DEPARTMENT OF THE INTERIOR
U.S. GEOLOGICAL SURVEY







Summary Distribution of Slides and Earth Flows in the San Francisco Bay
Region, California


By  Carl M. Wentworth, Scott E. Graham, Richard J. Pike, 
    Gregg S. Beukelman, David W. Ramsey, and Andrew D. Barron(1)



Future movement of slides and earth flows is most likely to occur 
within the delineated parts of the landscape where 
these landslides have previously moved.





Open-File Report  97-745 C





1997













This report is preliminary and has not been reviewed for conformity with U.S.
Geological Survey editorial standards or with the North American Stratigraphic
Code.  Any use of trade, product, or firm names is for descriptive purposes
only and does not imply endorsement by the U.S. Government.  

This database, identified as 'Summary Distribution of Slides and Earth Flows
in the San Francisco Bay Region, California', has been approved for release
and publication by the Director of the USGS. Although this database has been
reviewed and is substantially complete, the USGS reserves the right to revise
the data pursuant to further analysis and review. This database is released on
condition that neither the USGS nor the U.S. Government may be held liable for
any damages resulting from it use.

_______________________________
(1) MENLO PARK, CA 94025 



SUMMARY DISTRIBUTION OF SLIDES AND EARTH FLOWS IN THE 
SAN FRANCISCO BAY REGION, CALIFORNIA

by 

Carl M. Wentworth, Scott E. Graham, Richard J. Pike, 
Gregg S. Beukelman, David W. Ramsey, and Andrew D. Barron


INTRODUCTION

The distribution of landslides evident in the landscape -- most of which are
slumps, translational slides, and earth flows -- is of interest both for
evaluation of hazard and risk and for use in further study of landslides.
Future movement of such landslides is most likely to occur within and around
the places where they have previously occurred. A map showing the generalized
distribution of these landslides was published by Nilsen and Wright for the
9-county San Francisco Bay region in 1979. Original sources available at the
time of that compilation were incomplete for the region. Landslide mapping
that has since become available provides a basis for revision and extension of
that work, and modern procedures allow this to be done in digital form.

We have digitized category 5 (landslides) from the Nilsen and Wright map,
added equivalent information for Santa Cruz County, revised and filled in the
principal deficiencies of their original map, and added Quaternary surficial
deposits (to delimit areas largely invulnerable to these types of landslides).
The result is reasonably complete coverage for the 10-county region, available
in digital form as (1) vector polygon databases and (2) map-image files for
the whole region and separately for each of the ten included counties.

Nomenclature for landslides is complex (Varnes, 1978). For present purposes,
we use the term slide to include both slumps and translational slides, and
earth flow to represent flows of clayey earth. These kinds of landslides
typically move slowly, in contrast to the rapid movement of debris flows (see
companion report by Ellen and others, 1997). Slides and earth flows deform the
ground surface when they move and remain in the landscape as recognizable
landslide masses, whereas debris flows run down slope and form separate
deposits lower in the landscape.


SUMMARY(2)

Slides and earth flows are landslides that can pose serious hazard to property
in the hillside terrain of the San Francisco Bay region. They tend to move
slowly and thus rarely threaten life directly. When they move -- in response
to such changes as increased water content, earthquake shaking, addition of
load, or removal of downslope support -- they deform and tilt the ground
surface. The result can be destruction of foundations, offset of roads, and
breaking of underground pipes within and along the margins of the landslide,
as well as overriding of property and structures downslope.

_______________________________
(2) This section constitutes the text on the face of the maps.


The best available predictor of where movement of slides and earth flows might
occur is the distribution of past movements (Nilsen and Turner, 1975). These
landslides can be recognized from their distinctive topographic shapes, which
can persist in the landscape for thousands of years. Most of the landslides
recognizable in this fashion range in size from a few acres to several square
miles. Most show no evidence of recent movement and are not currently active.
Some small proportion of them may become active in any one year, with
movements concentrated within all or part of the landslide masses or around
their edges.

These maps and databases provide a summary of the distribution of landslides
evident in the landscape of the San Francisco Bay region. Original
identification and map delineation of these landslides required detailed
analysis of the topography by skilled geologists, a task generally
accomplished through the study of aerial photographs. Such original landslide
maps are now available for most of the region at scales of 1:24,000 - 1:62,500
(Pike, 1997). The summary map presented here makes selected use of these
original maps and the 9-county compilation by Nilsen, Wright, and others
(1979) to provide a basis for initial evaluation of areas vulnerable to
slumps, translational slides, and earth flows in the region. 

The summary map modifies and improves the compilation by Nilsen and Wright,
which was prepared from sources available in the mid-1970's. The generalized
landslide distribution shown on that map has here been improved in areas where
the 1970's sources were notably deficient (Figure 1), has been extended to
include Santa Cruz County, and includes the distribution of surficial deposits
that define landscape not generally vulnerable to these kinds of landslides.
The method of compilation and resolution of 1:125,000 (1 inch = 2 miles)
limits use of the map to regional considerations. For more detailed
information, see the maps listed by Pike (1997) or consult local officials or
private consultants. 


MAPS 

A regional map (1:275,000) and maps for each of the ten counties in the region
(1:125,000) are included here as postscript plot files; they show the
distribution of landslides in relation to details of the landscape and to
roads and towns. The several categories of landslide (see MAP UNITS) are shown
in color on a background of shaded relief (Graham and Pike, 1997) together
with roads and other cultural features (Aitken, 1997). 


MAP UNITS

Four units are distinguished on the maps:

  Mostly Landslide - consists of mapped landslides, intervening areas 
	typically narrower than 1500 feet, and narrow borders around 
	landslides; defined by drawing envelopes around groups of mapped 
	landslides.
  Many Landslides - consists of mapped landslides and more extensive
	intervening areas than in 'Mostly Landslide'; defined by excluding 
	areas free of mapped landslides; outer boundaries are quadrangle and 
	county limits to the areas in which this unit was defined.
  Few Landslides - contains few, if any, large mapped landslides, but locally
	contains scattered small landslides and questionably identified larger
	landslides;  defined in most of the region by excluding groups of mapped
	landslides but defined directly in areas containing the 'Many 
	Landslides' unit by drawing envelopes around areas free of mapped 
	landslides.
  Flat Land - areas of gentle slope at low elevation that have little or no
	potential for the formation of slumps, translational slides, or earth 
	flow except along stream banks and terrace margins; defined by the 
	distribution of surficial deposits (Wentworth, 1997).

A fifth unit, combined with 'Few Landslides' above, is distinguished in the
polygon databases. It was defined in northern Sonoma County by delineating
areas free of mapped landslides (see ADDITION OF OTHER LANDSLIDE INFORMATION,
below).


FLATLANDS

Slides and earth flows do not occur on nearly flat ground -- they require
slopes that are steep and long enough to permit failure. We can thus exclude
gently sloping ground from principal consideration. Nilsen and Wright used a
slope boundary of 15 percent for this purpose. A similar criterion is the
boundary between hillsides and areas of recent alluvial deposition. This
boundary typically occurs at a slope of about 15 percent. This criterion has
the advantage over slope alone of being limited to the lowland areas and
excluding such other areas of low slope as hilltops and sidehill benches. We
have obtained this alluvial boundary from the regional materials map
(Wentworth, 1997).


LANDSLIDE CATEGORY OF NILSEN AND WRIGHT

The principal source of information used to define the distribution of slides
and earth flows in the region is category 5 (landslides) of the regional slope
stability map of Nilsen and Wright (1979). The category 5 areas are a
generalization of the distribution of mapped landslide deposits recognizable
in the terrain, consisting principally of slumps, translational slides, and
earth flows. 

An important limitation of the Nilsen and Wright map is the varied character
of the landslide mapping used in its compilation. For many areas, landslide
inventory maps of various kinds, and even some detailed engineering geologic
maps, were available, whereas elsewhere only general geologic maps were
available. Some of the landslide inventory maps delineated only the most
obvious landslides in the landscape, whereas others represented a thorough
effort to identify all recognizable landslides.

From these varied sources, Nilsen and Wright prepared a generalization of the
distribution of the landslides by drawing envelopes around areas containing
any type or size of mapped landslide that was within 1,000-1,500 feet of
another landslide. Envelopes were also drawn around groups of landslides in
such topographic settings as the same hillslope or creek bank, under the logic
that such groups of landslides have a common local cause. Isolated landslides
were represented directly where large enough, and inversely, inlying areas
larger than 1,000-1,500 feet in diameter that lacked landslides were similarly
delineated. 

The result was subdivision of the hillside terrain of the region into two
categories, one that contained scattered landslides together with intervening
ground typically as wide as 1,000-1,500 feet, and a second that contained no
mapped landslides. (In the present report these two categories are described
as 'Mostly Landslide' and Few Landslides', respectively.) Although generally
consistent, in detail the content of each category depends on the type of
landslide mapping represented by each compilation source. Where the landslide
mapping was thorough, for example, the non-landslide category contains few if
any mappable landslides, whereas in the areas for which geologic maps were
used as sources, the non-landslide category may contain numerous landslides
not deemed important in depicting the areal geology. 


ADDITION OF OTHER LANDSLIDE INFORMATION

Additions to the Nilsen and Wright compilation focussed on those areas lacking
any information and those for which Nilsen and Wright used general geologic
maps as their source (see Figure 1).

Landslide areas were added for Santa Cruz County from the work of Cooper-Clark
and Associates (1975) by drawing envelopes around landslides and groups of
landslides in the fashion used by Nilsen and Wright. Some small landslides
were excluded in this process to avoid their overemphasis.

In northern Sonoma County, Huffman and Armstrong (1980) provide intricate
landslide mapping at a scale of 1:62,500. Rather than drawing envelopes around
these landslides -- a task for which time was not available -- envelopes were
drawn around areas lacking mapped landslides. The result is similar to
category 5 of Nilsen and Wright, but includes more area of non-landslide. This
category is combined with the 'Few Landslides' category on the maps, but is
distinguished in the map database.

In southeastern Sonoma County, Huffman and Armstrong (1980) provide more
tractable information. Envelopes were drawn around these landslides and groups
of landslides. Other, local additions were made from Huffman and Armstrong in
southern Sonoma County (outside the update areas shown on Figure 1).  In the
northeastern part of the Bay region, the 1:24,000 landslide maps of Dwyer and
others (1976) were similarly used to draw envelopes around landslides and
groups of landslides, although numerous small landslides and questionably
identified larger landslides were excluded on a case-by-case basis.

A large patch of northwestern Marin County is not addressed by any available
landslide mapping. In this area, a map of terrain types prepared by S.D. Ellen
(Ellen and others, 1988) was used in concert with a digital slope map(3)
(30-meter resolution) to interpret landslide distribution. Areas mapped by
Ellen as old erosion surface or 'Hard terrain' together with other areas with
slopes less than about 30 percent were categorized as having Few Landslides.
Those areas mapped by Ellen as 'Soft terrain' where slopes are greater than 30
percent were categorized as being Mostly Landslide, and intervening areas of
'Intermediate terrain' steeper than 30 percent were categorized as having Many
Landslides.

_______________________________
(3)  Digital slope data prepared by Graham from an unpublished compilation of
7.5 minute, 30-meter altitude grids by Graham, Bennett, Pike, and others,
1997.


DIGITAL COMPILATION

The digital compilation was performed in ARC/INFO, a commercial Geographic
Information System (Environmental Research Institute [ESRI], Redlands,
California). The three 1:125,000 map sheets of the region (Aitken, 1997) were
compiled separately and then combined. County databases were clipped from the
composite polygon coverage for the region.

The generalized landslide zones of Nilsen and Wright (1979, category 5 of
plates 1-3) were captured in digital form by scanning publication negatives.
Vector perimeters of the resulting category 5 areas in the raster scans were
determined automatically (GRIDPOLY) and processed to smooth the boundaries and
eliminate polygons with areas less than 4000-5000 square meters (noise in the
data). Registration proved to be a problem. Initial registration using
geographic locations of various marks on the negatives (determined by
comparison with a base map) was improved by rubbersheeting to better fit
quadrangle corners evident in the data. Most misfits in the result are
probably less than 50 meters on the ground.  

The revision lines in the northern part of the region were traced, scanned,
hand-edited on-screen, registered with latitude/longitude intersections and
intersections in the township/range network, and converted to closed polygons.
The information for Santa Cruz County was prepared in the same fashion, using
the original 1:24,000 quadrangles as a source. As an initial step, a coherent
original for tracing was prepared by scanning, projecting, and plotting the
7.5 minute quadrangles together at 1:62,500.

Ellen's terrain map in northwestern Marin county (Ellen and others, 1988) was
scanned and projected and then interpreted on-screen together with the slope
data to produce vector polygons.

The areas of surficial deposits were extracted from Wentworth (1997) by
selecting appropriate polygons, putting them into a separate data layer, and
then dissolving internal lines. The result was a vector polygon layer in which
the polygons were categorized as unmapped (within outer sheet boundaries),
water, surficial deposits, or bedrock.

Combining of the several polygon datasets was performed in bulk for each
sheet, using the INTERSECT and UNION commands in ARC. Because the polygons
from the different sources overlapped, this process resulted in many sliver
polygons of various sizes. These were largely eliminated in two steps: 

1. Assigning unit identities in a priority sequence -- water, surficial
   deposits, mostly landslide, many landslides, and finally the remaining few
   landslides category,

2. Dissolving the boundaries between adjacent polygons with the same resultant
   identities.

This was a complex process that yielded a usable result, although numerous
slivers remain. The hand editing required to produce a more elegant result was
not feasible.


DIGITAL DATABASE

The database for slides and earth flows consists of vector coverages for the
region as a whole and for each of the ten included counties, postscript
map-image files for each of those areas, a postscript version of this text and
separately of Figure 1 (Sources of Landslide Information), and an ASCII
version of this text.

The data files are as follows. (See San Francisco Bay Mapping Team (1997) for
information on how to obtain the data.)

sfbr-sef-dbdesc.ps	-  Postscript version of this text
sfbr-sef-dbdesc.txt	-  ASCII version of this text (without Figure 1)
sfbr-sef-fig1.ps		-  Postscript file for Figure 1 alone

  Exported ARC coverages
    sfbr-sef.e00	-  ten-county San Francisco Bay Region
    al -sef.e00	-  Alameda County 
    cc -sef.e00	-  Contra Costa County
    ma-sef.e00	-  Marin County
    na-sef.e00	-  Napa County
    scl-sef.e00	-  Santa Clara County
    scr-sef.e00	-  Santa Cruz County
    sf-sef.e00	-  San Francisco County
    sm-sef.e00	-  San Mateo County
    sol-sef.e00	-  Solano County
    son-sef.e00	-  Sonoma County
  Postscript map-image files:
    sfbr-sef.ps	-  ten-county San Francisco Bay region
    al-sef.ps	-  Alameda County 
    cc-sef.ps	-  Contra Costa County
    ma-sef.ps	-  Marin County
    na-sef.ps	-  Napa County
    scl-sef.ps	-  Santa Clara County
    scr-sef.ps	-  Santa Cruz County
    sf-sef.ps	-  San Francisco County
    sm-sef.ps	-  San Mateo County
    sol-sef.ps	-  Solano County
    son-sef.ps	-  Sonoma County

The map database itself is relatively simple, consisting of unattributed lines
and attributed polygons in UTM projection (Table 1), with the polygons
assigned to seven categories (see Tables 2 and 3). The polygon database
contains two parallel database items (see Table 4), one numeric and one a
character field.

Table 1.  Map Projection

	projection	UTM	(Universal Transverse Mercator)
	units 		meters
	zone 		10


Table 2.  Field Definition Terms

  Terms used in the description of the polygon database

	ITEM NAME 	name of the database field  (item) 	
	WIDTH		maximum number of digits or characters stored
	OUTPUT		output width
	TYPE 		B- binary integer, F- binary floating point number, 
			    I- ACSII integer, C- ASCII character string
	N.DEC		number of decimal places maintained for floating 
			    point numbers


Table 3.  Content of the Polygon Attribute Table

  The # and - ids are shown for the regional coverage SFBR-SEF;  equivalent
  database fields for the county coverages will be named according to those
  coverage names. 

       ITEM NAME        WIDTH  OUTPUT      TYPE   N.DEC	

	AREA              4    	12     	F      	3	area of polygon in 
							  square meters

	PERIMETER         4    	12     	F      	3	length of perimeter 
							  in meters

	SFBR-SEF#   	  4      5     	B      	-	unique internal control 
							  number

	SFBR-SEF -ID  	  4      5     	B      	-	unique identification 
							  number

	PTYPE            35    	35     	C      	-	text description of 
							  category (see Table 4)

	LSCAT             3      3     	 I      -	numeric category 
							  parallel to PTYPE


Table 4. Map Unit Categories 

  Note that no LSCAT values of 4 remain. 4 represented bedrock areas in the
  input geologic materials polygon layer that were later overlaid by LSCAT
  categories 5-8.

       	  LSCAT	  PTYPE

   	    1 	unmapped	

  	    2 	water

   	    3 	surficial deposits

  	    5 	no mapped landslides (combined with 'few landslides' 
		on the maps)

   	    6 	few landslides

   	    7 	many landslides

   	    8 	mostly landslide


REFERENCES CITED

Aitken, D. S., 1997, digital version of the topographic map of the San
  Francisco Bay region [California], three sheets, 1:125,000: U.S. Geological
  Survey Open-file Report 97-500.
Cooper-Clark and Associates, 1975, Preliminary map of landslide deposits in
  Santa Cruz County, in Seismic Safety Element of the County General Plan: Santa
  Cruz County, California, Planning Department, 1 sheet, map scale 1:62,500.
Dwyer, M.J., Noguchi, N., and O'Rourke, J., 1976, Reconnaissance
  photointerpretation map of landslides in 24 selected 7.5-minute quadrangles in
  Lake, Napa, Solano, and Sonoma Counties, California: U.S. Geological Survey
  Open-file Map 76-74, map scale 1:24,000.
Ellen, S.D., Cannon, S.H., and Reneau, S.L., 1988, Distribution of debris
  flows in Marin County, plate 6, in, Ellen, S.D., and Wieczorek, G.F., eds,
  Landslides, floods, and marine effects of the storm of January 3-5, 1982, in
  the San Francisco Bay region, California: U.S. Geological Survey Professional
  Paper 1434, map scale 1:62,500.
Ellen, S.D., Mark, R.K., Wieczorek, G.F., Wentworth, C.M., Ramsey, D.W., and
  May, T.E., 1997, Principal debris-flow source areas in the San Francisco Bay
  region, California: U.S. Geological Survey, Open-file Report 97-745 E, map
  scales 1:275,000 and 1:125,000.
Graham, Scott E. and Pike, R, J., 1997, Shaded-relief map of the San Francisco
  Bay region, California: U.S. Geological Survey Open-file Report 97-745 B, data
  resolution 30-m pixels.
Huffman, M.E. and Armstrong, C.F., 1980, Geology for planning in Sonoma
  County: California Division of Mines and Geology, Special Report 120,
  Landslides and relative slope stability, map sheets 2A and 2B, map scale
  1:62,500.
Nilsen, T.H., and Turner, B.L., 1975, Influence of rainfall and ancient
  landslide deposits on recent landslides (1950-71) in urban areas of Contra
  Costa County, California: U.S. Geological Survey Bulletin 1388.
Nilsen, T.H., Wright, R.H., and others, 1979, Relative slope stability and
  land-use planning in the San Francisco Bay region, California: U.S. Geological
  Survey Professional Paper 944, map scale 1:125,000.
Pike, R, J., 1997, Index to detailed maps of landslides in the San Francisco
  Bay region, California: U.S. Geological Survey Open-file Report 97-745 D.
San Francisco Bay Landslide Mapping Team, 1997, San Francisco Bay region
  landslide folio: U.S. Geological Survey Open-file Report 97-745 A.
Wentworth, C.M., 1997, General distribution of geologic materials in the San
  Francisco Bay region, California: a digital map database: U.S. Geological
  Survey Open-file Report 97-744, database resolution 1:125,000.
Varnes, David J., 1978, Slope movement and types and processes, Chapter 2 in
  Schuster, R.L., and Krizek, R.J., editors, Landslides: analysis and control:
  Transportation Research Board, National Academy of Sciences, Washington, D.C.,
  Special Report 176.