Ten Million Year Old Drainage At The Santa Rosa Plateau Reconstructed By Analysis

See Ten million year old drainages seen at the Santa Rosa Plateau: Overview for an introduction to this page.

All published geologic maps for the area of the Santa Rosa Plateau show basalt only on the flat-topped mesas. Until May 2007, we also thought those were the only locations of basalt, since those areas of basalt are distinctive by being so flat.

In 2006, we discovered the Santa Rosa Basalt Brodiaea, Brodiaea santarosae, growing at Clay Hill, 0.3 miles west of, and 100 feet below, the nearest mapped basalt on the Mesa de Burro. However, subsequent searches for this species, made with the considerable help of Steve Boyd, found that all other documented occurrences of this species were on areas mapped as basalt.

We initially trusted the geologic maps and our impressions about where the basalt occurred. There were two reasonable hypotheses as to why B. santarosae might grow off the basalt:

We submitted our paper with those hypotheses.

Fortunately, one of the Madroño reviewers asked: have the authors done any chemical analysis that might suggest why one or more populations of the new species grow on soils not derived from basalt?

As a result, we returned to the site in May 2007, and actually looked at the rocks there instead of trusting the maps. We were dumbfounded to find that the areas of Clay Hill where B. santarosae grows are all covered with basalt, even though they are far from being flat! As a result, we were able to revise the paper before it was published to state that all vouchered populations were found on basalt without exception.

In November 2007 we returned to this area to map the basalt in detail, to understand its origin, and made the following unexpected discoveries:

The rest of this page details how the ancient valley was reconstructed from our mapping of the basalt.

The detailed fine-scale map of the basalt from our November 2007 is presented below. See Basalt Survey Near Clay Hill for details about our survey, and the location of this area on a map of the entire Santa Rosa Plateau Ecological Reserve.

The Mesa de Burro is the flat area at extreme right capped with basalt. Two roads that approach the Mesa de Burro are labeled and drawn in purple. A modern-day creek that drains this area of Mesa de Burro is drawn in blue and labeled as Current drainage. Clay Hill is between that drainage and the next east-west drainage to the south.

There are three green / red contours of the current-day basalt. Clay Hill contains two mapped basalt areas that do not touch the basalt at the base of the Mesa de Burro. Three areas have not yet been surveyed, and are outlined in black, with black diagonal lines. It is thus possible that the two areas of basalt on Clay Hill are actually part of one larger area. It appears that those contours will fit together well.

The green contours delineate Santa Rosa Basalt; the red contours delineate basalt of the Santiago Peak Volcanics.

See also Photographs of these basalt areas taken from the Mesa de Burro.

Specimens of B. santarosae occur in each of the three mapped basalt areas that are farthest off the mesas (see map). Amazingly, our surveys for B. santarosae along the roads here in 2006 only found it in those locations. That survey was done without knowledge that these were areas of basalt or that this species was only confined to basalt.

From the above plot, it is a simple matter to draw elevation contours that were sequentially covered by the lava of the Santa Rosa Basalt at the time of its deposition, given the following assumptions (all references to lava and basalt in the reconstruction below now refer only to the Santa Rosa Basalt):

  1. There has been no tilting of the terrain here since the lava was deposited. This is an excellent assumption over the area of this map, since all the nearby basalt mesas are essentially as the same elevation.

    This assumption means that elevations of today's basalt equal the elevations from the time the lava was deposited to within a constant which can be ignored.

  2. There has been only minimal erosion to areas currently covered with basalt. This is again an excellent assumption, since basalt-covered areas essentially only erode when they are undermined. After all, that is why the mesas have persisted.

    Of course, outside the current basalt areas, there has been erosion. Elevations of those areas were higher in the past.

  3. In this small area, erosion removed the basalt at all identical elevations simultaneously.

    This assumption is quite reasonable, for two reasons. First, the lip of the Mesa de Burro is essentially at a constant elevation here. Second, for the most part we are just concerned with erosion from the single drainage indicated in the above map. That drainage is fairly symmetric for the elevations considered below.

Given our survey and assumptions, one can easily reconstruct the ancient drainage by plotting surveyed points by their elevation and whether they currently contain basalt or not:

In the above plot, the points are labeled by their elevation in feet. Black points currently are covered by Santa Rosa Basalt; pink points are not. Areas without points have elevations either at 1860 feet or below, below the lowest lava elevation, or above 1920 feet.

The plot also indicates the bottom of the observed ancient valley seen in the cross-section of Mesa de Burro itself (see Ancient Drainages Seen In Profile for a picture of that valley). The bottom is defined here as the extent of the valley shown in the close-up picture on the referenced webpage, which is a small portion of the total width of the valley near its top.

To reconstruct the ancient 1880 foot contour, consider the points at a current elevation of 1880 feet:

Therefore, when the lava flow first covered those points that still are covered with lava today, the points also at 1880 feet today were not covered with lava until there were later flows at higher elevation.

Hence the elevation contour of the ancient valley that corresponds to an ancient elevation of 1880 feet must be drawn to include only those points currently covered with lava, segregating the points not currently covered with lava to a higher ancient elevation.

The following plot gives the results of carrying out the above procedure for ancient elevations of 1880 feet and 1900 feet:

Amazingly, the reconstructed 1880 foot ancient valley contour runs exactly into the bottom of the ancient valley observed in the Mesa de Burro! We note that nothing about this observed ancient valley was used in the determination of the reconstructed valley; it is completely-independent information. Furthermore, we were taken by surprise at this finding, since we had been expecting the downstream portion of that observed ancient valley to form a ridge in today's topography (inverted topography; see below).

The derived 1880 foot contour is well-defined on the north side of the current valley, but is unconstrained on the south side, except at the uppermost end, since there is no residual lava from the Santa Rosa Basalt there. We have simply drawn the south side to be roughly symmetric to the well-defined north side contour, using dashes to indicate it is not constrained. The derived 1900 foot contour is also well-defined on most of the north side of the current valley, except near the Mesa de Burro since all points to the north are currently covered with lava. No attempt was made to draw a 1900 foot contour on the south side of the valley.

The map above shows that the lava flow of 1880 feet covered only near the bottom of an ancient valley, and it didn't quite extend to the portion of the valley exposed in the west face of the Mesa de Burro.

When 20 feet more depth of lava flowed into this area, it covered a much more extensive area since the ancient valley was so shallow.

The bottom of the reconstructed valley derived from the basalt survey, and the above assumptions, is a perfect fit to the valley exposed in the west face of the Mesa de Burro:

This agreement is strong support for the reconstructed valley, since there were no free parameters used in the reconstruction.


It is quite interesting, and unexpected, that the main course of the ancient drainage lines up almost exactly with today's drainage.

Normally, areas covered with basalt exhibit what is called inverted topography. As erosion removes the basalt layers covering the sides of ancient valleys, usually the exposed rock of those sides is less resistant to erosion than the basalt remaining in the valley. As a result, the ancient ridges between valleys become the valleys of today, and the drainage itself becomes a ridge temporarily topped by basalt.

This inverted topography is seen in the green tongue of Santa Rosa Basalt west of the valley seen in the west side of the Mesa de Burro (see first map above). The current drainage, shown by the blue line in the first map above, heads directly for that green tongue, following the ancient river valley, but then diverts to the north, as shown by the 1880 foot elevation contour in the first map above. The upper part of the valley was formed from mudstone of the Bedford Canyon Formation, which is very easily eroded compared to the basalt, and thus it would be expected to produce the inverted topography seen here.

A priori, it is not so easy to understand why the main course of the ancient drainage has not become inverted as well. In fact, it was our expectation that our mapping would show that the ancient drainage visually seen in the Mesa de Burro would form a ridge connecting to the top of Clay Hill, which is clearly not the case.

The key to solving this mystery was the discovery that the top of Clay Hill, and the sides of the ancient valley, were formed from the ancient basalt of the Santiago Peak Volcanics, just as resistant to erosion as the Santa Rosa Basalt. This removes the expectation of inverted topography for the main part of this valley.

One mystery still remaining is why the valley reformed in the same place, instead of producing a modern valley unrelated either to the ancient valley or inverted topography. After all, at the time the basalt was eroded from the top of the sides of the ancient valley, there could have been a flat area of basalt everywhere, and a new valley might form at any location.

A possible solution to this mystery is that the ancient upper drainages were never totally eliminated by basalt. In the interval between lava flows, the drainages would be re-established on the top of the youngest lava flow.

It is possible that the interval between lava flows was as long as one million years or so, since the variance in the dates for different pieces of lava is on the same order. Even with much shorter intervals, it is possible, and perhaps even probable, that the drainages would persist between lava flows.

There are two possible ways in which the drainages might persist. First, since it took many lava flows to completely cover even the shallow valleys near the Mesa de Burro, deeper valleys lower in the drainage system might have persisted, allowing the upstream portions to become reestablished even after they were covered. Second, any upstream portions above the highest lava flow might recreate lower sections, providing the lava flows did not disrupt the entire drainage system. Of course, both of these processes might occur simultaneously.

A persistent drainage is demonstrated by the valley seen in the top of the Mesa de Burro. Even though the basalt was flat at the end of its deposition, a broad valley has been reestablished on top at the location of the ancient and current drainage.

We thank Phil Armstrong for providing the senior theses of Dan Loera (2003) and Steve Turner (2005) on the Geology of the Santa Rosa Plateau Ecological Reserve to help us in doing our geologic survey. Those theses helped us greatly to distinguish the Bedford Canyon argillite from the Santa Rosa Basalt.

We thank Norrie Robbins for her field assistance on 27 November 2007 which resulted in the discovery that some of the basalt at Clay Hill was from the Santiago Peak Volcanics.

Note on geologic maps: All geologic maps for the Santa Rosa Plateau are large-scale, and it is not possible to note very small patches of one rock type within another larger unit on such maps. Even the map made specifically for the Santa Rosa Plateau Ecological Reserve by Loera and Turner could not be expected to show the geology of every hill on the Reserve.

Go to:

Copyright © 2007 by Tom Chester, Wayne Armstrong and Kay Madore.
Permission is freely granted to reproduce any or all of this page as long as credit is given to us at this source:
Comments and feedback: Tom Chester
Updated 30 November 2007.