Overview of Day 4 route (purple line), looking south
Leaving Tygh Valley, we climb slowly across a gently sloping surface, which is the top of the Juniper Flat lava flow that came from the High Cascades 6 to 10 million years ago. As we cross the flat we’ll have good views of Mt. Hood to our right, and Mt. Jefferson ahead. Jefferson is the third High Cascade volcano we encounter, and is smaller and less active than either Mt. Adams or Mt. Hood. Its last eruption was twenty to thirty thousand years ago, during the last ice age.
Shortly after we turn off on Reservation Road, we will begin to descend Paquet Gulch, a narrow canyon cut into the rising Laughlin Hills in front of us. The hills are once again made of Columbia River Basalt, and here they have been folded into another long wrinkle so that the originally horizontal layers are now tilted toward us. This makes for an interesting topographic pattern, as you can see the tilted beds making striking “V” shapes in the topography, where gullies cut into the tilted layers.
As we enter the first canyon on Reservation Road, we pass these inverted “V”s on the ridges, which show that the originally flat layers of Columbia River Basalt have been tilted toward us. This means that as we proceed down the canyon, we see progressively older rocks.
This is also a good area to try to spot Mima mounds. Mima mounds are regularly spaced, circular mounds a foot or two high and 10 to 20 feet across. They stud the eastern Oregon landscape in areas mainly where the natural ground surface has not been disturbed. Geologists still argue about the cause, but the best theory to date suggests that long-lived colonies of burrowing rodents slowly build the mounds.
In this image you can see the regular round mounds that cover the ground; they’re quite low and broad, and may be hard to see from the road.
Although they look like goose bumps or turkey skin, these are Mima mounds near Madras, each about 30 feet across. This image is made using lidar, an airborne laser scanner that makes incredibly detailed 3-D maps of the earth’s surface.
Because the rock layers are tilted toward us, descending Paquet Gulch takes us deeper into the layers of rock, so that we will be headed back in time. Shortly we’ll start to see white ash and clay, sand and gravel in the road cuts, along with bright red patches of soil that mean we have entered into rocks of the John Day Formation. This highly varied pile of volcanic rocks was formed 20 to 30 million years ago, as part of an earlier cycle of subduction-related volcanic activity. Shortly after we switchback down into the Warm Springs River canyon, the valley will cut a narrow rock-walled gorge through lava of an ancient volcano. The volcanoes of the John Day formation alternated between producing lava flows like these, and explosive eruptions that blew huge amounts of ash into the air, which settled out to make the layers of bright-colored ash and clay we also see.
At Kah-nee-ta hot springs and resort, there are over 20 different springs, the hottest of which are around 180° F. Why are there hot springs here? Oregon has over 690 known hot springs, some nearly at the boiling point, and they are either associated with the active volcanoes of the High Cascades or with fault zones in eastern Oregon. Shallow bodies of magma beneath the active volcanoes can heat the groundwater, causing hot water rise to the surface; in eastern Oregon, where the earth’s crust is thin and highly fractured, ground water can circulate very deeply along faults to
bring hot water to the surface. Although Kah-nee-ta hot springs are within sight of two major Cascade volcanoes, they are probably the result of deep faulting.
After a short climb out of the Warm Springs River canyon we cross a broad pass and descend toward Simnasho Creek, and as we do we will cross a huge prehistoric landslide that came from the mountain to our right. You may be able to see the headscarp, the high cliff at the top of the landslide, and the jumbled terrain between there and the road that is the body of the slide. This mixture of ash and clay beds with lava flows is very landslide-prone. The landslide we cross is half a mile wide and extends a mile from the headscarp to the bottom of the slide, about the same size as the recent deadly slide in Oso, Washington.
Our route (purple line) crosses a large prehistoric landslide (white outline) a few miles past Kah-nee-ta. You can see the rocky cliff of the landslide headscarp to your right as you ride by (on the left in this image looking north) An older landslide (green outline) covers the opposite hill.
After we reach Highway 26 at Warm Springs, we turn to ride up the canyon of the Deschutes River. The annual pattern of flow of the Deschutes is unusual, and the tables below shows the flow levels for the Deschutes over the last year, and also show the long-term average. What is unusual is how constant the flow is: The yearly average ranges from about 3,800 cubic feet per second (cfs) to about 5,300 cfs – or 1.7 million gallons per minute to 2.4 million gallons per minute – with peak flows about one and a half times the lowest. The John Day River, which enters the Columbia a few miles from the Deschutes, also drains mountainous terrain with a big annual snowpack. Its flow is what one would expect: highly variable, with a huge peak in spring during snowmelt and then very low levels in late summer before the fall rains start. Its range is 100 cfs (4,500 gallons per minute) to 5,000 cfs (2.25 million gallons per minute), with peak flows 50 times the lowest.
The Deschutes lacks the huge snowmelt peak and extreme summer lows because it drains a High Cascades landscape largely made of young lava. Lava can be a very permeable rock, so most of the snowmelt goes directly into the ground and into a huge volcanic aquifer that slowly and steadily feeds hundreds of springs that make up most of the flow of the Deschutes River. This steady flow of cold clear water makes the Deschutes a world-class steelhead river, and a popular summer rafting and kayaking destination.
These graphs show the amount of flow (in cubic feet per second) of the Deschutes and John Day rivers over the last year (blue line) and the long-term average (orange line). The flow axis is logarithmic, so 900 cfs appears much closer to 1,000 than does 1,100.
For the last mile we’re on Highway 26; before we turn off on Pelton Dam Road, we will again cross a huge landslide. This one came from the canyon wall on our left and is a mile wide and a mile from top to bottom. As we proceed up the Deschutes River on Pelton Dam Road we’ll cross several more. See if you can spot the characteristic jumbled topography. A few miles after we leave Highway 26, we will come around a corner and see Pelton Dam ahead, and as we approach the dam the road cuts on our left will have excellent exposures of lava flows of the Columbia River Basalt, which erupted far to the east, overlaying ash and sand and gravel produced by eruptions from the Cascades.
As we finally leave the river and climb out of the canyon on our way into Madras, you’ll notice that the skyline ahead is capped with a rimrock layer. This is a very common phenomenon in these areas where the canyons are eroded into layers of soft sediment and ash with lava flows in between. Erosion will strip away the soft rock, leaving a plateau supported by the harder lava, and the edge of the lava will form a cliff at the top of the canyon.
Once out of the canyon and riding across the plateau, we’ll have a great view of Mt. Hood behind us, Mt. Jefferson on our right, and the Three Sisters ahead.