The role of helicopter sluicing in earthquake response and recovery in Aotearoa New Zealand
Long-term rockfall observations post-Kaikōura earthquake
Abstract
Helicopter sluicing is an uncommon, and relatively novel active landslide mitigation approach. Following the 2016 Kaikōura earthquake, the resulting helicopter sluicing campaign was the largest use of this technique to date in New Zealand, and possibly globally. The aim was to reduce immediate rockfall hazard and clear slopes back to bedrock, where possible. Yet the legacy effect of sluicing in reducing post-earthquake rockfall activity remains unstudied. Here, we provide the first quantitative assessment of the role of helicopter sluicing on residual rockfall hazard following the Kaikōura earthquake using a long-term rockfall dataset from NZTA. Comparing rockfall rates from sluiced and non-sluiced landslides showed that north of Kaikōura, sluiced slopes produced 51% of all rockfall events and 87% of all rockfall volume, despite making up just 20% of the landslides along this stretch of SH1. South of Kaikōura, sluiced slopes account for 40% of the road exposure and produced 59% of all events but only 26% of the volume. The results suggest sluicing does not seem to meaningfully reduce long-term rockfall hazard. Rather, sluicing appears to be an effective remediation approach in a short-term operational setting only. Following future earthquakes, sluicing may be an important tool in the activate mitigation of landslides, but only if used in a targeted and focused way. However, sluicing should not be seen as a mitigation technique that can reduce the chronic post-earthquake rockfall hazard to a meaningful extent.
1. Introduction
Co-seismic landslides cause extensive social and economic damage, as witnessed in recent events including the 2008 Wenchuan, 2010-11 Canterbury, 2015 Gorkha and 2016 Kaikōura earthquakes (Kincey et al., 2021). Immediately post-event, the damage caused by these slope failures to critical infrastructure hinders the emergency response and increases the impact of the event on affected communities by blocking and damaging critical transport links (Budimir et al., 2014; Dahlquist & West, 2019). Importantly, increased rockfall activity associated with post-event landsliding can long outlast the initial surface rupture and ground motion hazards, resulting in a chronically elevated hazard and risk that may persist for years after the earthquake (Marc et al., 2015; Kincey et al., 2021). Understanding how the post-seismic elevated rockfall hazard might be effectively reduced is critical to post-earthquake recovery planning.
One approach may be by applying active mitigation measures, such as scaling, to landslide source areas, removing loose debris and allowing landscapes to recover faster, thus reducing the timeframe of elevated post-event hazard. The 2016 Kaikōura earthquake provides a case-study example of active mitigation being used immediately post-earthquake to clear co-seismic landslides impacting critical infrastructure (Green & Finlan, 2019; Mason & Brabhaharan, 2021). The earthquake triggered landslides that damaged 200 km of New Zealand’s road and rail network. Crews were rapidly brought in to begin mitigation at these sites, using a combination of active techniques including helicopter sluicing, a form of hydro-scaling (Mason & Brabhaharan, 2021).
Helicopter sluicing is a novel and uncommon approach to landslide mitigation globally but was used extensively post-Kaikōura earthquake. More than 220 million litres of seawater were dropped on 42 coastal landslides over the length of the transport corridor recovery area (Green & Finlan, 2019). Time, rather than cost, was the driving factor that led to the selection of helicopter sluicing as the primary means of active landslide mitigation. The slopes were identified as too steep and inaccessible for conventional techniques. Overall helicopter sluicing was credited by the North Canterbury Transport Infrastructure Recovery Alliance (NCTIR) as being critical to the swift recovery of the road and rail assets.
Despite its growing use in New Zealand, helicopter sluicing is rarely used elsewhere, and little has been published on the approach and its effectiveness. In particular, there are no studies on how removing the loose material on earthquake damaged slopes through heli-sluicing influences post-earthquake rockfall hazard. A detailed analysis of the long-term effects of sluicing on earthquake damaged slopes is therefore useful for informing whether helicopter sluicing is considered as an active mitigation technique in New Zealand’s next major earthquake.
2. 2016 Kaikōura earthquake
2.1 Landslide impacts
The Mw7.8 Kaikōura earthquake triggered up to 30,000 landslides over approximately 10,000 km2 of North Canterbury, including rock falls, rock and debris avalanches, shallow debris flows, and rotational slides (Massey et al., 2018; Massey et al., 2020). Slope failures ranged from large scale debris avalanches through to shallow upper slope failures and widespread ridge cracking (Justice et al., 2018). Approximately 60% of the landslides occurred in the highly fractured Torlesse Greywacke and tended to be shallow translational slides (Massey et al., 2018).
Co-seismic landslides significantly affected the transport networks, particularly State Highway 1 (SH1) and the Main North Line railway (MNL) along the Kaikōura coastline, resulting in the isolation of communities (Robinson et al., 2018). Damage occurred at more than 1500 sites in total, with more than 200 sites between the Clarence River and Cheviot (Mason & Brabhaharan, 2021). Almost one million cubic metres of landslide debris inundated the coastal transport corridor and had to be removed to allow traffic and trains back along the corridor (NCTIR, 2021).
The earthquake damaged the rock mass above the transport corridor, reducing its strength and increasing its susceptibility to further instability during aftershocks and rainfall events. Post-earthquake, three notable heavy rainfall events remobilised existing landslide features and trigged new instabilities: Cyclones Debbie and Cook in April 2017, and Cyclone Gita in February 2018 (GeoNet, 2017). These cyclone events delayed the recovery efforts by remobilising material that re-impacted the transport network, highlighting the ongoing threat to infrastructure and users.
2.2 Landslide Remediation: Heli-sluicing
In response, the NCTIR alliance was established as a partnership between NZTA Waka Kotahi and KiwiRail, along with Fulton Hogan, Higgins, Downer and HEB Construction. NCTIR was tasked to reopen the coastal transport corridor and the inland road between Kaikōura and Waiau. Importantly, one of the key aims of the programme was to increase the resilience of the transport infrastructure to future events, including ongoing rockfall.
Helicopter sluicing was quickly identified as a technique that could be used on the steep and often inaccessible slopes. As a form of scaling, the aim was to reduce the rockfall hazard by removing loose debris that could remobilise in future to allow earthworks crews to work safely beneath slopes (Figure 1). South of Kaikōura, sluicing was used on 29 landslides that damaged the road and rail infrastructure, compared to 13 major landslides to the north.
The sluicing process involved the helicopter pilots filling their monsoon buckets with seawater and releasing the water onto the slope, directed by site engineers and geologists (Figure 2). This water was targeted on slopes in a few ways, including washing loose material down a slope through exerting a force with the water, or trying to saturate a rock mass by filling discontinuities (Avery & Barrett, 2021; Green, 2024). Sluicing tended to focus on the headscarp areas, initially aiming to dislodge loose material and increase the safety for the rope-access teams by reducing the likelihood of future rockfall. At its peak in mid-2017, up to 13 helicopters were actively sluicing, with up to seven targeting a single location at a time. In total, over 220,000 litres of seawater were dropped onto the coastal landslides over a period of 12 months (Avery & Barrett, 2021; Green, 2024).
Sluicing south of Kaikōura was used at 29 sites along SH1. Public road access was regained by 21 December 2016, five weeks after the earthquake, however, additional landslides came down in subsequent cyclone events which caused temporary road closures and renewed the need for sluicing (Green, 2024). Landslide damage north of Kaikōura was significantly worse than that in the south, due to the much larger landslide volumes. Sluicing was used extensively at seven sites (Green, 2024). More than 90% of the recorded helicopter hours were spent on this northern section and it was not until mid-2017 that the full stretch of road became accessible again for construction work, some five months after the earthquake (Green & Finlan, 2019).
3. Methods
To evaluate the effect of heli-sluicing on longer term rockfall behaviour, we analysed a spatio-temporal record of all rockfall events recorded along the coastal transport corridor between 10 January 2017 and 3 March 2023. This dataset is owned by NZTA Waka Kotahi and managed by WSP New Zealand Ltd, and consists of a tabular dataset documenting the attributes of post-earthquake rockfall events that originated along the coastal slopes adjacent to the Transport Corridors. With data spanning almost 6 years post-event, this provides one of the most detailed and complete records of post-earthquake rockfall globally.
The dataset comprises a georeferenced tabular record with over 1150 slope failure entries with 55 attribute fields equating to over 63,000 individual data points. We narrowed these 55 attributes down to five key attributes that describe the location, style and physical dimensions of the failure event.
Importantly, the dataset does not identify whether rockfall originated from a sluiced landslide or not. To determine this, the provided geospatial coordinates for each rockfall were compared to available landslide maps and helicopter flight logs, with those coordinates within or immediately downslope of sluiced landslides considered to have originated from a sluiced landslide. All other rockfall records were assumed to have occurred from non-sluiced source zones. From these categories total rockfall event count and cumulative daily volume statistics were calculated.
4. Results
4.1 Heli-sluicing usage
In total, there were approximately 6700 hours of heli-sluicing undertaken, with >90% focussed on the slope failures north of Kaikōura. Two landslides, NRP6 (Ohau Point) and NRP7 (Paparoa South), account for 75% of the total heli-sluicing hours undertaken.
Initially, heli-sluicing was first used along the Southern Transport Corridor (Oaro-Kaikōura), where of the 5.2 km stretch of SH1 between Raramai Twin Tunnels and Peketa, 48% was partially or fully inundated by landslide or rockfall debris. Earthworks teams were exposed to continuous rockfall from the damaged slopes, with sluicing being employed to remove the most unstable material. After Christmas 2016, sluicing efforts were redirected to the Northern Transport Corridor (Kaikōura-Waipapa Bay), where the damage was more extensive. In the north, the plan area of sluiced landslides totals approximately 350,000 m2, across a transport corridor that spans 14.8 km. By mid-May 2017, most of the full sluicing days were complete, with occasional sluicing continuing until the end of 2017 when the road was reopened under limited serviceability.
4.2 Long-term rockfall rates
Overall, between 10 July 2017 and 3 April 2023, there were 883 rockfall events recorded, with sluiced landslides producing more rockfall (493 events) than non-sluiced landslides (390 events). The total rate of rockfall decreased over time, with an average rate from all landslides of 1 rockfall per 0.8 days between July 2017 and July 2018, decreasing to 1 rockfall per 9 days from July 2021 onwards (Figure 3). Importantly, sluiced landslides appear to initially produce rockfall at a higher rate than non-sluiced landslides, with a rockfall from a sluiced landslide occurring every 1.5 days prior to July 2018, compared to every 1.9 days from non-sluiced landslides (Figure 3). By July 2021 both sluiced and non-sluiced landslides were producing rockfall at similar rates (1 per 18 days).
In terms of rockfall volume, a total of ~7275 m3 of material was recorded, coming in roughly equal distributions from sluiced (~3618 m3) and non-sluiced landslides (~3657 m3). However, the nature of the accumulations are different. Of the total 3675 m3 to fail from non-sluiced landslides, ~2873 m3 failed within just one 24-hour period, accounting for 78% of all recorded rockfall (Figure 4). Sluiced landslides produced a similar total volume but with 70% of that coming from five separate days. In total, 90% of the volume of rockfall from sluiced landslides and 95% from non-sluiced landslides had failed by December 2018, just 2 years after the Kaikōura earthquake.
While the rockfall rates and volume from sluiced and non-sluiced landslides appear similar across the entire network, key differences appear when considering the sections north and south of Kaikōura independently. Firstly, to the north, sluiced landslides make up just 20% of the total failure area above the transport network, yet contribute 51% of the rockfall events, and 87% of the total rockfall volume since 2017 (Figure 5). Comparatively, south of Kaikōura sluiced landslides account for 40% of the failure area above the transport corridor, but 59% of the total rockfall events. However, in terms of rockfall volume, the majority here came from non-sluiced sources, which account for 74% of the total rockfall volume recorded along the southern corridor (Figure 5).
4.3 Ohau Point and Paparoa South
NRP6 (Ohau Point) and NRP7 (Paparoa South) occurred in close proximity and were the largest slope failures by plan area across the transport network. These two slope failures produced rock avalanches from near continuous bluffs, 180 m above SH1 (Hodgkinson et al., 2017). NRP6 was sluiced for ~2,600 hours across its plan area of almost 90,000 m2. NRP7 was larger at ~105,000 m2 and was sluiced for ~2,400 hours. Both failed within typical lithology for the coastal cliffs north of Kaikōura with similar failure styles and volumes.
Ohau Point was identified as one of the most challenging sites to remediate due to its scale, the residual hazard the slopes posed, and the physically constrained corridor that is situated on a narrow bench between the toe of the high cliffs and the sea. Persistent tension cracks were evident at the ridge crest and headscarp, presenting a possibility for regression of the source zones. Rockfall material was entrained up on the slope and remained in precarious locations, especially along the southern face. Large boulders were also exposed on the cliff face and were a key target of remediation.
Similarly, for Paparoa South, residual hazards included a large area of displaced rock mass (2000 m2) and tree island approximately 2-5 m thick (Hodgkinson et al., 2017a). Evidence of tension cracks throughout the mass and the front face fretting away resulted in the call for targeted remediation. Headscarp regression was also identified as a hazard with tension cracks present above headscarps towards the crest of the slope. Continued generation of rockfall from the highly dilated and weakened slope crest was identified as a lasting hazard at this site along with boulders lodged partially up the slope.
Despite the similarities of the two sites, their long-term response to sluicing in terms of rockfall has been very different. Ohau Point produced the highest rockfall count of any sluiced landslide, with 51 rockfall events, and the second highest cumulative volume, with 624 m3. In comparison, Paparoa South produced just 24.5 m3 of rockfall from June 2017 – March 2023 over 23 individual rockfall events.
5. Legacy Effects of Sluicing
The results of this study show a complicated picture of the interaction between sluicing and the long-term behaviours of a slope. The Northern and Southern Transport Corridors were sluiced to different extents and produced unique rockfall trends and post-event behaviours. Overall, extensive sluicing of slopes such as in the Northern Transport Corridor appears to have limited long-term benefits and does not effectively reduce post-event rockfall hazard. Ultimately, the legacy effect of sluicing on a slope is likely to be highly site specific.
The data can be interpreted in multiple ways, and there is no one reason that can be assigned to the patterns seen without further studies. Potential reasons for the different rockfall behaviour include:
a) We didn’t sluice enough.
b) Sluicing made the slopes moreunstable.
c) These were the most active sites and would have produced heightened rockfall regardless.
All three reasons could explain why sluiced landslides produced more rockfall than non-sluiced landslides. Too much material may have been left on the slopes which later failed. This likely occurred in the south as the aim wasn’t to flush the scarps back to bedrock as it was in the north. Alternatively, in some cases sluicing may have degraded the slopes more by saturating the landslide debris on the slopes and increasing its weight, unevenly eroding and destabilising areas of marginally stable debris, and infiltrating into the underlying fractured rock mass, reducing its strength and further dilating discontinuities. All three of these processes are not mutually exclusive and may have occurred concurrently, as the post-earthquake slope conditions and response of a slope to sluicing are site specific.
Ascertaining which of these reasons might be the cause of these observations requires a more in-depth analysis of the sluicing data and individual behaviour of the slopes. Most likely it is a combination of these factors, heavily influenced by site-specific characteristics of the individual failures and operational decision making.
5.1 The Role of Sluicing in the Next Earthquake
The context surrounding the choice to use sluicing in a future earthquake is important to understand. In the next event, sluicing is likely to form part of a wider earthquake response plan. If sluicing is to be effectively utilised in a future event, its limitations need to be clearly understood and a cost-benefit analysis undertaken prior to committing to a significant sluicing campaign. This includes only targeting slopes that are geographically, morphologically, and geologically suitable. There are high upfront costs associated with sluicing, but it has some benefits in certain circumstances, including the speed and safety elements required in the immediate emergency response period. Due to this, there is likely a place for sluicing in a future event where a corridor must be urgently re-opened and safety of earthworks crews is a central focus.
An Alpine Fault earthquake has a 75% probability of occurring in the next 50 years, with modelled landslide numbers as high as 70,000 across the Southern Alps (Howarth et al., 2021; Robinson et al., 2016). The South Island Alpine Fault Earthquake Response (SAFER) Framework has identified segments on six state highways where landslides will likely leave the corridor impassable (Emergency Management Southland, 2018). Key among them are Lewis Pass (SH7), Arthurs Pass (SH73), Buller Gorge and Haast Pass (SH6) connecting the West Coast Region to the rest of the South Island (Orchiston et al., 2016). Helicopter sluicing may be one of the tools used to regain access into and across the Southern Alps. Sluicing will likely be most effective in an approach like what was used in the Southern Transport Corridor, targeting the most unstable debris to allow safer access for earthworks crews. Geological conditions where slope failures are shallow, relatively constrained in size, and have occurred along a competent basal plane would likely lead to the greatest sluicing effect over the shortest duration. These conditions, in combination with a geographically suitable location near a water source, will be key to any sluicing campaign after an Alpine Fault Earthquake.
Published studies on the Wellington Fault put the probability of a Wellington-Hutt Valley segment rupture at 11% in the next 100 years (Rhoades et al., 2011). SH2 is a critical link between the CBD and the Hutt Valley and will likely be inundated with landslides from the coastal cliffs which reach up to 100 m at angles between 40°- 47°. As this 4.6 km section of the road is coastal, Helicopter Sluicing may be an appropriate and effective method of reducing rockfall hazard to earthworks crews regaining access across the road link. Rakaia Terrane Greywacke is the main bedrock lithology on the western side of the Wellington Fault (Begg & Johnston 2000) and is similar to the Pahau Terrane Greywacke which underlies most of the sluiced Kaikōura slope failures. As both locations have similar geographic and geological conditions, the Kaikōura Sluicing campaign may be a good analogy to assess how sluicing may be used on potential co-seismic failures along SH2.
6. Key Findings
The 2016 Kaikōura Earthquake produced more than 30,000 co-seismic landslides, including 1,500 which directly impacted critical transport infrastructure. This critical infrastructure required comprehensive recovery and resilience works to clear and stabilise the adjacent earthquake-damaged slopes, along with reinstating and re-aligning the transport corridor. Helicopter sluicing played a critical part of this slope repair and recovery process due to the steep and inaccessible terrain, in combination with the significant time pressure to reinstate SH1. Although helicopter sluicing is a novel and unconventional method, previously untested at a corridor wide scale, >6500 hours of helicopter time were invested in the sluicing campaign, with more than 220 million litres of sea water dropped on 42 coastal landslides.
Our quantitative analysis found that while just 20% of the slopes adjacent to the Transport Corridor north of Kaikōura were sluiced, they account for 51% of all rockfall events and 87% of all rockfall volume to fail in the first 6 years post-earthquake. In comparison, 40% of the southern slopes were sluiced and these produced 59% of all rockfall events but just 26% of the rockfall volume in the same period. Possible explanations for this disparity in long-term rockfall rates include:
a) We didn’t sluice enough.
b) Sluicing made the slopes more unstable.
c) These were the most active sites and would have produced heightened rockfall anyway.
Overall, extensive sluicing of slopes like what was seen in the Northern Transport Corridor is shown to have limited long-term benefits and does not appear to effectively reduce post-event rockfall hazard. Nevertheless, sluicing does appear to play an important role in the immediate remediation of slopes when used under the right conditions.
The result of this study shows that the benefit of sluicing is seen most clearly in the short-term recovery of a Transport Corridor. It can be a useful tool to reduce short-term rockfall hazard safely and rapidly if targeted on landslides in appropriate locations close to a water source. The landslides must also have appropriate geology and morphology that make sluicing or scaling effective.
References
Asadi, M., Orense, R.P., Asadi, M., Pender, M.J., 2019. Maximum dry density test to quantify pumice content in natural soils. Soils & Foundations. 59(2), 532–543.
Avery, M. B., & Barrett, C. (2021). Innovation Under Pressure – Ohau Point Slope Stabilisation NZGS Symposium, Dunedin.
Begg J.G., & Johnston M.R. (2000). Geology of the Wellington area. Institute of Geological & Nuclear Sciences.
Budimir, M. E. A., Atkinson, P. M., & Lewis, H. G. (2014). Earthquake-and-landslide events are associated with more fatalities than earthquakes alone. Natural hazards (Dordrecht), 72(2), 895-914. https://doi.org/10.1007/s11069-014-1044-4
Dahlquist, M. P., & West, A. J. (2019). Initiation and runout of post-seismic debris flows: Insights from the 2015 Gorkha Earthquake. Geophysical Research Letters, 46(16), 9658-9668.
Emergency Management Southland. (2018). Safer Framework: South Island/Te waipounamu Apline failure Earthquake Response. AF8.
GeoNet. (2017). Debbie and the landslide dams. https://www.geonet.org.nz/news/6m9tYfIAKIGI8CKYME8Cg4
Green, R. (2024). Helicopter Sluicing for Mitigating Rockfall Risk – Documenting experience from the 2016 Kaikōura Earthquake. Rori Green consulting Limited.
Green, R., & Finlan, S. (2019). Helicopter Sluicing for Rockfall Risk mitigation in responce to the 2016 Kaikōura Earthquake. Prepared for the 70th Highway Geology Symposium.
Hodgkinson, J., McMorran, T., Mason, D., Mabin, M., Ridl, R., Barnard, S., & Dykstra, J. (2017a, 24/04/2017). Geology of Kaikōura Transport Corridor – Kaikōura North. NCTIR.
Hodgkinson, J., McMorran, T., Mason, D., Mabin, M., Ridl, R., Barnard, S., & Dykstra, J. (2017b, 17/07/2017). Geology of Kaikōura Transport Corridor – Kaikōura South. NCTIR.
Justice, R., Saul, G., & Mason, D. (2018). Kaikōura earthquake slope hazards – risk mitigation and network resilience. NZ Geomechanics News, 96, 26-37.
Kincey, M. E., Rosser, N. J., Robinson, T. R., Densmore, A. L., Shrestha, R., Pujara, D. S., Oven, K. J., Williams, J. G., & Swirad, Z. M. (2021). Evolution of Co-seismic and Post-seismic Landsliding After the 2015 Mw 7.8 Gorkha Earthquake, Nepal. Journal of geophysical research. Earth surface, 126(3), n/a-n/a. https://doi.org/10.1029/2020JF005803
Marc, O., Hovius, N., Meunier, P., Uchida, T., & Hayashi, S. (2015). Transient changes of landslide rates after earthquakes. Geology (Boulder), 43(10), 883-886. https://doi.org/10.1130/G36961.1
Mason, D., & Brabhaharan, P. (2021). Characterisation of transport resilience and measures to enhance resilience in the recovery after the 2016 Kaikōura earthquake. Bulletin of the New Zealand Society for Earthquake Engineering, 54(2), 69-81. https://doi.org/10.5459/bnzsee.54.2.69-81
Massey, C. I., Townsend, D., Rathje, E., Allstadt, K. E., Lukovic, B., Kaneko, Y., Bradley, B., Wartman, J., Jibson, R. W., Petley, D. N., Horspool, N., Hamling, I., Carey, J., Cox, S., Davidson, J., Dellow, S., Godt, J. W., Holden, C., Jones, K., Kaiser, A., Little, M., Lyndsell, B., McColl, S., Morgenstern, R., Rengers, F. K., Rhoades, D., Rosser, B., Strong, D., Singeisen, C., & Villeneuve, M. (2018). Landslides triggered by the 14 November 2016 Mw 7.8 Kaikōura earthquake, New Zealand. Bulletin of the Seismological Society of America, 108(3B), 1630-1648. https://doi.org/10.1785/0120170305
Massey, C. I., Townsend, D. T., Lukovic, B., Morgenstern, R., Jones, K., Rosser, B., & de Vilder, S. (2020). Landslides triggered by the MW7.8 14 November 2016 Kaikōura earthquake: an update. Landslides, 17(10), 2401-2408. https://doi.org/10.1007/s10346-020-01439-x
Orchiston, C., Davies, T., R., L., Wilson, T., Mitchell, J., & Hughes, M. (2016). Alpine Fault Magnitude 8 Hazard Scenario. AF8.
Rhoades, D., Van Dissen, R., Langridge, R., Little, T., Ninis, D., Smith, E., & Robinson, R. (2011). Re-evaluation of conditional probability of rupture of the Wellington-Hutt Valley segment of the Wellington Fault. Bulletin of the New Zealand Society for Earthquake Engineering, 44(2), 77-86.
Robinson, T., Davies, T., Wilson, T., & Orchiston, C. (2016). Co-seismic landsliding estimates for an Alpine Fault earthquake and the consequences for erosion of the Southern Alps, New Zealand. Geomorphology, 263, 71-86.
Robinson, T. R., Rosser, N. J., Davies, T. R. H., Wilson, T. M., & Orchiston, C. (2018). Near-real-time modeling of landslide impacts to inform rapid response; an example from the 2016 Kaikōura, New Zealand, earthquake. Bulletin of the Seismological Society of America, 108(3B), 1665-1682. https://doi.org/10.1785/0120170234
