ABSTRACT

Johnstone, E.; Raymond, J.; Olsen, M.J., and Driscoll, N., 2016. Morphological expressions of coastal cliff erosion processes in San Diego County. In: Brock, J.C.; Gesch, D.B.; Parrish, C.E.; Rogers, J.N., and Wright, C.W. (eds.), Advances in Topobathymetric Mapping, Models, and Applications. Journal of Coastal Research, Special Issue, No. 76, pp. 174–184. Coconut Creek (Florida), ISSN 0749-0208.

High-resolution, Terrestrial Laser Scanning (TLS) data have been acquired seasonally since 2006 to define the style and magnitude of cliff erosion along the southern 20 km of coastline within the Oceanside Littoral Cell (OLC). In particular, twelve sites with cliff collapses were mapped repeatedly to examine how these collapses propagate along the cliffs and to identify feedback mechanisms between the liberated material and subsequent cliff failures. Grain size analyses of the failed material (retention cutoff) were performed to estimate the contribution to the beach sand inventory. Despite a relatively short time series (only six years) on a geologic scale, the high spatial and temporal resolution of the study has provided important insights into the fine details of processes controlling cliff erosion in the OLC. In addition, the seasonal TLS established a quantitative baseline from which future change may be assessed. Both lithological and environmental conditions are known to play a major role in governing the rate and style of cliff erosion; however, other factors such as beach width, elevation, and precipitation also exert control on rates and styles of cliff failures. The findings of this study reveal that cliff erosion is subaerially dominated where the beaches are wider and elevation is higher. Alternatively, erosion is marine dominated where the beaches are narrow and have lower average elevation. A direct relationship exists between beach elevation and undercutting and erosion along the failure edges and thus might provide a mechanism to create the observed linear retreat of the cliffs in the OLC rather than the formation of promontories and embayments. Other morphological expressions on the cliff face, such as honeycomb patterns and sawtooth-style frontage, indicate mechanisms that control predominant styles of erosion in particular locations. This time series documents seasonal and short-term erosional patterns and rates as well as establishes a baseline to understand cliff erosion in response to rapid sea level rise (>3 mm/yr).

INTRODUCTION

Beaches in San Diego County and many locales throughout the world are an important natural resource for a variety of socioeconomic reasons. For example, economic studies indicate that beach-related tourism and associated services contribute more than $200 million a year to the local economy (California Department of Boating and Waterways, 2002). Currently, concern is growing that this resource is at risk because the damming of local rivers (Inman and Jenkins, 1999), urbanization (Warrick and Rubin, 2007), and armoring of the bluffs (Haas, 2005; Runyan and Griggs, 2003) are reducing the natural sand supply to the beach. It is also important to understand how coastal erosion and sediment supply will be impacted by increasing sea level rise as well as changes from climate shifts, such as the Pacific Decadal Oscillation (PDO) and El Niño Southern Oscillation (ENSO).

Recent studies suggest that seacliff erosion currently contributes more to the beach sand supply than previously thought (Haas, 2005; Young and Ashford, 2006). For example, Raymond (2010) and Young et al. (2010b) determined that the OLC seacliffs are composed of >50% sand content. Sediment contributions from the erosion of seacliffs may become a larger percentage of the sediment source as the effects of damming (Slagel and Griggs, 2008; Willis and Griggs, 2003) and urbanization of the coastal watershed continue to increase (Warrick and Rubin, 2007).

Most researchers have quantified regional erosion rates for San Diego County using long-term averaging methods (e.g., Griggs and Patsch, 2004; Hapke et al., 2009). While this technique provides a valuable estimate for average erosion rates in a region and can be used for management decisions, it is often difficult to assess, visualize, and understand the processes governing cliff erosion, which occurs across multiple spatial and temporal scales. Due to the complex nature of the composition and morphology of the cliffs and forcing functions controlling the erosion, it is difficult to assign a single, regional erosion rate that accurately captures the local variability. Overall styles of erosion are highly dependent on localized factors such as geological composition, wave dynamics, and regional climate patterns.

An improved understanding of the relationship between bluff erosion and beach sand supply is necessary for proper management and enjoyment of this valuable resource. Unfortunately, obtaining reliable, quantitative information in dynamic coastal environments can be challenging due to high spatial and temporal variability of the processes and limitations in technologies that are able to acquire this information. Given the current >3 mm/year eustatic sea level rise (Rahmstorf, 2007) due to thermal expansion as well as accelerated glacial retreat, it is imperative to establish a baseline from which future erosion can be assessed to understand implications for coastal communities.

Quantitative mapping of cliff erosion on temporal and spatial scales similar to the operative processes provides insights into the mechanisms that control cliff erosion, complementing longer-term studies that yield average erosion rates. In particular, three-dimensional change modeling using lidar, when combined with grain size information, enables a determination of what percentage of the collapsed material is likely to remain on the beach given the average grain size population of sand on the beach and the wave climate (Runyan and Griggs, 2003; Young et al., 2010a). Such detailed assessments provide the link between processes and product.

To this end, the objective of this study is to document and illustrate processes observed with repeat three-dimensional change modeling using high-resolution TLS and sediment analyses. Various styles of seacliff erosion observed in the OLC (Figure 1), the controlling parameters, and potential contribution to the beach sand budget were examined. This study does not intend to quantify long-term erosion rates, but rather highlight processes and insights that may not be observed in data sources commonly used for long-term studies. To date, given the relatively short duration of this study and the episodic nature of cliff failures, it is difficult to make any long-term conclusions about contributions to the beach sand budget. Nevertheless, the quantitative baseline dataset for the seacliffs in the OLC with repeat lidar scans enables measuring of the quantities of materials associated with seacliff collapses due to different processes (e.g., wave undercutting versus subaerial processes) and how the collapses vary from natural environments to developed landscapes.

Figure 1.

Location map of lidar study sites in the Oceanside Littoral Cell, southern California, and select failure sites (orange circles) where repeat rapid response and seasonal surveys (red lines) were conducted.

Figure 1.

Location map of lidar study sites in the Oceanside Littoral Cell, southern California, and select failure sites (orange circles) where repeat rapid response and seasonal surveys (red lines) were conducted.

Study Area

The study area is within the Oceanside Littoral Cell (OLC) between La Jolla and Encinitas (Figure 1). The coastal sediments of San Diego County consist of several well-sorted, cemented marine terraces that have formed as a result of sea level rise and fall in combination with tectonic activity throughout during California's geologic history (Kennedy and Tan, 2005). The seacliffs are composed of Eocene marine sedimentary rocks mantled by Pleistocene terrace and alluvium deposits (Kennedy, 1973; Kennedy and Tan, 2005). The Eocene rocks belong to the La Jolla Group; the members found within the study area ordered from oldest to youngest are the Delmar formation (~49–45 Ma), Torrey Sandstone (~49–45 Ma), Ardath Shale (49–45 Ma), and the Scripps formation (47–44 Ma).

The area is characterized by a semiarid Mediterranean climate, with a winter rainy season from October to April and a warm dry summer from May to September. The climate is influenced by changes in ENSO and PDO events resulting in alternating decades of strong and weak El Niño. Strong El Niño events are associated with anomalously high precipitation during the winter rainy season, whereas La Niña events are associated with moderate to low rainfall.

While a variety of mechanisms contribute to seacliff erosion, the two dominant modes of cliff erosion in the OLC can be distinguished as marine and subaerial erosion, which may work in concert (Young et al., 2009b). Marine erosion occurs when the wave runup (Ruggerio et al., 2001) exceeds the beach elevation, enabling wave energy to make contact and erode the base of the cliffs. Sand and small rocks often mix with the water to accelerate the erosion. These processes erode the cliff from the bottom up. Subaerial erosion, in contrast, predominately occurs above the wave impact zone (Benumof et al., 2000) and generally works from the top downward. Faults and joints form zones of weakness, enhancing mechanical erosion and providing conduits for groundwater seepage, which further destabilize the seacliffs.

Benumof et al. (2000) suggest that lithology is a more important control on cliff erosion than the wave climate in California. Nevertheless, certain climatological conditions such as increased storminess also may cause higher rates of cliff erosion (Storlazzi and Griggs, 2000). Sea level is rising at a rate of 2–3 mm/per year (IPCC, 2007) in the study area, and this trend is predicted to continue, if not increase by an order of magnitude (Pfeffer et al., 2008; Rahmstorf, 2007). Therefore, it is important to understand how this will influence marine erosional processes and ultimately how the coastal cliffs and beaches will be impacted.

Human development, including extensive cliff top construction and armoring, plays a major role in the erosive nature of seacliffs. With heightened development from coastal properties and golf courses in close proximity to the coast, water runoff and groundwater seepage has increased along the cliffs. This accumulation and drainage of water simulates increased precipitation, creating zones where sapping, which used to be seasonal, now occurs year-round. In addition to cliff top alterations, approximately 30% of the seacliff faces within the OLC have been altered in some manner. Armoring of cliffs in the study region alters the natural process of sediment liberation from the cliff face and artificially limits local inputs of sand supply. This sediment supply reduction is reflected in the lower beach elevation profiles and width, which subsequently diminishes the ability of the beach to act as a natural barrier to wave energy. Within the study area, beach width and height tends to be greatest in areas without cliff armoring (e.g., Torrey Pines south to La Jolla Cove; Haas, 2005; Young et al., 2010a).

The coastal armoring has implications on the amount of sediment provided to the beach. Based on airborne lidar data acquired in San Diego County between 2001 and 2004, it was estimated that seacliffs and gullies yielded 76,900 m3/yr and 20,000 m3/yr of sediment, respectively (Young and Ashford, 2006). The average annual sediment flux estimated from the coastal rivers for the study period was 15,700 m3/yr. The volumetric rate of sediment yield per length of shoreline ranged from 0.47 (Carlsbad) to 3.61 (San Onofre) m3/m-yr with a weighted average of 1.80 m3/m-yr for the OLC.

Through sediment provenance (Haas, 2005) and airborne lidar data (Young and Ashford, 2006; Young et al., 2009a), 50% or more of the sand supplied to the beach is estimated to be sourced from the seacliffs. This estimate is in marked contrast to the established paradigm that rivers in the OLC supply the majority of sand to the beach system with the cliffs being a minor source (Brownlie and Taylor, 1981; Inman and Jenkins, 1999). Based on these data, linear rates of seacliff retreat ranged from 3.1 to 13.2 cm/yr with a weighted average for the littoral cell of 8.0 cm/yr. The highest retreat rates were observed in the Del Mar, Solana Beach, and San Onofre sections, all of which were greater than 10 cm/yr. Hapke et al. (2009) reported erosion rates of ~20 cm/yr on average for seacliffs in the OLC.

METHODS

This study involved continuous TLS surveys along the seacliffs in the OLC between La Jolla and Encinitas starting in 2006 to establish a quantitative topographic baseline from which future change can be assessed (Figure 1). In addition to these baseline scans, several sites of heightened activity have been repeatedly scanned as part of a rapid response program (Olsen et al., 2008).

Field surveys were conducted using a Maptek I-Site 4400 laser scanner. The scanner emits light pulses at a wavelength of 905 nanometers that radiate in 360 degrees around the origin and 80 degrees in the Z-axis (Figure 2). An X, Y, Z coordinate is calculated from each signal based on two-way time travel from the scan origin and the direction of the emitted pulse. The scanner also records signal intensity and color information (Red, Green, Blue values) for each data point in the cloud. Scan time varies based on the resolution and rotational swath of each survey. This study mainly consisted of 180-degree scans oriented orthogonal to the cliffs with a sampling increment of 0.104°. An interscan distance of 40–50 m was used to obtain adequate point cloud overlap and complete coverage along the coast based on the optimal scan protocols determined by Olsen et al. (2009) for beaches backed by high seacliffs and equipment limitations (Figure 3).

Figure 2.

Image of I-Site 4400 terrestrial scanner field setup with diagram of laser pulse function and head rotation. Scanner emits infrared wavelength laser pulses (905 nm) that radiate in a 360-degree rotational scan field with a vertical look angle of 80 degrees. The signal is returned and distance is calculated based on time of return. Data clouds are georeferenced based on the real-time kinematic (RTK) GPS survey origin of the receiver.

Figure 2.

Image of I-Site 4400 terrestrial scanner field setup with diagram of laser pulse function and head rotation. Scanner emits infrared wavelength laser pulses (905 nm) that radiate in a 360-degree rotational scan field with a vertical look angle of 80 degrees. The signal is returned and distance is calculated based on time of return. Data clouds are georeferenced based on the real-time kinematic (RTK) GPS survey origin of the receiver.

Figure 3.

Georeferenced point cloud data (approximately 50 million points) of the Torrey Pines study area. Semi-circles on the beach represent the scan origins with approximately 40-m spacing.

Figure 3.

Georeferenced point cloud data (approximately 50 million points) of the Torrey Pines study area. Semi-circles on the beach represent the scan origins with approximately 40-m spacing.

Scanning efficiency (survey coverage area/time) was taken into consideration during survey planning and acquisition. The I-Site 4400 laser scanner has several scanning parameters including high, medium, and low resolution. These settings enable users to find an appropriate balance to maximize scan effort to data ratio because of the labor-intensive nature of TLS, especially in dynamic coastal areas (Olsen et al., 2009). Calibration surveys were completed in order to determine the optimal scanning parameters. High-resolution scans were performed along sections with large failures or other features of interest. Successful data collection was highly dependent on working around weather and tidal conditions. Negative low tides were an environmental requirement to maximize survey distance from the cliffs in order to capture data from the cliff tops. The optimal distance from the cliff was dependent on cliff height based on a 40-degree laser angle from the scanner origin.

Scans were georeferenced to minimize uncertainty using the methodologies outlined in Olsen et al. (2009 and 2011) for seacliff erosion studies. This approach has an estimated 3D accuracy of +/− 7 cm with this equipment. The scanner position (X, Y, Z) was determined using observations from a Trimble Real-Time Kinematic GPS (RTK-GPS) receiver located above the scanner origin, corrected for offset. Control points (USACE survey markers) were occupied with the GPS receiver for calibration and quality control of the GPS data. The scanner was approximately leveled within a few degrees and level compensator readings on the scanner were used to level each scan. A “backsighting” technique was used during surveys to increase the efficiency of inter-scan alignment. The algorithms presented in Olsen et al. (2011) were then applied to minimize the error throughout the survey section.

Post-collection data processing and analysis was performed using Maptek I-Site Studio software. Surface modeling was completed using a combination of the rapid triangulation algorithm presented in Olsen et al. (2013) and modeling tools in Maptek I-Site Studio. To complement the TLS surveys, grain size analysis was performed on ten sediment samples from several formations, including Torrey Sandstone, Del Mar, and Bay Point alluvium in selected regions. Sand percentages were calculated using a Littoral Cutoff Diameter (LCD) limit of 125 μm (Limber et al., 2008; Patsch and Griggs, 2007) based on the grain size results of Haas (2005).

Observed environmental conditions were obtained for the study period to understand subaerial and marine factors (Figure 4), predominately precipitation and wave runup, respectively, that impact the rates and timing of cliff erosion. Daily-accumulated precipitation data was acquired from NOAA station 23188 (NOAA, 2012) to measure both frequency and intensity of rainfall relative to cliff failure events. Wave, height, period, and direction were obtained from the Coastal Data Information Program (CDIP buoy 093 – Mission Beach) in order to estimate the Total Water Level (TWL) based on the method developed by Ruggiero (1996; 2001).

Figure 4.

Failure event timeline overlain on environmental data during study period including (A) Precipitation levels (NOAA, 2006–2011), (B) Swell direction (CDIP offshore Buoy 093), (C) Significant wave height, and (D) Total Water Level (TWL).

Figure 4.

Failure event timeline overlain on environmental data during study period including (A) Precipitation levels (NOAA, 2006–2011), (B) Swell direction (CDIP offshore Buoy 093), (C) Significant wave height, and (D) Total Water Level (TWL).

RESULTS

A substantial database of TLS data has been collected throughout the study area. Given that the primary focus of this paper is to document operative processes of cliff erosion and reworking rather than to define regional erosional rates and trends, only detailed results for select sections and sites are presented.

Observed Collapses and Conditions

Twelve seacliff collapses (Table 1) in the study region (Figure 1) have been imaged routinely since 2006. Sites were prioritized since not all failures observed in the study area could be scanned on a regular basis because of limited access to the beach and tidal constraints. Thus, sites of low priority such as those with relatively small volumes of sediment (<30 m3) were not rescanned on a regular basis. Hence, the list presented in Table 1 is not a complete inventory of all collapses that occurred during this time but records collapses that were monitored and periodically rescanned during the course of the study. Additional detail and observations made at these sites are discussed in Olsen et al. (2016), which is in this special issue.

Table 1

Large failures occuring betwen 2007–2010 within the study region. Geologic Units: DM=Del Mar, TS=Torrey Sandstone, SF=Scripps Formation, AD=Ardath Shale

Large failures occuring betwen 2007–2010 within the study region. Geologic Units: DM=Del Mar, TS=Torrey Sandstone, SF=Scripps Formation, AD=Ardath Shale
Large failures occuring betwen 2007–2010 within the study region. Geologic Units: DM=Del Mar, TS=Torrey Sandstone, SF=Scripps Formation, AD=Ardath Shale

At each site, environment conditions varied, and the most probable cause of each failure is listed in Table 1. These classifications are based on determining whether wave runup, as described by Ruggerio et al. (2001), reached the base of the cliff face, if there was local precipitation in the region recently before the time of the failure (Figure 4), or evidence of groundwater sapping was observed at the site during the initial post-collapse survey. If a plausible triggering mechanism could not be identified, then a failure was deemed to have occurred from general instability.

Wave undercutting was more prevalent during periods with extreme wave-run-up from tidal conditions or large storm systems. Sites most impacted by wave action tended to have lower beach elevations in front of the cliff and talus debris piles were quickly reworked (within days to weeks of wave contact). An example is presented in Figures 5 and 6, illustrating the evolution of wave undercutting at the base of site TP2, where a wave cut notch grows until the upper material fails. In August 2007, semi-consolidated sediment in the Torrey Pines State Beach detaches from an oversteepened cliff and falls to the beach. The collapsed material is then reworked by waves. During the period of reworking, the base of the cliff remains protected so additional undercutting is not observed behind the material; however, adjacent areas on the cliff face continue to be undercut. Finally, once the material is largely removed, the process of undercutting continues and a much larger collapse occurs the following year in September 2008.

Figure 5.

Time-series evolution of cliff failure site TP2 imaged in three-dimensional lidar models. Lower bar of each image shows date of survey and calculated volume change since last survey. Erosion is depicted in orange shading and accretion in blue. Areas in gray have no detectable change. Note the continual reworking of failed cliff material between most collapses.

Figure 5.

Time-series evolution of cliff failure site TP2 imaged in three-dimensional lidar models. Lower bar of each image shows date of survey and calculated volume change since last survey. Erosion is depicted in orange shading and accretion in blue. Areas in gray have no detectable change. Note the continual reworking of failed cliff material between most collapses.

Figure 6.

Series of cross-section profiles of selected adjacent surveys (A–E) at Site TP2 illustrating the evolution of the cliff morphology. Subsequent surveys acquired post-collapse events imaged with zones of erosion shown in orange and accretion in blue. Note the large talus from September 2008 failure was eroded by winter wave action.

Figure 6.

Series of cross-section profiles of selected adjacent surveys (A–E) at Site TP2 illustrating the evolution of the cliff morphology. Subsequent surveys acquired post-collapse events imaged with zones of erosion shown in orange and accretion in blue. Note the large talus from September 2008 failure was eroded by winter wave action.

Subaerial failures within the study area were frequently initiated along aquatards where increased sapping was documented (Figure 7) and often occurred after large precipitation events. Groundwater failure sites tend to be zones where the lithologic bedding plane is saturated along free surfaces where concentrated fluid flows toward the coast. At sites experiencing subaerial erosion, talus deposits tended to build up and remain on the beach for a longer period of time. Figure 8 shows an example of talus deposit build up at sites TP6 and TP7. This section of the study was where the beach was inflated above the Mean High Water level and, therefore, not impacted by marine processes.

Figure 7.

(A) Lidar scan model of seacliffs backing Black's Beach with points colored in photographic color data (RGB) values. (B) Intensity return values draped on surface of same region. Note: Red colors indicate high signal return and blues indicate lower return signals. These wet areas of low intensity return (B) correspond well with areas of dark staining on the photograph (A).

Figure 7.

(A) Lidar scan model of seacliffs backing Black's Beach with points colored in photographic color data (RGB) values. (B) Intensity return values draped on surface of same region. Note: Red colors indicate high signal return and blues indicate lower return signals. These wet areas of low intensity return (B) correspond well with areas of dark staining on the photograph (A).

Figure 8.

Surface change DTM for Site A showing erosion and deposition patterns from November 2006 to May 2010. The boundary between the Delmar (claystone) and Torrey (sandstone) formations is delineated with a dashed line.

Figure 8.

Surface change DTM for Site A showing erosion and deposition patterns from November 2006 to May 2010. The boundary between the Delmar (claystone) and Torrey (sandstone) formations is delineated with a dashed line.

Erosion Volumes by Geologic Unit

Use of the TLS yielded quantitative estimates of cliff erosion and placed important constraints on the volume of sediment liberated by each geologic layer within the cliffs. Figure 9 presents semi-annual erosion volumes calculated for sites TP2, TP6, and TP7 separated by each predominant geological unit and their combined total (colored gray). For this analysis, sites TP6 and TP7 were grouped together given their close spatial proximity. At sites TP6 and TP7, the Delmar and Torrey Sandstone formations are the lower and upper units, respectively. Due to a downward dip of the cliffs toward the south, the Delmar formation is no longer visible at site TP2. Hence, the lower unit at site TP2 is the Torrey Pines formation and the upper unit consists of Baypoint Alluvium.

Figure 9.

Erosion volumes (m3) for lower geological unit (blue), upper geological unit (green), and total for both units (shaded gray) for semi-annual surveys at (A) Sites TP6 and TP7 and (B) Site TP2.

Figure 9.

Erosion volumes (m3) for lower geological unit (blue), upper geological unit (green), and total for both units (shaded gray) for semi-annual surveys at (A) Sites TP6 and TP7 and (B) Site TP2.

The total volume eroded from the cliff face at sites TP6 and TP7 was 2,169 m3. A fairly equal contribution was observed from both geological formations: 1,073 m3 from the Delmar and 1,096 m3 from the Torrey Sandstone. While the total volumes were fairly equal in total, examining the differences observed with the finer temporal resolution provided by the TLS surveys (Figure 9) shows that erosion volumes were nearly equal for both units in the May 2009 survey, the October 2009 and May 2010 surveys show significantly more erosion in the Torrey Sandstone formation and Del Mar formations, respectively. In contrast to sites TP2 and TP6, the total volume eroded from the cliff face at TP2 was 897 m3 with 625 m3 from the Delmar and Torrey, and 272 m3 from the Baypoint.

To relate these volumes to potential material retained on the beach, average cumulative frequency grain size distribution was determined for the formations in the study area (Figure 10) to determine the littoral sediment contribution. For sediment yields from each site, sediments above the cut-off diameter (125 m) are assumed to remain on the beach and finer-grained sediments are deposited offshore in lower energy environments.

Figure 10.

Average cumulative frequency grain size distribution plot for each geological formation in the study area. A littoral cutoff diameter (LCD) of 125 μm (3 Φ) was selected to compare the differences in the contribution of sand from each formation to the littoral sediment budget. The “littoral window” (shaded in gray) denotes the weight percentages of sand above the LCD observed for each geological formation.

Figure 10.

Average cumulative frequency grain size distribution plot for each geological formation in the study area. A littoral cutoff diameter (LCD) of 125 μm (3 Φ) was selected to compare the differences in the contribution of sand from each formation to the littoral sediment budget. The “littoral window” (shaded in gray) denotes the weight percentages of sand above the LCD observed for each geological formation.

Sediment Reworking of Talus Deposits

Figure 11 shows the cumulative cliff input and talus volume measured at Site TP2 from the first failure event on August 23, 2007, to the last survey on June 16, 2010, comprising 26 repeat scans (Figure 6). The slope of the talus volume curve is steepest after the largest failure, which indicates that reworking speed is a function of sediment availability because the talus material is also more exposed to more wave action and energy the farther it extends seaward. The difference between the two curves indicates the volume of material transferred from the cliffs to the beach. For instance, the last data point plotted for each curve shows that the total cliff input was 722 m3 and the talus volume that remains at the base of the cliff is 15 m3, suggesting that 707 m3 of talus material is reworked.

Figure 11.

Cumulative cliff input and talus volume at Site TP2 measured over the course of this study.

Figure 11.

Cumulative cliff input and talus volume at Site TP2 measured over the course of this study.

Geomorphologic Styles

In addition to quantifying the volumes of erosion, the high resolution TLS data enables documentation of the style and resulting cliff morphologies formed through the erosional processes. For example, the Torrey Sandstone formation typically erodes in a honeycomb fashion, resulting in a rough, hummocky cliff surface. When collapses occur, however, the exposed failure surface appears smoother. Rugosity, or surface roughness, appears to increase over time after the initial failure of the cliff surface at the detachment site through subaerial processes (e.g., wind, salt, and precipitation). Figure 12 captures the rugosity before and after collapses at sites TP6 and TP7. In this figure, it is observed that at both sites, the failed surface is smoother than the unfailed surface. Periodic scans visually examined following these events showed a gradual increase in roughness of the surface with time.

Figure 12.

(A) Topographic triangulation of northern Torrey Pines study region. Survey was conducted on 11/13/2008 and images the pre-failure sites TP6 and TP7. (B) Topographic triangulation of northern Torrey Pines study region. Survey was conducted on 05/13/2009 and images the post-failure sites TP6 and TP7.

Figure 12.

(A) Topographic triangulation of northern Torrey Pines study region. Survey was conducted on 11/13/2008 and images the pre-failure sites TP6 and TP7. (B) Topographic triangulation of northern Torrey Pines study region. Survey was conducted on 05/13/2009 and images the post-failure sites TP6 and TP7.

Both the width and elevation of the beach vary spatially and temporally across the study site. Figure 13 presents the results of an analysis of beach widths using airborne lidar surveys for the Black's Beach and Solana Beach sections of the study area. In general, wider beach widths and greater beach elevation are observed in locations with higher seacliffs. Figure 14 tracks beach elevations at two locations. Sites with higher beach elevations are predominantly eroded from subaerial processes, whereas sites with lower beach elevations are dominated by marine erosion and undercutting.

Figure 13.

Aerial lidar digital elevation model of Black's Beach (top) and Solana Beach (bottom) from 2009. Above 4 m is in grayscale and the beach (0–4-m elevation) is shown in color. The red line indicates the location of the 2D profile for above beach elevation and the black line indicates the location of the 2D profile for beach. Note the change in vertical and horizontal scale between beach and land profiles.

Figure 13.

Aerial lidar digital elevation model of Black's Beach (top) and Solana Beach (bottom) from 2009. Above 4 m is in grayscale and the beach (0–4-m elevation) is shown in color. The red line indicates the location of the 2D profile for above beach elevation and the black line indicates the location of the 2D profile for beach. Note the change in vertical and horizontal scale between beach and land profiles.

Figure 14.

Images of two failure sites (A, B) within the study area and graphical representation of beach elevations at each site through time with TWL data. Site A is the location of failures TP6 and TP7. Site B is located just south of the “Flat Rock” in Torrey Pines State Park where failure TP2 occurred.

Figure 14.

Images of two failure sites (A, B) within the study area and graphical representation of beach elevations at each site through time with TWL data. Site A is the location of failures TP6 and TP7. Site B is located just south of the “Flat Rock” in Torrey Pines State Park where failure TP2 occurred.

DISCUSSION

This five-year, repeat lidar study conducted in the OLC provided new insights into short-term processes that control seacliff erosion. The cliff morphologies observed are controlled by the lithological composition of sedimentary units and the physical forces acting upon them. Subaerial and marine-based erosion were both present along the entire study site; however, the degree to which each end-member played a role was highly dependent on the intrinsic character of the cliff face and local beach width and elevation.

Subaerial Erosion Processes

Weathering of sediments from above the waterline is a common form of erosion in southern California because the cliffs consist of weakly cemented, semi-consolidated sedimentary deposits (Kennedy, 1973). Erosion is activated by several processes including wind, precipitation, bioerosion, and mass-wasting. Many sites in the northern end of Torrey Pines where the cliffs are higher and the beaches are wider are dominated by subaerial erosion. As shown for site TP6 and TP7 in Figure 14, these sites tend to have a higher beach elevation at a given point in time. In particular, the cliffs composed of Torrey Sandstone tend to erode in a hummocky fashion as highlighted in Figure 12. Whereas, in the Del Mar formation, consisting predominately of claystone, rilling is more common, and erosion tends to be uniform and produce a lower vertical angle profile. Both styles produce significant stockpiles of talus deposits at the cliff base (Figure 8).

Failures initiated during periods of higher precipitation (>1 cm) occurred at sites TP4 and TP5 in October 2007 and January 2008, respectively. Storm events increased the groundwater flow and sapping was evident where the block-style cliff erosion occurred. TP4 liberated 168 cubic meters of sediments from the Torrey Sandstone formation. Failure TP6 and TP7 (Figure 8) occurred in close proximity to one another near the northern extent of the park. TP6 initially released 77 cubic meters of sandstone in June 2008 with an additional failure event placing 233 cubic meters on the beach. TP7 experienced a major failure during December 2008 with 678 cubic meters of sediment from the Torrey formation eroding off the cliff face. TP8 occurred on the northern side of a headland within Torrey Pines State Park. This failure released 141 cubic meters of sediments and occurred on the north side of the headland discussed in regard to TP2 (Figure 14).

Bioerosion was also observed at several sites. Failure sites TP6 and TP7 appear to be the result of bioersoion from root swelling and general instability (Figure 12). Both failures occurred near the cliff tops where root material is exposed on the cliff face. Although TP8 appears to be a result of general instability from oversteepening of the cliff, significant plant material (roots) exists just above the observed failure zone. At this site, most material was liberated from the upper part of the cliff in the Torrey formation (Figure 12) where a significant overhang was located prior to the failure event.

Groundwater sapping appears to be a significant factor causing erosion at several sites (Figure 7). For example, the SC1 failure occurred just below the Institute of Geophysics and Planetary Physics (IGPP) building where groundwater seepage is regularly observed in the Scripps formation. In addition, Site BB1 appears to be initiated by groundwater leeching from the cliffs. The sapping zone, with concentrated groundwater discharge, is an aquatard developed along the lower lens of relict channel fill conglomerate, which also represents the contact between the Scripps formation and the less permeable Ardath Shale. This stratigraphic low point likely focused the water transport causing the increased moisture to destabilize the cliff. At site DM1 significant groundwater sapping was also observed.

Profiles of the cliffs in sections that are controlled mainly by subaerial processes tend to exhibit a more relaxed angle as the sediments move toward the angle of repose. This angle is dependent on the grain size and structure, but intrinsic sediment strength characteristics may vary based on the level of cementation and lithology. Interestingly, the Torrey Sandstone formation tends to be very steep, whereas the Del Mar formation is generally less steep whether subaerial or wave-based erosion dominates.

Marine Erosion Processes

Oceanographic conditions, such as swell direction and amplitude, play an important role in the stability of seacliffs (Sunamura, 1977); unfortunately, the ability to model or quantify the direct impact of these conditions is limited because of lack of current in situ data. Wave information from offshore buoys has been used to model wave runup conditions, but due to the complexity of the bathymetry associated with the La Jolla and Scripps canyons in the coastal zone of southern California, these models do not accurately capture actual wave runup along the beach and cliffs (Raymond, 2010). Offshore topography and coastal headlands refract wave energy (Guza and Thornton, 1981; Inman and Frautschy, 1966), complicating the resultant conditions reaching and acting on the coast.

Furthermore, the beach elevation in southern California exhibits seasonal inflation and deflation, both based on the precursor wave climate and where offshore sand bodies and hardgrounds are located (Yates et al., 2009). During boreal winter, Northern Hemisphere swells dominate, keeping the sand bars in deeper waters. During the summer, the Southern Hemisphere swells transport sand to shallower depths attaching a sand bar to land, inflating the regional beach elevations. This landward shift of additional sand onshore provides increased protection for the cliffs from marine erosion. Moreover, the summer wave energy is lower and thus is typically less erosive than winter swells (Inman and Jenkins, 1999). If the actual wave runup exceeds the elevation at the cliff-beach interface, the base of the cliff is exposed to the hydraulic energy of the waves and then the sediment resistance to erosion controls the rate of basal notching (Figure 5).

Sites at the southern end of the Torrey Pines and Solana Beach sections resulted predominantly from wave-based erosion. These sites have a narrower beach, lower cliff height and lower beach elevation (Figure 13). Site TP2, in particular, experienced numerous retrogressive failures (Figures 5 and 6) that were originally caused by wave undercutting at the base of the cliff. After the original bluff collapsed, the cliff face angle was oversteepened, and most likely retrogressive instability caused the following series of failures observed. Figures 5 and 6 shows the evolution of the nucleation point elevation migrating upward through time. The pattern of beach erosion during winter 2008–2009 confirms the talus from this failure did not remain on the beach long, but rather was reworked within weeks by wave energy. Moreover, this site exists on part of a south facing natural headland created where relict, highly cemented lagoonal deposits from the Del Mar formation intersect the beach because of the southward dip of the formation. These lithified hard grounds are resistant to erosion and cause the cliffs to step seaward with increased exposure to waves.

Combined Erosion Processes

Two failures in Encinitas (EN1 & EN2) nucleated within the Delmar formation and were initiated by wave undercutting. These sections were exhibiting heavy groundwater sapping at the time of collapse. The seacliffs in this region of Encinitas are topped with high population density and development. Low intensity returns in lidar data and visual observation suggest that groundwater seepage is occurring regularly at numerous expulsion sites along the cliff face.

Indeterminate Processes

A direct cause for a few of the sites could not be determined based on environmental parameters. For example, general instability from oversteepened cliff faces appears to have initiated the failures documented at Del Mar. Failure TP1 also occurred where oversteepened slopes were present; however, no direct triggers were identified for this failure. The failure occurred during a dry period (August) when the beach berm was built to a high elevation preventing water from reaching the base of the cliff. More details of this particular site are discussed in Olsen et al. (2008).

Anthropogenic Influences

Lower, narrower beaches, especially those with exposed hardgrounds tend to be backed by wave-cut seacliffs (Figure 13). Notably, these regions are also zones where development and human population density are higher. Although it is difficult to assess directly how anthropogenic factors affect erosion rates, increased anthropogenic alteration of the seacliffs correlates with regions of narrow and low elevation beaches (Figure 13) throughout the study area. Seawalls are more prevalent along these stretches of coastline that are armored against cliff retreat, which cuts off sand supply to the beaches (Haas, 2005; Young and Ashford, 2006). The lack of sediment to the system intensifies deflation of the beach elevation and basement rock is exposed. There was more wave-based cliff erosion in the areas with lower beaches, instigating more artificially armored cliff faces, which in turn leads to less mobile sand on the beach to buffer the cliff. Increased levels of human development also lead to higher amounts of irrigation and non-native runoff, causing year-round groundwater sapping along aquatards that may only have been seasonally active in undisturbed settings. The increased water supply in these highly developed sections has led to artificial alteration in the character of the cliff face where drainage persists all year. Groundwater sapping destabilizes sections of the seacliff causing mass wasting.

CONCLUSIONS

The OLC is dominated by two modes of cliff erosion: marine and subaerial weathering. These two processes are observed to play an important role in controlling how the seacliffs contribute sediments to the OLC, whereas the composition of the cliffs (grain size) determines whether the failed material will remain on the beach. This research, along with current findings by others (Haas, 2005; Young and Ashford, 2006), illustrates the significance of these geological features for coastal resource management. The establishment of a baseline dataset in the region provides insight for quantifying changes based on longer-term geomorphological impacts from climate events (e.g., El Niño and/or sea level rise).

Subaerial weathering appears to be the predominant style of erosion in areas where beaches have high elevation and are wide; this weathering pattern tends to dominate in the southern portion of the study area. Results are consistent with previous research on grain size (Haas, 2005) analysis, revealing that the cliffs are contributing a significant amount of sand to the beaches (>50%), particularly in the southern study area where the highest seacliffs occur (Figure 13). The sandstone-rich formations that make up the upper portion of much of the OLC's composite seacliffs provide an abundant source of sand to the local beaches, and this, in turn, creates an elevated beach platform. The increase in topological structure acts as a temporary physical barrier against wave erosion until erosion of the beach platform occurs. Interestingly, much of the subaerial erosion occurred during periods of little to no precipitation, indicting another process forcing the instability of the seacliffs. Salt crystallization from exposure to sea spray may cause pressure alterations in the pore spaces and instigate subaerial erosion on the seacliffs. Further investigations of microstructural weathering are required to elucidate more results on this topic.

Marine erosion was observed to be the predominant style of erosion in locations with lower, narrower beaches, which tend to be backed by smaller seacliffs. Wave undercutting and notching were observed at these locations. Many of these areas have been anthropogenically armored, leading to reduced sediment supply (Young and Ashford, 2006), intensifying deflation of the beach elevation through wave action and erosion. During winter months, basement rock is often exposed in these locations. As a result, the amount of wave energy reaching the base of the cliffs increases, which subsequently increases marine-based cliff erosion.

High-resolution terrestrial lidar is an effective tool to quantify volumes of sediment eroded from seacliffs. These techniques also provide valuable information to understand the style and patterns of seacliff erosion along a geologically complex and variable coastline. While a significant amount of information is gleaned from using TLS to evaluate short-term processes, longer-term complementary studies are needed to measure change through climatological oscillations to gain insight on how southern California seacliffs will be impacted by sea level rise, ENSO, and PDO cycles.

Sediments liberated from the OLC seacliffs are an important source for beach sand supply, especially considering the modern rate of sea level rise (>3 mm/yr; IPCC, 2007; Pfeffer et al., 2008; Rahmstorf, 2007) and increased erosion along southern California's coastline. It is imperative to consider the significant contribution of a natural source of sand when making decisions regarding coastal management, especially as southern California enters a drier than average climatic cycle (PDO) and sediment input from fluvial systems is likely to be further reduced. At the same time, our understanding of the beach sand budget and the relative importance of the different sources are evolving (Haas, 2005). Longer time-series data are required to adequately document impacts from climate change and population increase (coastal development) in the OLC. This research provides valuable information regarding the processes that control the different styles of cliff erosion in southern California and a baseline from which future change can be evaluated.

ACKNOWLEDGMENTS

This research was funded by the California Seagrant number R/OE-39 and the Coastal Environmental Quality Initiative (CEQI) under grant number 04-T-CEQI-06-0046. Scott Schiele and John Dolan (Maptek I-site) provided technical assistance and Travis Thompson (CALVRS) provided assistance with the CALVRS network. Additional support was provided by Falko Kuester (UCSD) and Scott Ashford (Oregon State University).

LITERATURE CITED

LITERATURE CITED
Benumof
,
B.T.
;
Storlazzi
,
C.D.
;
Seymour
,
R.J.
, and
Griggs
,
G.B.
,
2000
.
The relationship between incident wave energy and seacliff erosion rates: San Diego County, California
.
Journal of Coastal Research
,
16
(
4
),
1162
1178
.
Brownlie
,
W.R.
and
Taylor
,
B.D.
,
1981
.
Sediment management for southern California mountains, coastal plains and shoreline; Part C, Coastal sediment delivery by major rivers in southern California: Pasadena
.
California Institute of Technology, Environmental Quality Laboratory Report 17-C
,
314
p
.
California Department of Boating and Waterways
,
2002
.
California Beach Restoration Study
.
Sacramento, California
.
Griggs
,
G.B.
and
Patsch
,
K.
,
2004
.
Cliff erosion and bluff retreat along the California coast - Coast of California caught between increasing numbers of people and ongoing process of shoreline retreat
.
Sea Technology
,
45
(
9
),
36
40
.
Guza
,
R.T.
and
Thornton
,
E.B.
,
1981
.
Wave set-up on a natural beach
.
Journal of Geophysical Research-Oc Atm
,
86
(
5
),
4133
4137
.
Hapke
,
C.J.
;
Reid
,
D.
, and
Richmond
,
B.
,
2009
.
Rates and trends of coastal change in California and the regional behavior of the beach and cliff system
.
Journal of Coastal Research
,
25
(
3
),
603
615
.
Haas
,
J.K.
,
2005
.
Grain Size and Mineralogical Characteristics of Beach Sand in the Oceanside Littoral Cell: Implications of Sediment Provenance
.
San Diego, California: University of California, San Diego, Master's thesis
,
41
p
.
Inman
,
D.L.
and
Frautschy
,
J.D.
,
1966
.
Littoral processes and the development of shorelines
.
Proceedings of the Coastal Engineering Specialty Conference
,
American Society of Civil Engineers
,
pp
.
511
536
.
Inman
,
D.L.
and
Jenkins
,
S.A.
,
1999
.
Climate change and the episodicity of sediment flux of small California rivers
.
Journal of Geology
,
107
(
3
),
251
270
.
IPCC
,
2007
.
Climate change 2007: The physical science basis
.
Cambridge
:
Cambridge University Press
,
996
p
.
Kennedy
,
M.P.
,
1973
.
Bedrock Lithologies, San Diego Coastal Area, California
.
In
:
Ross
,
A.
and
Dowlin
,
R.J.
(
eds.
),
Studies on the Geology and Geologic Hazards of the Greater San Diego Area, California
.
San Diego, California
:
San Diego Association of Geologists
,
pp
.
9
15
.
Kennedy
,
M.P.
and
Tan
,
S.S.
,
2005
.
Geologic Map of the San Diego 30′ × 60′ Quadrangle, California
.
California Department of Conservation, 1 plate
.
Limber
,
P.W.
;
Patsch
,
K.B.
, and
Griggs
,
G.B.
,
2008
.
Coastal sediment budgets and the littoral cutoff diameter: a grain size threshold for quantifying active sediment inputs
.
Journal of Coastal Research
,
24
(
2A
),
122
133
.
National Oceanographic and Atmospheric Administration, NOAA
,
2012
.
National Climatic Data Center (NCDC)
. .
Olsen
,
M.J.
;
Johnstone
E.
;
Ashford
S.A.
;
Driscoll
N.
;
Young
A.P.
;
Hsieh
T.J.
, and
Kuester
F.
,
2008
.
Rapid response to seacliff erosion in San Diego County, California using terrestrial Lidar
.
Proceedings of Solutions to Coastal Disasters
(
Oahu, Hawaii, ASCE
),
pp
.
573
583
.
Olsen
,
M.J.
;
Johnstone
,
E.
;
Driscoll
,
N.
;
Ashford
,
S.A.
, and
Kuester
,
F.
,
2009
.
Terrestrial laser scanning of extended cliff sections in dynamic environments: Parameter analysis
.
Journal of Surveying Engineering
,
135
(
4
),
161
169
.
Olsen
,
M.J.
;
Johnstone
,
E.
;
Kuester
,
F.
;
Driscoll
,
N.
, and
Ashford
,
S.A.
,
2011
.
New automated point-cloud alignment for ground-based light detection and ranging data of long coastal sections
.
Journal of Surveying Engineering
,
137
(
1
),
14
25
.
Olsen
,
M.J.
;
Johnstone
,
E.
, and
Kuester
,
F.
,
2013
.
Hinged, pseudo-grid triangulation method for long, near linear cliff analysis
.
Journal of Surveying Engineering
,
139
(
2
),
105
109
.
Olsen
,
M.J.
;
Johnstone
,
E.
;
Driscoll
,
N.
;
Kuester
,
F.
, and
Ashford
,
S.A.
,
2016
.
Fate and transport of seacliff failure sediment in southern California
.
In
:
Brock
,
J.C.
,
Parrish
,
C.E.
, and
Wright
,
W.
(
eds.
),
Advances in Topobathymetric Mapping, Models, and Applications. Journal of Coastal Research
,
Special Issue, No. 76
,
pp
.
185
199
.
Patsch
,
K.B.
and
Griggs
,
G.B.
,
2007
.
Development of sand budgets for California's major littoral cells: Eureka, Santa Cruz, Southern Monterey Bay, Santa Barbara, Santa Monica (including Zuma), San Pedro, Laguna, Oceanside, Mission Bay, and Silver Strand littoral cells
.
California Coastal Sediment Management Work Group Report, California Department of Boating and Waterways
,
266
p
.
Pfeffer
,
W.T.
;
Harper
,
J.T.
, and
O'Neel
,
S.
,
2008
.
Kinematic constraints on glacier contributions to 21st-century sea-level rise
.
Science
,
321
(
5894
),
1340
1343
.
Rahmstorf
,
S.
,
2007
.
Response to comments on “A semi-empirical approach to projecting future sea-level rise.”
Science
,
317
(5846
).
Raymond
,
J.H.
,
2010
.
New Constraints on the Processes that Control Cliff Erosion and Sediment Dispersal Using Ground-Based Lidar
.
San Diego, California: University of California, San Diego, Master's thesis
,
73
p
.
Ruggiero
,
P.
;
Komar
,
P.D.
;
McDougal
,
W.G.
, and
Beach
R.A.
,
1996
.
Extreme water levels, wave run-up, and coastal erosion
.
Proceedings of the 25th Coastal Engineering Conference
(
Orlando, Florida, ASCE
),
pp
.
2793
2805
.
Ruggiero
,
P.
;
Komar
,
P.D.
;
McDougal
,
W.G.
;
Marra
,
J.J.
, and
Beach
,
R.A.
,
2001
.
Wave run-up, extreme water levels and the erosion of properties backing beaches
.
Journal of Coastal Research
,
17
(
2
),
407
419
.
Runyan
,
K.
and
Griggs
,
G.B.
,
2003
.
The effects of armoring seacliffs on the natural sand supply to the beaches of California
.
Journal of Coastal Research
,
19
(
2
),
336
347
.
CDIP offshore Buoy 093, SCRIPPS Institution of Oceanography
,
2015
.
The Coastal Data Information Program (CDIP)
.
Integrative Oceanography Division, Scripps Institution of Oceanography
,
San Diego
. .
Slagel
,
M.J.
and
Griggs
,
G.B
,
2008
.
Cumulative losses of sand to the California coast by dam impoundment
.
Journal of Coastal Research
,
24
(
3
),
571
584
.
Storlazzi
,
C.D.
and
Griggs
,
G.B.
,
2000
.
Influence of El Nino-Southern Oscillation (ENSO) events on the evolution of central California's shoreline
.
Geological Society of America Bulletin
,
112
(
2
),
236
249
.
Sunamura
,
T.
,
1977
.
Relationship between wave-induced cliff erosion and erosive force of waves
.
Journal of Geology
,
85
(
5
),
613
618
.
Warrick
,
J.A.
and
Rubin
,
D.M.
,
2007
.
Suspended-sediment rating curve response to urbanization and wildfire, Santa Ana River, California
.
Journal of Geophysical Research -Earth
,
112
(F2
).
Willis
,
C.M.
and
Griggs
,
G.B.
,
2003
,
Reductions in fluvial sediment discharge by coastal dams in California and implications for beach sustainability
.
Journal Geology
,
111
(
2
),
167
182
.
Yates
,
M.L.
;
Guza
,
R.T.
;
O'Reilly
,
W.C.
, and
Seymour
,
R.J.
,
2009
.
Overveiw of seasonal sand level changes on southern California beaches
.
Shore and Beach
,
77
(
1
),
39
46
Young
,
A.P.
and
Ashford
,
S.A.
,
2006
.
Application of airborne lidar for seacliff volumetric change and beach-sediment budget contributions
.
Journal of Coastal Research
,
22
(
2
),
307
318
.
Young
,
A.P.
;
Flick
,
R.E.
;
Gutierrez
,
R.
, and
Guza
,
R.T.
,
2009a
.
Comparison of short-term seacliff retreat measurement methods in Del Mar, California
.
Geomorphology
,
112
(
3–4
),
318
323
.
Young
,
A.P.
;
Guza
R.T.
;
Flick
R.E.
;
O'Reilly
W.C.
, and
Gutierrez
,
R.
,
2009b
.
Rain, waves, and short-term evolution of composite seacliffs in southern California
.
Marine Geology
,
267
,
1
7
.
10.1016/j.margeo.2009.08.008
.
Young
,
A.P.
;
Raymond
,
J.H.
;
Sorenson
,
J.
;
Johnstone
,
E.A.
;
Driscoll
,
N.W.
;
Flick
,
R.E.
, and
Guza
,
R.T.
,
2010a
.
Coarse sediment yields from seacliff erosion in the oceanside littoral cell
.
Journal of Coastal Research
,
26
(
3
),
580
585
.
Young
,
A.P.
;
Olsen
,
M.J.
;
Driscoll
,
N.
;
Flick
,
R.E.
;
Gutierrez
,
R.
;
Guza
,
R.T.
;
Johnstone
,
E.
, and
Kuester
F.
,
2010b
.
Comparison of airborne and terrestrial lidar estimates of seacliff erosion in Southern California
.
Photogrammetric Engineering and Remote Sensing
,
76
(
4
),
421
427
.