3.9. Analogue studies in the western USA and the Middle East

The questions to be addressed by the TOPO-EUROPE Initiative will benefit from ongoing research activities in other broadly deforming, but semi-arid intra-continental regions such as the Basin-and-Range Province (USA) or the Eurasian-Arabian plate boundary region across Iran. In order to model the 4D evolution of topography in response to internal and external Earth processes it is essential to know the signal intensity, duration, and the time since when a particular process has been active at a particular location. Such parameters are especially important in regions with large transient gradients in topography or complex fault system geometries. Of particular interest for modelling of the underlying dynamic process also is the rate at which such signals propagate in time and space. A related concern is whether it will be possible to close the observational gap between processes operating at historic time scales and those operating at geologic time scales.

Much progress towards solving these questions has recently been made along the Pacific-North American plate boundary region, western USA. The Basin-and-Range Province is one of the best-monitored examples of an active, diffusely deforming intraplate region, and its location in a semi-arid climate grants excellent exposure and preservation of climatic and tectonic proxy records. Measurements with > 50 GPS stations (BARGEN, e.g. Bennett et al., 2003), which have been recording continuously since 1996, have confirmed that about 1 cm/y or about one third of the contemporary strain accumulation across the plate boundary occurs up to 1000 km from the San Andreas transform system. On a historic time scale only a small number of faults have experienced large ground-rupturing seismic events, whereas on intermediate time scales (10s to 100’s of ky) nearly all faults have been active. Some of the faults, such as the Wasatch fault, exhibit a spectral deformation character with order of magnitude variations in measurable fault slip parameters on annual, millennial, and million-year time scales. First order variations in fault slip rates are attributed to changes in tectonic boundary conditions and occur on the million-year time scale (Fig. 75, Table 1). Second order variations are likely related to fault system dynamics and occur on time scales of thousands of years. Third order variations provide key information on the time scale of the seismic cycle. For several other faults (e.g. Owens Valley fault, Crescent Valley fault, and Garlock fault; Peltzer et al., 2001; Friedrich et al., 2004), paleoseismic data in conjunction with space-geodetic data yield fluctuations in contemporary strain on time scales significantly shorter than the seismic cycle. On one hand, this behaviour implies that measurements of deformational parameters with just a single method over a short time window are insufficient to capture short-, intermediate- and long-term processes. On the other hand, this implies that it is possible to detect real variability in a process-related parameter if time-series are measured over sufficiently long intervals, and at three neighbouring frequencies. This approach will be adopted at the TOPO-EUROPE focus sites whenever possible.

Variations in fault interaction also occur on a range of spatial scales, particularly expressed in terms of alternating slip rates on neighbouring fault systems. For example, on a scale of several hundred kilometres the slip-rate histories of the San Andreas and San Jacinto faults in California have co-varied over the past 5 My (Bennett et al., 2004). On a scale of a few hundred kilometres, variation of fault activity also appears to co-vary, as has been observed in the Los Angeles Basin-Mojave Desert regions (e.g. Rockwell et al., 2000). On the scale of a single mountain front, segmentation into semi-independent seismotectonic blocks, however, also demonstrates that sustained faulting can occur over timescales of 104 to 105 years, apparently without influencing activity on adjacent segments (Wallace, 1987; Strecker et al., 2003).

Taken together, these observations clearly demonstrate that instrumentally and historically recorded deformation does not provide sufficient information to fully understand the long-term behaviour of linked fault systems (e.g. Friedrich et al., 2003; Friedrich et al., 2004). Consequently, records of short-term fault activity may not be a good proxy for long-term behaviour, and vice versa. This dilemma becomes even more complex when regions are assessed in which plate convergence is accommodated by the reactivation of pre-existing structures, faults with different kinematics or where faulting may be associated with the effects of rapidly changing boundary conditions.

Spatial variability in strain release on geological time scales has recently been described for the western portion of California, where the San Andreas fault probably jumped inland around 5 Ma (e.g. King et al., 2004). Consequently, the rate of activity changed on several previously existing fault systems. The precise initiation times and the magnitude of the rate change is subject of ongoing studies.

Another example of spatial deformation variability is related to the collision zone of the Arabian and the Eurasian plate, representing one of the largest regions of active intracontinental convergent deformation on Earth that extends from Turkey to Afghanistan (Fig. 76). To the west, shortening is mainly accommodated by westward escape of Anatolia (see Section 3.3. above). To the east, in Iran, convergence is mainly accommodated by shortening in the Zagros and Elborz mountains, the Aspheron-Sill, as well as a small component of SE-directed extrusion of the Central Iranian Plateau (Berberian and Yeats, 1999). Preliminary space geodetic measurements suggest NNE directed, distributed shortening across the Iranian Plateau at a rate of 22±2 mm/y (Vernant et al., 2004a). Shortening in the Zagros fold and thrust belt is 7±2 mm/y, 8±2 mm/y in the Elborz Mountains, and 7±2 mm/y along the Aspheron Sill (Vernant et al., 2004a). In the Alborz Mountains the geodetic measurements show that in addition to a shortening component there is a component of left-lateral motion at 4±2 mm yr-1 (Vernant et al., 2004b). In order to obtain a better understanding of the tectonic and topographic evolution of the Iranian collision zone, however, a larger number of space-geodetic surveys, remote sensing studies and field campaigns will be needed to examine the kinematics of active fault zones.