Currently browsing category

volcanoes

A very pretty map

Upper Tana landforms and rivers

The flickr caption reads: “Map of the the Upper Tana landforms and rivers published in ‘Nature’s Benefits in Kenya Nature’s Benefits in Kenya: An Atlas of Ecosystems and Human Well-Being,’ 2007 (image credit: ILRI and the World Resources Institute, the Department of Resource Surveys and Remote Sensing of the Kenya Ministry of Environment and Natural Resources, and the Central Bureau of Statistics of the Kenya Ministry of Planning and National Development).”

I really like how the topography and rivers show up on this map, but I wish it had a clearer key of the colors and of the black-outlined areas. Are they watersheds or political subdivision? [Note the original source clears it up: They are political district boundaries]

Used under a Creative Commons license from ILRI on Flickr.

EGU Abstract: Potential impact of lava flows on regional water supplies: case study of central Oregon Cascades volcanism and the Willamette Valley, USA

This abstract was just submitted to the European Geosciences Union meeting for a session on “NH9.9. Natural hazard impact on technological systems and urban areas.” I won’t get to go to Vienna in April, but at least a little bit of my science will. Thanks to Natalia for finding a graceful way to integrate our work.

Potential impact of lava flows on regional water supplies: case study of central Oregon Cascades volcanism and the Willamette Valley, USA

Natalia I. Deligne, Katharine V. Cashman, Gordon E. Grant, Anne Jefferson

Lava flows are often considered to be natural hazards with localized bimodal impact – they completely destroy everything in their path, but apart from the occasional forest fire, cause little or no damage outside their immediate footprint. However, in certain settings, lava flows can have surprising far reaching impacts with the potential to cause serious problems in distant urban areas. Here we present results from a study of the interaction between lava flows and surface water in the central Oregon Cascades, USA, where we find that lava flows in the High Cascades have the potential to cause considerable water shortages in Eugene, Oregon (Oregon’s second largest metropolitan area) and the greater Willamette Valley (home to ~70% of Oregon’s population). The High Cascades host a groundwater dominated hydrological regime with water residence times on the order of years. Due to the steady output of groundwater, rivers sourced in the High Cascades are a critical water resource for Oregon, particularly in August and September when it has not rained for several months. One such river, the McKenzie River, is the sole source of drinking water for Eugene, Oregon, and prior to the installation of dams in the 1960s accounted for ~40% of river flow in the Willamette River in Portland, 445 river km downstream of the source of the McKenzie River. The McKenzie River has been dammed at least twice by lava flows during the Holocene; depending the time of year that these eruptions occurred, we project that available water would have decreased by 20% in present-day Eugene, Oregon, for days to weeks at a time. Given the importance of the McKenzie River and its location on the margin of an active volcanic area, we expect that future volcanic eruptions could likewise impact water supplies in Eugene and the greater Willamette Valley. As such, the urban center of Eugene, Oregon, and also the greater Willamette Valley, is vulnerable to the most benign of volcanic hazards, lava flows, located over 100 km away.

Galapagos Conference Website

I’m working on a review paper on evolution of volcanic ocean islands coming out of the Chapman Conference on the Galapagos I participated in last summer. Rather handily, the conference organizers have put together a nice website with all of the talks, posters, and field trip guides. If you are interested in any of the earth science aspects of volcanic ocean islands, ridge-plume interactions, etc., check out the website. Later this year there will also be an AGU monograph arising from the Conference. I’ll post details when it is published.

basalt, water, waves, and mangroves

Lagoon-Ocean tie channel, Isabella Island, Galapagos, July 2011. Photo by A. Jefferson. Click the image to see more Galapagos photos from me on flickr.

AGU 2011 abstract: Controls on the hydrologic evolution of Quaternary volcanic landscapes

The following talk will be presented in the 2011 AGU fall meeting session on “EP41F. Posteruptive Processes Operating on Volcanic Landscapes I” on Thursday, December 8th from 9:15 to 9:30 am.

Controls on the hydrologic evolution of Quaternary volcanic landscapes
Anne J. Jefferson and Noemi d’Ozouville

1. Geography and Earth Sciences, University of North Carolina at Charlotte, Charlotte, NC, United States.
2. UMR 7619 Sisyphe CNRS & UPMC, Universite Paris 6, Paris, France.

Conceptual models that explain the evolution of young volcanic landscapes require the prominent inclusion of processes which affect partitioning of water between surface and subsurface flows. Recently emplaced lava flows have no surface drainage, with infiltration to groundwater as the dominant hydrologic process. Older volcanic landscapes are often dominated by extensive drainage networks, fed by permanent or intermittent streams, which have deeply dissected the constructional topography. Drainage density, topography, and stream and groundwater discharge provide readily quantifiable measures of hydrologic and landscape evolution on volcanic chronosequences. We will use examples from the High Cascades, Galapagos, and elsewhere to illustrate the trajectories and timescales of hydrologic evolution.

We suggest that the surface-subsurface water partitioning is a function of volcanic architecture, climate-driven processes, and water-rock interactions. We will show that in mafic volcanic areas, climate-driven processes (such as weathering and dust deposition) control landscape evolution, while explosive eruptive products may be important for local hydrology. In the High Cascades, where precipitation exceeds 2 m/yr, landscape dissection has obliterated constructional morphology within 1 million years, while in the more arid Galapagos, million year old landscapes are largely undissected. Conversely, localized groundwater perching on pyroclastic layers or paleosols has been characterized in the Galapagos, but not in the Cascades, where pyroclastic activity is more limited in extent. In areas where explosive activity, including phreatomagmatism, dominates volcanism, the evolution of hydrology and topography occurs much more rapidly than in landscapes created by effusion. Hydrothermal circulation and water-rock interactions may play an important role in reducing deep permeability and altering subsurface flowpaths in some volcanic landscapes. Observed chronosequences can be complicated by juxtaposition of different age deposits, post-emplacement faulting, uplift or subsidence, and climate change, so detailed understanding of the landscape’s geologic history is a prerequisite for appropriate interpretation of hydrologic evolution in volcanic landscapes.

Lush vegetation in a pit crater on Santa Cruz Island

A "pit crater" in the highlands of Santa Cruz Island in the Galapagos shows preferential vegetation growth at the contact between lava flows, probably where water is more available. Photo by A. Jefferson.

Chapman Abstract: Top down or bottom up? Volcanic history, climate, and the hydrologic evolution of volcanic landscapes

In July 2011, Anne was a plenary speaker at the Chapman Conference on The Galápagos as a Laboratory for the Earth Sciences in Puerto Ayora, Galapágos. Anne was tasked with reviewing the state-of-knowledge of volcanic island hydrology and identifying pressing questions for future research in this 40 minute talk. The following is the abstract which she submitted when she began the task.

Top down or bottom up? Volcanic history, climate, and the hydrologic evolution of volcanic landscapes

Volcanic landscapes are well suited for observing changes in hydrologic processes over time, because they can be absolutely dated and island chains segregate surfaces of differing age. The hydrology of mafic volcanic landscapes evolves from recently emplaced lava flows with no surface drainage, toward extensive stream networks and deeply dissected topography. Groundwater, a significant component of the hydrologic system in young landscapes, may become less abundant over time. Drainage density, topography, and stream and groundwater discharge provide readily quantifiable measures of hydrologic and landscape evolution on volcanic chronosequences. In the Oregon Cascades, for example, the surface drainage network is created and becomes deeply incised over the same million-year timescale at which springs disappear from the landscape. But chronosequence studies are of limited value if they are not closely tied to the processes setting the initial conditions and driving hydrologic evolution over time.

Landscape dissection occurs primarily by erosion from overland flow, which is absent or limited in young, mafic landscapes. Thus, volcano hydrology requires conceptual models that explain landscape evolution in terms of processes which affect partitioning of water between surface and subsurface flows. Multiple conceptual models have been proposed to explain hydrologic partitioning and evolution of volcanic landscapes, invoking both bottom up (e.g., hydrothermal alteration) and top down processes (e.g., soil development). I suggest that hydrologic characteristics of volcanic islands and arcs are a function of two factors: volcanic history and climate. We have only begun to characterize the relative importance of these two drivers in setting the hydrologic characteristics of volcanic landscapes of varying age and geologic and climatic settings.

Detailed studies of individual volcanoes have identified dikes and sills as barriers to groundwater and lava flow contacts as preferential zones of groundwater movement. Erosion between eruptive episodes and deposits from multiple eruptive centers can complicate spatial patterns of groundwater flow, and hydrothermal alteration can reduce permeability, decreasing deep groundwater circulation over time. Size and abundance of tephra may be a major geologic determinant of groundwater/surface water partitioning, while flank collapse can introduce knickpoints that drive landscape dissection. The combination of these volcanic controls will set initial conditions for the hydrology and drive bottom up evolutionary processes.

Climatic forcing drives many top down processes, but understanding the relative effectiveness of these processes in propelling hydrologic evolution requires broader cross-site comparisons. The extent of weathering may be a major control on whether water infiltrates vertically or moves laterally, and we know weathering rates increase until precipitation exceeds evapotranspiration. Weathering by plant roots initially increases porosity, but accumulation of weathered materials, such as clays in soils, can reduce near-surface permeability and promote overland flow. Similarly, eolian or glacial inputs may create low permeability covers on volcanic landscapes.

View into the crater of Sierra Negra Volcano on Isabella Island, Galapagos

View of the 2005 lava inside the crater of Sierra Negra Volcano on Isabella Island, Galapagos. Photo by A. Jefferson

Scenic Saturday: Ropy pahoehoe on a biogenic beach

Cross-posted at Highly Allochthonous

Anne on a ropy pahoehoe flow on the beach

Anne enjoying the scenery on Isabella Island, Galápagos, July 2011

In this inaugural Scenic Saturday post, I offer up very happy volcano/landscape nerd enjoying the stunning geologic scenery on Isabella, Galápagos Islands, July 2011. I was there as a participant in the Chapman Conference on the Galápagos as a Laboratory for the Earth Sciences. I may manage to blog in more detail about the islands and the conference, but for now enjoy the scenery, just as I did on my first few days in the archipelago.

In the image above, I’ve got my back to the village of Villamil, on the southern flank of Sierra Negra volcano, and I’m actually sitting on some of the oldest exposed lavas from that volcano. You’re looking at the crust of a pahoehoe flow that is probably about 5000 to 9000 years old. A short distance up the beach, I could peek under the skin of the lava and walk a few meters into a lava tube. The floor of this lava tube was below sea level and covered by sea water, so this was really a chance to experience the water table in a very macro-pore.

Lava tube, geologist for scale

Lava tube, USGS scientist for scale. Isabella Island, Galápagos, July 2011

The lava along this stretch of seafront is largely covered by sand that is clearly not basaltic. Instead it is made of little bits of broken shells and sea urchins from the incredibly rich marine ecosystem that surrounds the islands.

Beach sediments

Beach close-up, near Villamil, Isabella Island, Galápagos, July 2011

Elsewhere, the biogenic beach was covered by rather more living parts of the marine ecosystem. This sea lion put on a quite a performance for some appreciative visitors to a mangrove lagoon (and freshwater spring).

Isabella 105

Sea lion, sand, and mangrove roots, near Villamil, Isabella Island, Galápagos, July 2011

When a tree falls in a stream, there's always something around to make use of it.

Cross-posted at Highly Allochthonous (for obvious reasons)

ResearchBlogging.org Allochthonous may have some obscure usage related to rocks, but in ecology, allochthonous material is a major concept that underpins thinking about nutrient cycling and food web dynamics. In its most general definition, allochthonous material is something imported into an ecosystem from outside of it. Usually, ecologists are thinking about organic matter and the nutrients (C, N, and P) that come with it.

Allochthonous material in the form of coarse particulate organic matter in a mountain stream in Oregon.

Allochthonous material in the form of coarse particulate organic matter in a mountain stream in Oregon.

In streams, allochthonous material includes leaves that fall or are washed into the water and branches and trees that topple into the stream. These would both be called “coarse particulate organic matter” or “CPOM” in the lingo of stream ecologists. In headwater streams, especially in forested areas, there is a lot of CPOM, and the community of aquatic organisms has a high proportion of “shredders” – the critters that that feed on CPOM and break it up into tinier bits called “fine particulate organic matter” or FPOM. In turn, organisms called “collectors” make use of the FPOM by filtering it from the water or accessing it in the sediments. [Allochthonous material can also include dissolved organic matter (DOM) carried into the stream by overland or subsurface flow.]

Schematic illustration of the River Continuum Concept, as modified from Vannote et al. (1980)

Schematic illustration of the River Continuum Concept, as modified from Vannote et al. (1980)

As you move downstream from the headwaters toward medium-sized rivers, the stream channel becomes wider and allochthonous input from overhanging forest and riparian vegetation decreases in abundance and importance relative to primary production (or autochthonous organic mattter) driven by available sunlight. In other words, algae and aquatic plants become the most important food producers. Organisms called “grazers” who scrape algae from surfaces become an important component of the aquatic food web, and grazers become less abundant.

Farther downstream, the ecosystem shifts again, as there is so much FPOM moving with the water and sediment, that collecters far outnumber either shredders or grazers. There’s still allochthonous input from the banks and being carried in by tributaries, and there’s still primary production occurring in the stream, but upstream “system inefficiency” or “leakage” in the processing of nutrients and organic material lets large river aquatic communities be based on material washing in from upstream.

The adjustment of river ecosystems in a downstream fashion that I’ve described above is part of the “river continuum concept”, described by Vannote and colleagues in 1980 in the Canadian Journal of Fisheries and Aquatic Science, and it is one of the unifying principles of modern stream ecology. At its root, the river continuum concept is driven by the relative proportion of allochthonous to autochthonous organic matter inputs to the stream.

While I’m not an ecologist, I was raised by one and I work with them, so when I hear the word allochthonous, I pictures leaves and logs in streams, rather than anything to do with rocks. So, I’ll end this post with some nice pictures of allochthonous material.

An overwhelming amount of allochthonous material in a headwater stream, Gaston County, North Carolina

An overwhelming amount of allochthonous material in a headwater stream, Gaston County, North Carolina. One of my MS students showed that debris jams like this were the biggest driver of groundwater-stream interactions, variations in sediment size, and changes in water chemistry in these tiny streams.

Allochthonous organic material in Clark Creek, Charlotte. High water has washed branches and leaves into the creek, where they got hung up on the riffle (or riprap).

Allochthonous organic material in Clark Creek, Charlotte. High water has washed branches and leaves into the creek, where they got hung up on the riffle (or riprap). What role do natural and artificial geomorphic structures (with their FPOM trapping abilities) play in promoting ecosystem health in urban streams? My colleagues and I are trying to find out.

Large wood jam on Mallard Creek, near Harrisburg, NC

Large wood jam on Mallard Creek, near Harrisburg, NC. For several years, I've taken my Fluvial Processes class to this spot, in part so that they can observe the geomorphic effects of wood in streams.

Wood in streams is utilitarian. During my PhD, I used stable large logs to cross streams and attach equipment.

I use large logs to cross streams and attach equipment. Here, in a spring-fed stream in Oregon, with extremely stable water levels and no floods, allochthonous material that falls into the stream stays where it falls and forms a substrate for a fabulous community of mosses and ferns.

Not a stream. Allochthonous input onto the surface of a lava flow, from the edge of a forest.

Not a stream. Here we are looking at allochthonous input onto the edge of a lava flow, from the forest beyond. On this young lava flow (in the Oregon Cascades), I found substantially greater soil depth near the edge of the flow, where organic acids from decaying allochthonous organic matter had probably sped up the weathering process, as well as contributing directly to the soil. In my PhD dissertation, one subsection had "allochthonous inputs" for a title.

Vannote, R., Minshall, G., Cummins, K., Sedell, J., & Cushing, C. (1980). The River Continuum Concept Canadian Journal of Fisheries and Aquatic Sciences, 37 (1), 130-137 DOI: 10.1139/f80-017

New publication: Coevolution of hydrology and topography on a basalt landscape in the Oregon Cascade Range, USA

ResearchBlogging.org

How does a landscape go from looking like this…

<2000 year old landscape on basaltic lava with no surface drainage

~1500 year old basaltic lava landscape with no surface drainage

to looking like this?

2 Million year old landscape on basaltic lava

2 Million year old landscape on basaltic lava. Note steep slopes and incised valleys

Find out in my new paper in Earth Surface Processes and Landforms.

Hint: Using a chronosequence of watersheds in the Oregon Cascades, we argue that the rates and processes of landscape evolution are driven by whether the water sinks into the lava flows and moves slowly toward springs with steady hydrographs or whether the water moves quickly through the shallow subsurface and creates streams with flashy hydrographs. Further, we suggest that this water routing is controlled by an elusive landscape-scale permeability which decreases over time as processes like chemical weathering create soil and clog up pores in the rock. And as a bonus, because of the high initial permeability of basaltic landscapes, the formation of stream networks and the dissection of the landscape appears to take far longer than in places with less permeable lithologies.

Jefferson, A., Grant, G., Lewis, S., & Lancaster, S. (2010). Coevolution of hydrology and topography on a basalt landscape in the Oregon Cascade Range, USA Earth Surface Processes and Landforms, 35 (7), 803-816 DOI: 10.1002/esp.1976

When it rains a lot and the mountains fall down

Cross-posted at Highly Allochthonous

2006 debris flow deposit in the Eliot Glacier drainage, north flank of Mount Hood (Photo by Anne Jefferson)

The geo-image bonanza of this month’s Accretionary Wedge gives me a good reason to make good on a promise I made a few months ago. I promised to write about what can happen on the flanks of Pacific Northwest volcanoes when a warm, heavy rainfall hits glacial ice at the end of a long melt season. The image above shows the result…warm heavy rainfall + glaciers + steep mountain flanks + exposed unconsolidated sediments are a recipe for debris flows in the Cascades. Let me tell you the story of this one.

It was the first week of November 2006, and a “pineapple express” (warm, wet air from the tropic Pacific) had moved into the Pacific Northwest. This warm front increased temperatures and brought rain to the Cascades…a lot of rain. In the vicinity of Mt. Hood, there was more than 34 cm in 6 days, and that’s at elevations where we have rain gages. Higher on the mountain, there may even have been more rain…and because it was warm, it was *all* rain. Normally, at this time of year, the high mountain areas would only get snow.

While it was raining, my collaborators and I were sitting in our cozy, dry offices in Corvallis, planning a really cool project to look at the impact of climate change on glacial meltwater contributions to the agriculturally-important Hood River valley. Outside, nature was opting to make our on-next field season a bit more tricky. We planned to install stream gages at the toe of the Eliot and Coe glaciers on the north flank of Mt. Hood, as well as farther downstream where water is diverted for irrigation. But instead of nice, neat, stable stream channels, when we went out to scout field sites the following spring, we were greeted by scenes like the one above.

Because sometime on 6 or 7 November, the mountain flank below Eliot Glacier gave way…triggering a massive debris flow that roared down Eliot Creek, bulking up with sediment along the way and completely obliterating any signs of the pre-existing stream channel. By the time the flow reached the area where the irrigation diversion occur, it had traveled 7 km in length and 1000 m in elevation, and it had finally reached the point where the valley opens up and the slope decreases. So the sediment began to drop out. And debris flows can carry some big stuff (like the picture below) and like the bridge that was washed out, carried downstream 100 m and turned sideways.

2006 Eliot Glacier debris flow deposit (photo by Anne Jefferson)

2006 Eliot Glacier debris flow deposit (photo by Anne Jefferson)

In this area, the deposit is at least 300 m wide and at least a few meters deep.

Eliot Creek, April 2007 (photo by Anne Jefferson)

Eliot Creek, April 2007 (photo by Anne Jefferson)

With all the big debris settling out, farther downstream the river was content to just flood…

[youtube=http://www.youtube.com/watch?v=J4eduMJU710]
Youtube video from dankleinsmith of the Hood River flooding at the Farmers Irrigation Headgates

and flood…

West Fork Hood River flood, November 2006 from http://elskablog.wordpress.com/2006/11

West Fork Hood River flood, November 2006 from http://elskablog.wordpress.com/2006/11/. For the same view during normal flows, take a look at my picture from April 2007: http://scienceblogs.com/highlyallochthonous/upload/2009/10/IMG_1108.JPG.

and create a new delta where Hood River enters the Columbia.

Hood River delta created in November 2006 (photo found at http://www.city-data.com/picfilesc/picc30876.php)

Hood River delta created in November 2006 (photo found at http://www.city-data.com/picfilesc/picc30876.php

And it wasn’t just Mt. Hood’s Eliot Glacier drainage that took a beating in this event. Of the 11 drainages on Mt. Hood, seven experienced debris flows, including a rather spectacular one at White River that closed the main access to a popular ski resort. And every major volcano from Mt. Jefferson to Mt. Rainier experienced debris flows, with repercussions ranging from downstream turbidity affecting the water supply for the city of Salem to the destruction of popular trails, roads, and campgrounds in Mt. Rainier National Park (pdf, but very cool photos).

In the end, our project on climate change and glacial meltwater was funded, we managed to collect some neat data in the Eliot and Coe watersheds in the summer of 2007, and the resulting paper is wending its way through review. The November 2006 debris flows triggered at least two MS thesis projects and some serious public attention to debris flow hazards in the Pacific Northwest. They also gave me some really cool pictures.