land use change can propagate over human timescales.

These observations extend the known depths to which human modification of natural biogeochemical cycles penetrates Earth’s surface.

Investigators at the Calhoun CZO in South Carolina have quantified lower root densities beneath long-term agricultural fields versus those beneath old-growth forests growing on never-plowed soils [Billings et al., 2018]. Such differences are evident even after roughly 80 years of forest regeneration on former agricultural land. Deep roots perform many functions, such as nutrient uplift, generation of deep soil carbon stocks, deep soil weathering via carbon dioxide (i.e., carbonic acid) generation, and creation of preferential flow paths for water, dissolved solutes, and gases. Persistent modification of these deep soil processes means that many ecosystem services are unintentionally altered for many decades after land use change decisions are implemented.

This work also demonstrated that greater root abundances in old-growth forests versus agricultural land were linked to the greater availability of some forms of soil organic carbon well below the depths at which root abundance differences were observed. The consequences of these differences for soil organic carbon preservation are not yet known, but the result hints that old-growth forests are more effective contributors to deep stores of soil organic carbon than agricultural systems.

A researcher crouches in a deep pit collecting soil samples while another takes notes.
Kathleen Lohse collects sterile soil samples from a pit more than 2 meters deep dug near Upper Sheep Creek in the Reynolds Creek CZO while research assistant Mitch Price takes notes. The effort was part of a cross-CZO project investigating deep microbial life in soils. Credit: Nicholas Patton

These observations extend the known depths to which human modification of natural biogeochemical cycles penetrates Earth’s surface. They prompt compelling hypotheses about the depth of human fingerprints in other anthropogenically modified systems around the world. For example, we might hypothesize that the truncated rooting depths of most annual agricultural systems, compared with perennial systems, limit carbon input and root modification of soil porosity deep in soil profiles.

Critical zone science demonstrates the value of placing physical and biogeochemical ecosystem processes into the context of the Anthropocene by delving deeper in space and time to understand critical zone responses and resiliency to both pulse and press disturbances. The studies highlighted here document the acceleration in modern times of soil and soil carbon movement across landscapes and their redistribution both vertically and laterally. These processes directly influence how vulnerable soil carbon is to being lost back to the atmosphere, where it contributes to rising temperatures.

In many cases, these studies illuminate the powerful role of life in processes historically studied by geologists, as well as how geologic legacies continue to shape responses of the critical zone to human influences today. Discoveries from this work help us understand how ecosystems will function and inform landscape management in the Anthropocene.


Abney, R. B., and A. A. Berhe (2018), Pyrogenic carbon erosion: Implications for stock and persistence of pyrogenic carbon in soil, Front. Earth Sci., 6, 26, https://doi.org/10.3389/feart.2018.00026.

Aciego, S. M., et al. (2017), Dust outpaces bedrock in nutrient supply to montane forest ecosystems, Nat. Commun., 8, 14800, https://doi.org/10.1038/ncomms14800.

Bacon, A. R., et al. (2012), Coupling meteoric 10Be with pedogenic losses of 9Be to improve soil residence time estimates on an ancient North American interfluve, Geology, 40(9), 847–850, https://doi.org/10.1130/G33449.1.

Billings, S. A., et al. (2018), Loss of deep roots limits biogenic agents of soil development that are only partially restored by decades of forest regeneration, Elementa Sci. Anthropocene, 6(1), 34, https://doi.org/10.1525/elementa.287.

Billings, S. A., et al. (2019), Distinct contributions of eroding and depositional profiles to land-atmosphere CO2 exchange in two contrasting forests, Front. Earth Sci., 7, 36, https://doi.org/10.3389/feart.2019.00036.

Del Vecchio, J., et al. (2018), Record of coupled hillslope and channel response to Pleistocene erosion and deposition in a sandstone headwater valley, central Pennsylvania, Geol. Soc. Am. Bull., 130(11–12), 1,903–1,917, https://doi.org/10.1130/B31912.1.

Field, J. P., et al. (2015), Critical zone services: Expanding context, constraints, and currency beyond ecosystem services, Vadose Zone J., 14, 1–7, https://doi.org/10.2136/vzj2014.10.0142.

Galanter, A., D. Cadol, and K. Lohse (2018), Geomorphic influences on the distribution and accumulation of pyrogenic carbon (PyC) following a low severity wildfire in northern New Mexico, Earth Surf. Processes Landforms, 43, 2,207–2,218, https://doi.org/10.1002/esp.4386.

Kumar, P., et al. (2018), Critical transition in critical zone of intensively managed landscapes, Anthropocene, 22, 10–19, https://doi.org/10.1016/j.ancene.2018.04.002.

Li, L., et al. (2018), The effect of lithology and agriculture at the Susquehanna Shale Hills Critical Zone Observatory, Vadose Zone J., 17, 180063, https://doi.org/10.2136/vzj2018.03.0063.

Orem, C. A., and J. D. Pelletier (2016), The predominance of post-wildfire erosion in the long-term denudation of the Valles Caldera, New Mexico, J. Geophys. Res. Earth Surf., 121, 843–864, https://doi.org/10.1002/2015JF003663.

Stacy, E. M., et al. (2015), Soil carbon and nitrogen erosion in forested catchments: Implications for erosion-induced terrestrial carbon sequestration, Biogeosciences, 12, 4,861–4,874, https://doi.org/10.5194/bg-12-4861-2015.

Vega, S. P., et al. (2020), Interaction of wind and cold-season hydrologic processes on erosion from complex topography following wildfire in sagebrush steppe, Earth Surf. Processes Landforms, 45, 841–861, https://doi.org/10.1002/esp.4778.

Yan, Q., et al. (2018), Three-dimensional modeling of the coevolution of landscape and soil organic carbon, Water Resour. Res., 55, 1,218–1,241, https://doi.org/10.1029/2018WR023634.

Author Information

Kathleen A. Lohse (klohse@isu.edu), Idaho State University, Pocatello; Sharon A. Billings, University of Kansas, Lawrence; Roman A. DiBiase, Pennsylvania State University, University Park; Praveen Kumar, University of Illinois at Urbana-Champaign, Urbana; Asmeret Asefaw Berhe, University of California, Merced; and Jason Kaye, Pennsylvania State University, University Park


Lohse, K. A.,Billings, S. A.,DiBiase, R. A.,Kumar, P.,Berhe, A. A., and Kaye, J. (2020), Soil signals tell of landscape disturbances, Eos, 101, https://doi.org/10.1029/2020EO148736. Published on 24 September 2020.

Text © 2020. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.