I was fortunate enough to spend some time bouldering in Red Rocks Canyon, NV at the end of 2018 and early 2019. During my time there, I heard a lot of various opinions about how long to wait for the rock to dry before climbing again. Most of the opinions ranged from 24 hours to 3 days (e.g., [1]), but then someone at the local gym said at least a week because the rock loses the majority of its integrity. So many different opinions got me thinking about rock mechanics… specifically, “what does the sandstone mechanics research show?”
Yes, the mechanical strength of sandstone is reduced when wet, so clearly you should not climb on wet rock… also, friction is important for climbing and water reduces friction (think of the caution signs that state ‘slippery when wet’). So, what if it looks dry but may still be ‘wet’, how much of a difference is there in mechanical properties? I’ve heard 75% decrease quoted multiple times within the climbing community. BUT, it’s important to understand how the tests were performed to come to those values, because tests performed in a laboratory are often conducted to evaluate extreme cases that a material might experience ‘in the wild’. This way engineers or designers can have better understanding of the full range of properties.
To create ‘dry’ samples, rocks are heated in an oven over a period of days – even with climate change these temperatures are quite high (105oC or 221oF). On the flip side, ‘wet’ sandstone is created by submerging specimens in a water bath for up to a week. Again, this is not representative of rainstorms, where most of the water drips off of the rock immediately…. You would need flooding of biblical proportions. If you’re experiencing either one of these conditions, you have bigger issues to deal with than ‘is the rock OK to climb on’. In any case, the drop in mechanical strength from extremely Dry to extremely Wet sandstone ranges between 15%-75%. (The wide range of values are likely due differences in sandstone locations: 15-40% from references [2-3], 75% from reference [4]*).
It’s helpful to actually look at some of the data (image below modified from [3]). To break a material, you would need to stretch it until it breaks into multiple pieces. Imagine pulling a rubber band until it snaps. Any increase in the material’s length would be the displacement (x-axis on the graph). The material will resist failure, requiring more force to be applied in order to continue stretching the material (y-axis on the graph). The peak point of this curve is the failure point (black X’s on the graph), so we can compare the failure loads… but we need to know which lines to compare.
Water in completely saturated (‘Wet’) samples only accounts for 3.5% of the rock volume, but I could not find the initial water content in the papers… so I emailed one of the authors to find out.** Before the rock drying process, the water weight in the rock was close to 1% (red line on graph). The difference between the 1% and 2% failure loads was ~100N or 20lbs… That’s 20lbs on a single hold, not distributed across two-four points of contact, which is easily within the range of weight differences between a set of climbers on any given day. (If you’re an engineer, yes, stress would be the more appropriate term to use here and is easy to calculate for comparison making some assumptions about hold sizes.)
So why does everyone think the rock will break shortly after rain?
I don’t know of that many holds that broke during the rainy month of January in 2019. However, during my time there, I did see and observe some climbs experience breakages. Most of them happened on the same day at the end of the December (2018) holiday season, where three holds broke on a dry day – a dry day after many many dry days. However, this is just an anecdote that will, hopefully, make more sense shortly.
Another important material failure mode that engineers need to consider is failure from fatigue loading, which essentially means that a material has a limited number of times that it can be loaded before failing. If you’ve ever discarded an old credit card by folding it back and forth until it breaks into two, this is a great example of fatigue loading. If you’ve only ever used the more efficient method of cutting your credit cards, I’ve provided a video showing fatigue failure with a credit card. Therefore, the more people climb on a climb, the more loading cycles the rock and hold experiences, increasing the likelihood of the rock failing from fatigue.
These failures occur suddenly, but damage had been accumulating at at a microscopic level over time. Failure starts with defects/cracks that go unnoticed or undetected. With the credit card example, you can usually visualize cracks form and spread well before the catastrophic failure (breakage into two pieces). Fatigue loading and the eventual cracks that from are a significant issue in engineering – it is also why plane fuselages and pipelines experience unexpected failures (e.g., Fuselage failure of Southwest flight 182 in 2011).[5]
It is also important to keep in mind that the more frequently phrases are repeated the more likely people are to believe the statement to be good and true,[6] regardless of new data presented.
- https://goo.gl/XQtY1e – blog with citations only to other blog posts
- Kim E & Changani H, 2016 Int. J of Rock Mechanics & Mining Sciences 88:23-28
- Zhou et al, 2018 Engineering Fracture Mechanics 193(47-65)
- Yadav S & Hagan P Coal Operators Conference, 2017 ** conference paper
- https://goo.gl/JCgXSr
- Thinking, Fast and Slow by Daniel Kahneman
* Interestingly, the mechanical properties of wet sandstone is not that far off of dry limestone…
**Researchers love it when someone actually finds and reads their work. It takes a lot of time and effort to put a paper together!… Speaking from experience.
Disclaimer – I’m not a rock engineer/civil engineer. I am a mechanical engineer who specializes in material failure (soft tissue failure) and failure mechanics with hydration.