By U.S. Geological Survey, Geologic Division, Eastern Region (1)
We have performed a preliminary interpretation of the geomorphology and the landslide susceptibility of Honduras based on limited pre-Hurricane Mitch topographic and geologic information available from the U.S. Geological Survey (USGS) library. This assessment demonstrates an approach for providing scientific information that can focus well-informed decisions for relief and recovery efforts following the catastrophic effects of Hurricane Mitch in Central America. This assessment is only intended as an initial framework for such investigations; it is likely that more comprehensive scientific investigations will be conducted.
by Gerald F. Wieczorek, Wayne L. Newell, Peter G. Chirico, and
Gregory S. Gohn (USGS), Remo Nardini (George Washington
University), and Trevor Putbrese (Mary Washington College)
During late October, 1998, Hurricane Mitch released heavy amounts of rain as it passed inland over Honduras, Nicaragua and Guatemala. Reports indicate that up to 2 meters or more of rainfall during one week triggered widesprea landsliding and flooding. Many of these landslides were probably debris flows that started on steep slopes and traveled long distances onto flatter terrain burying villages and disrupting roads and other lines of communication. The geomorphology of Honduras strongly reflects the geologic framework; the countryside can be partitioned into four main types of terrain. These general terrain types include: 1) major stratigraphic groups of bedded sequences, including continental assemblages of volcanic rocks and intermontane basin sediments, marine assemblages of carbonate deposits, and continental assemblages of fluvial and slope deposits, 2) metamorphic complexes of metasedimentary rocks, schists, gneisses, and plutonic rocks, 3) volcanic and associated ejecta blankets that are superimposed on the antecedent topography, and 4) major river deltas along the marine margins.
The major stratigraphic groups (terrain 1) commonly exhibit massive slope failures including rotational slumps, rock block slides, and rock avalanches, which are commonly bounded by breakaway scarps. Shear strength of the weakest interbeds, dip angle of the bedding, slope steepness, presence of faults, joints, and other discontinuities, and undercutting of slopes are important factors. The slopes forming on areas underlain by metamorphic complexes and plutonic rocks (terrain 2) have a deeply weathered saprolite with highly dissected drainage patterns with dense gullying and erosion likely attributable to debris flows during intense rainfall. Thick and widespread deposits of Quaternary alluvium are present at the bases of these slopes.
The topographic relief ranges from a few hundred meters to more than 2000
meters. Some of the relief is constructional, some is erosional, and some is
result of tectonic uplift. Constructional relief is the result of the
materials from the volcanic sources. Erosional relief is most apparent in the
many gorges and canyons, 100 to 300 meters deep, cut through the thick
sequences of continental sediments. The local relief of upland areas
underlain by metamorphic and plutonic rocks may approach 1000 meters.
One high massif of a carbonate assemblage, bounded by faults, has local
relief in excess of 2000 meters. The faulted margins of this massif are
by extensive massive slumps.
The map of annual rainfall for Honduras indicates that normal precipitation
patterns vary from 1 to more than 3 meters per year, an amount sufficient to
develop a deeply weathered regolith. From inspection of the rainfall and
topographic maps, it appears that there is a significant orographic effect
the higher mountains and volcanoes.
Landslide hazards can be assessed by different methods depending upon the complexity of the geologic setting and the availability of necessary data. At the time we conducted this study, information on the locations of landslides triggered by Hurricane Mitch was not available. Consequently, we created an inventory of landslides that had been mapped on previously published 1:50,000-scale geologic maps of Honduras. Most individual landslides that are shown on the 1:50,000-scale maps are at least one-half kilometer in maximum dimension; the largest are several kilometers in both width and length. These moderate- and large-sized landslides are probably mostly deep-seated, in excess of 10-m thick; smaller and shallower landslides, such as most debris flows, are probably under-represented and not individually shown on these maps. These maps (listed in Table 1) cover the general area from the capital city of Tegucigalpa to Lago de Yogoa, which represents a relatively small portion of the entire country.
Three physical factors are generally related to landslide incidence-- topography, geology, and hydrology. If we consider only the type of geologic material, i.e. bedrock geologic units, from which past landslides have initiated, we can obtain an overview of particular geologic materials that have been most prone in the past to slope instability. However, this characterization of geologic units in terms of a landslide incidence map has several limitations for assessing susceptibility. Geologic units consist of rocks of similar age and origin, but some geologic units may contain different lithologies, such as alternating beds of sandstone and shale, that differ widely in their mechanical strength and slope stability. Brief lithologic descriptions of those geologic bedrock units on the 1:50,000-scale maps that are prone to landsliding are given in Table 2.
The Padre Miguel Group (also referred to as the Jutiapo Group on some older maps) is one of the largest individual bedrock geologic units in Honduras and is generally susceptible to landsliding. Within the area of the San Pedro Zacapa 1:50,000-scale quadrangle, approximately 13% of the area mapped as underlain by the Padre Miguel Group has experienced landsliding in the past. Because the geologic map of Honduras was compiled at 1:500,000, many of the individual geologic members shown separately on maps with a scale of 1:50,000 were depicted as a single unit on the national map. For example, the Padre Miguel ignimbrite is a landslide-prone member of the Padre Miguel Group. Using just the high local landslide susceptibility of the Padre Miguel ignimbrite to characterize the entire Padre Miguel Group (which consists of several members regionally) would bias the inferred regional landslide susceptibility of that unit. In this specific example, however, there is additional evidence for a generally high incidence of landsliding throughout much of the area underlain by the Padre Miguel Group.
We interpreted the fan-shaped topography of extensive Quaternary alluvial deposits below some steep, deeply eroded bedrock units to suggest that successive and voluminous debris flows have been generated from those particular bedrock units (Table 3). A few geologic units in this group, without mapped large deep-seated landslides, are probably very susceptible to small and shallow landslides, including debris flows. Some geologic units in this group, such as the Padre Miguel Group, may be prone to both large deep-seated landsliding and more shallow debris flows.
Another caution with this approach for characterizing landslide
is that we know nothing of the cause or trigger of the past landslides.
From studies elsewhere (e.g. Harp and others, 1982) we know that slopes
respond differently to earthquakes than to rainstorms, resulting in
distributions of landslides that characteristically differ depending on
the triggering event. Without knowing the trigger for the inventoried
landslides, the characterization of geologic units in terms of landslide
susceptibility for conditions of intense rainfall from events such as
Mitch is a problematic first approximation that could be improved with
This study has used the geologic map of Honduras (1:500,000) as a basis
for portraying three different levels of landslide susceptibility--high,
and low. The association of geologic units on the 1:500,000-scale
map to degrees of landslide susceptibility is listed in Table 4. The specific
geologic units (Table 1) that included landslides on previously published
geologic maps that we examined have been characterized as having a high
susceptibility to landsliding. Those geologic units (Table 3) which had large
areas of Quaternary deposits suggestive of multiple episodes of debris-flow
deposition on alluvial fans have been termed as having a moderate
susceptibility to landsliding. Those geologic units which did not have either
the above characteristics are presumed to have a low susceptibility to
landsliding, although in specific instances landslides may occur. For
units which were not included in our review of 1:50,000 scale geologic maps,
we did not evaluate their susceptibility to landsliding. Generally, the
units that were not evaluated did not cover large areas on the geologic map
Honduras. During an event of the scale and intensity of Hurricane Mitch, the
specific geologic units we evaluated may also have been stripped of surficial
soils by shallow erosion resulting in downslope inundation or heavy
The geologic maps at 1:50,000 scale (Table 1) were published during the
1960's and early 1970's by the Ministry of Communications and Public
Works, Institute of National Geography and the Director General of Mines
and Hydrocarbon of the Ministry of Natural Resources of Honduras. Some of
the mapping was completed by the University of Texas in Austin. Remo
Nardini, one of the members of this team, is a USGS Volunteer for Science.
Harp, E.L., Wilson, R.C., and Wieczorek, G.F., 1981,
Landslides from the
February 4, 1976, Guatemala earthquake: U.S. Geological Survey
Professional Paper 1204-A, 35 p.
Maintained by: EESP Team Staff
Contact: Peter Chirico (email@example.com)