An Investigation into the Impacts of Human Presence and Activities on the Environment and Ecosystems in the Caves of the Mendip Hills.

Andrew Chamberlain

FdSc Countryside Management


Purpose of the Project:AimsObjectivesHypothesis5555
Literary Review6
Equipment and Methodology:EquipmentMethodology111112
Control Experiment – Waterwheel SwalletAtmospheric PressureHumidity and TemperatureCarbon Dioxide Levels 13141519
ResultsCave Visitor NumbersShowcavesGoatchurch CavernShowcaves – Wookey Hole and Goughs Cave.Comparisons in relationships (Showcaves)Goatchurch Cavern20202020212325 
Comparative Results Between Showcaves, Goatchurch Cavern and Waterwheel SwalletCO2 LevelsHumidity LevelsTemperatures 28282930
Changes Made Since Proposal34
Further Studies35
Appendix A – Locations of Caves Studied36
Appendix B Research Project Proposal44
Appendix C – Research Project (Cave Surveying) Risk Assessment46
Appendix D – Raw Data Tables48
Appendix E – PowerPoint Presentation Slides52
Appendix F – Data Calculations from Minitab62
Word count 4997 not including Titles, Tables, illustrations, footnotes and appendices


The Mendip Hills are a limestone range in Somerset in the Southwest of England. They cover an area of 78345 hectares.

As in nearly all limestone areas, the hills are peppered with caves. The  Mendip Hills have in excess of 200 explorable caves, with 195 being listed and described in Mendip Underground (Irwin 1999) and more being discovered and extended each year.

Much is understood about the physical impact of people visiting these caves[1], but little is known about the damage done to these fragile places simply by the unseen hazards  of simply being within them.

Humans entering a cave system bring with them many pollutants. Apart from the obvious dirt and detritus, people bring with them:

  • Body temperature
  • Noise
  • Humidity (both from breath and perspiration)
  • Carbon Dioxide.

It is known that CO2 in concentrations of over 2400 parts per million can dissolve limestone (Sargent 2006). This project is looking at the possible combined effects of a range of factors.

This project will be looking at how large numbers of visitors both in show caves and caves used by commercial caving operators can possibly adversely effect the cave environment.

Purpose of the project


  • To study the impacts of the presence of people in the cave environment.
  • To examine the CO2 levels in the caves in relation to visitor numbers, and other environmental factors.
  • To examine the humidity levels in caves in relation to visitor numbers, temperature and CO2 levels.
  • To examine the temperature in caves in relation to visitor numbers and CO2 levels.
  • To determine what effects these levels may have on the caves and fauna within them.
  • To examine the effects of the artificial infrastructure on the environment of showcaves.


  • To measure CO2 levels in each caves over a reasonable timescale.
  • To measure relative humidity levels in each cave over a reasonable timescale.
  • To measure temperature in each cave over a reasonable timescale.
  • To provide tangible figures and statistics to improve how commercial operators care for caves they own or use.


  • As the number of visitors to a cave system increases, so does the level of CO2 in the atmosphere.
  • As the number of visitors to a cave system increases, so do the relative humidity and temperature.
  • The levels of the above will in certain situations reach levels that pose a threat to both the fauna of the cave and the integrity of the cave itself.
  • Greatest effects will be noticed in showcaves with high numbers of visitors and commercial venues in peak seasons.
  • The artificial environment presented in showcaves will pose a treat to both the caves and their ecosystem.

Literary Review

Cave Science in British Caving Press and Literature

Since caving began seriously as a pastime and speleology as a science in the early twentieth century, knowledge, techniques and equipment have advanced beyond recognition.

Club caving properly started around 1900 with around 30 clubs in existence by 1930. Although the idea for a national caving body was proposed in 1935, the Second World War largely interrupted British caving. It was not until 1948 and the recommencement of caving by adults[2] that national bodies were formed. Amongst these was the Cave Research Group[3].

Interestingly the relationship with Carbon Dioxide and water in caves has from the beginning been of interest to researchers. In Volume one, issue one of the CRG Transactions (CRG 1948) an article entitled “Some Notes on he Relationship of Carbon Dioxide with Water in Caves” by M. H. Chantry was published. 

In this article, Chantry discusses how that during warmer times when water enters a cave, it is likely to cool (as natural caves have a relatively constant temperature) and therefore loose some of it’s ability to dissolve CO2 which will be released this into the caves atmosphere[4]. On colder days, the reverse happens and as the water travels through the cave system it increases in temperature along with it’s ability to dissolve CO2. When this happens, not only will it dissolve atmospheric CO2 but also due to it’s increased acidity will also now be able to dissolve the very rock of the cave.

The formula he gives for this reaction is:

CaCO3 + H2O + CO2 = Ca(HCO3)2

Much of the article by Chantry uses presumptive and assumptive language, so it’s possible to presume that he is basing his theory on research done outside of the cave environment. This could either be because of required scientific equipment not being available or more likely that it was to fragile and cumbersome o consider taking into a speleological environment.

For many years, little was written on this subject in the British press, with most cavers concentrating on the discovery of new cave systems and techniques for their exploration.

Most caving books written are either descriptions of specific caves, their discovery, features and geology such as “Swildons Hole – 100 Years of Exploration” (Irwin et al 2007) or guide books (such as Mendip Underground (Irwin 1999)) to help the sporting caver discover and navigate around new cave systems. Both types of book include only the most minor scientific detail.

When “Caves and Cave Life” (Chapman 1993) was published it was the first time that cave science had been presented in a way that was accessible and understandable to the layperson. Caving and caving books had been for many years the domain of legend and the eccentric. Limestone and Caves of the Mendip Hills (Drew et al 1975) had presented science, but in a way that only people familiar with caves and their terminology would understand.

In “Wookey Hole – Its Caves and Cave Dwellers” (Balch 1914) combines basic science and knowledge of the human history of the caves with ancient poems and legends, passing these off as relevant as the facts. It is interesting to note how Balch notes that most of the legends now associated with these and other caves can be founded in science. Balch does explain that the shapes thought to be amongst other things, a petrified witch are in fact just a massive stalagmite formations and that the loud banging noises heard in Wookey Hole and the Howlings heard in Goatchurch were in fact just movements of the (as yet undiscovered) underground river and thermally induced air movements respectively.

In Caves and Cave Life, Chapman discuses further the effect of increased humidity in caves and where in caves that are not evolved in parallel with this, an increase in humidity levels can in fact drown many of the small invertebrates on which the cave ecosystem relies. At the other extreme, some cave life has developed to thrive in more humid conditions. An example of this is some species of Choleva beetle make a cocoon out of mud or clay in which it seals itself over the dry summer to avoid desiccating. It will then break out when the damp weather returns and carry on with its life.

Chapman (1993) does explain how certain British bats such as the Greater Horseshoe (Rhinolophus ferrumequinum) has a very high tolerance to some environmental changes but not to others. They can allow their body temperature to range from 5° to 40°C to allow for equilibrium with atmospheric conditions (so as not to waste valuable energy reserves keeping their bodies warm). These bats do have a low tolerance to changes in CO2 levels, noise and other disturbance and will seek out new locations for their roost if required. The problem seems to occur when all suitable locations in a cave have been exhausted causing the bats to leave and hopefully find new venues. Being mammals, bats have a similar requirement for air quality as humans, therefore easy rises in this; either by natural or manmade can have serious consequences.

Humidity levels are influenced by the conditions outside the cave. If a cave has either one entrance higher than another, or significant mesocavern[5] cracks then air will be drawn through the caves causing the cave to dry out (known as the chimney effect). This is much greater in cold weather when the difference between the subterranean temperature and that above ground is at it’s greatest (Chapman 1993). In Mendip caves, this effect is most noticeable in Goatchurch cavern, where on a cold day the rock inside can be extremely dry compared to outside with a noticeable warm draught coming out of the entrance. This cave both has two entrances and numerous mesocavern cracks along the adjoining hillside. 

Sargent (2006) talks about CO2 levels in Mendip caves and how they are affected both by the depth of the passage and the external conditions. Significantly though, she looks at the effect that a various numbers of visitors have on different types of caves.

Sargent notes that a CO2 level of 2400 ppm will corrode limestone. She records that at 2pm on 24th August 2005 a level of 3800ppm[6] was recorded in the “Diamond Chamber” of Goughs Cave. This level was reached after 1351 people had passed through the cave that day. It can therefore be presumed on the basis of these figures that this must be having a detrimental effect on the structure of and the formations within the cave. 

As caves are both created by and destroyed by CO2 in various compounds it can be assumed that the processes in Goughs Cave are being greatly sped up by these levels.

Most caves are created by the erosive effects of water. This can either be physical or chemical erosion. Usually physical in passages with fast flowing water and chemical in more sedentary areas.

phreatic vadose diagram.psd
Fig 1.1 – Showing how the position of Phreatic and Vadose Passageways relate to water levels and depths in the cave.

As water runs over the ground above the cave, it dissolves mater and other limestone types compounds and slowly becomes more acidic, therefore causing it to dissolve even more rock.

Drew (1975) suggests that the caves in the Burrington Combe area are mainly formed by phreatic erosion. This is where the passages are formed underwater with the erosion happening evenly on all surfaces. This type of passage usually has a characteristic round profile (fig 1.2) as seen in the famous “Drainpipe”, the lowest part of Goatchurch Cavern.

As water levels fell during the creation of the cave, the passage turned from phreatic to vadose. This is where the water only erodes the bottom of the passageway as in a normal stream way. The underwater sections in Goughs Cave and Wookey Hole will still be undergoing phreatic processes, were as the sections with airspace above the water will be undergoing vadose processes. These processes are continual throughout the life of the cave with continual processes of erosion and deposition constantly changing the “live” cave.

Fig 1.2 – The Drain pipe in Goatchurch Cavern, a Classic Vadose Passageway. Photo – Creative Commons

Equipment and Methodology


  • Datalogger – Dataharvest Easysense Q5+ Bluetooth
  • Dataharvest Easysense CO2 sensor
  • Microsoft Excel Mac 2011
  • Dataharvest Easysense Software
  • Apple iMac running WindowsTM XP via VM ware

The Easysense Q5 datalogger was used connected to the CO2 sensor for all underground surveys, 

The datalogger itself is water resistant, but the CO2 monitor is not, so survey locations had to be chosen where the equipment would not be effected by running water or drips.

The Easysense Q5 can record the following data:

  • Light in Lux (lx)[7]
  • Sound in decibels (dBA)1
  • Temperature (°C selected)
  • Air pressure (kPa)
  • Humidity (%RH)
  • CO2 (ppm) via external sensor

The datalogger has a built in rechargeable battery with a claimed thirty day battery life. It was notice that when the COsensor is used, this falls to around seventeen hours. This limited the experiment in the non show caves, although, in the show caves it was possible to connect to mains electricity to give more recording.

For later experiments, a second logger was acquired. This was used outside the caves to record external data.


The surveys were undertaken between 7th February and 26th March 2012[8]

Four caves were surveyed as part of this experiment. These are:

  • Waterwheel Swallet – 15th March 2012.
  • Goatchurch Cavern – February 7th, 8th, 26th and March 8th 2012.
  • Wookey Hole – February 25th and 26th 2012.
  • Goughs Cave (Cheddar) – February 11th and 12th 2012.

The datalogger was positioned in and area of each cave that is on the main route through the system, but an area where people pause for one reason or another.

Raw data was received into the Easysense software where it was exported into Microsoft Excel to allow for detailed analysis.

The timescale of the experiment was limited by the battery life of the datalogger in Goatchurch and Waterwheel. In Goughs and Wookey, this was overcome by using the mains power supply in the cave.

In Waterwheel and Goatchurch, the logger was set in “Easylog” mode to record as much data as possible within the limited life of the battery. 

In Goughs and Wookey Hole, the datalogger was set to record every 2 minutes, starting at 6am on a Saturday and recording for 48 hours.

Control Experiment – Waterwheel Swallet

Waterwheel Swallet was decided upon as the venue for the control. The reasoning behind this is that it’s a “Key Controlled[9]” cave where it was possible to guarantee there would be no-one visiting the cave during the time of the experiment.

One data logger was placed in the cave at an area known as “The Folly” (As marked on fig 3.1). This area was chosen as it was both a reasonable distance (20 metres) underground so as not to be affected by external factors. Also this area is relatively dry, so the readings should not be affected by standing water or condensation. This is as much to protect the equipment, as it is the readings.

waterwheel survey.jpg
Fig 3.1 – Survey of Waterwheel Swallet showing locations of surveying equipment (Stanton 1987)

Another datalogger (without the C02) was placed in long grass at the surface about ten metres (horizontally) from and 4 metres above the cave entrance.

Both loggers were started at the same time and (other than C02) were set to record the same information. The outside logger would be subject to all the conditions on the surface, whilst at the same time, the underground logger would record the same data (plus CO2 levels) for the underground environment.

Atmospheric Pressure

As shown in the table (fig 3.2) the pressure at both sites maintains a similar downward trend. At the beginning, the readings are 0.4kPa apart increasing to roughly 0.5kPa at the end of the test. The difference in the two sets or readings can be easily explained by the difference in altitude. The norm is, the greater the altitude, the lower the atmospheric pressure. This data does not form part of the main research, but is included here as part of the control.

Fig 3.2 – Atmospheric pressure at Waterwheel Swallet

Humidity and Temperature

Fig 3.3 shows that the Relative humidity inside the cave is higher than that outside. This would be expected, as the outside weather was warm and spring like, whereas the inside of the cave was (although not wet) quite damp.

It can be noted that the humidity levels in the cave have a slight upwards trend whilst those on the surface trend slightly downwards. Chapman (1993, p63) states that the relative humidity of caves decreases when the outside temperature increases. Therefore if you reverse this to allow for the rising humidity in the cave it would suggest that the temperature should be falling on the outside. This is proven in fig 3.4 that shows the underground temperature being relatively constant whilst the outside temperature rises towards mid day and then falls to between four and six degrees over night.

Fig 3.3 – Waterwheel Swallet Humidity %RH

Interestingly, it can be seen in fig 3.5 , that when looked at alone, the graph for the temperature underground follows a very similar shape to that of the underground CO2 as shown in fig 3.9. Both graphs show a downward trend and have their relative peaks and troughs in the same places, it is therefore possible to assume that these two levels are directly related and proportional to each other.

Fig 3.4 – Waterwheel Swallet Comparative Temperatures
Fig 3.5 – Waterwheel Underground Temperatures

After looking at the above results, it was decided to compare the relationship between the humidity levels and temperature both underground and on the surface. Fig 3.8.1 shows surface humidity compared against surface temperature while fig 3.8.2 shows the same for the underground environ.

The surface figures show an R2 value of 0.80004 and a P value of 0.000. These figures indicate a significant negative relationship between surface temperature and humidity.

The same readings taken underground show an R2 value of 0.46988 and a P value of 0.000, this again shows a significant negative relationship between underground temperatures and humidity levels.

These figures (Table 3.6) indicate that the underground environment follows the same correlations in atmospheric conditions as those found on the surface.

Table 3.6 – Regression results and significance in the relationship between temperature and humidity levels for Waterwheel Swallet.
Table 3.7 – Temperature Levels (?C) Recorded for Waterwheel Swallet
Surface2.4612.60 7.57 2.46 
Fig 3.8.1 – Humidity in Relation to Temperature – Waterwheel Swallet (Surface)
Fig 3.8.2 – Humidity in Relation to Temperature – Waterwheel Swallet (Underground)

Waterwheel Carbon Dioxide (CO2) Levels

Table 3.9 – COLevels (ppm) Recorded in Waterwheel Swallet

As can be seen in table 3.9 and fig 3.9.1, the CO2 levels are relatively low. They fluctuate at about the same times as the temperatures shown in fig 3.5. It would be viable to assume that the two are naturally linked, with the peaks and troughs on the graph appearing at the same times.

The mean CO2 level found in the cave was 450 ppm where as a snapshot reading taken outside immediately after retrieving the data logger showed a reading of 229ppm. This would indicate that caves are naturally higher in CO2 than the terrestrial environment. This could be either due to the temperature difference or that due to poor air circulation, CO2naturally accumulates and disperses slower in caves possibly simply due to CO2 being heavier than air, thus falling into the caves. 

Fig 3.9.1 – Waterwheel Underground CO2 Levels


Cave Visitor Numbers


During the time the recordings were being made at the two showcave locations, they had much lower visitor numbers than would normally be expected. Cheddar had approximately 260 visitors per day of the experiment[10]. Wookey hole had higher numbers 494 and 418 respectively for the Saturday and Sunday of the experiment.

Goatchurch Cavern

Goatchurch Cavern was surveyed on four occasions to hopefully ensure contact with a wide number of groups and group sizes.

The datalogger was placed in the same position for each recording (Appendix A – fig 6.1).

The details of the survey sessions are[11]:

  1. GC1 – Saturday 4th February 2012  – started at 11.30am.
  2. Group 1, 4 people at 12pm.
  3. Group 2 4 people (route unknown).
  4. Group 3, School group, 8 people, paused by logger.
  5. GC2 – Tuesday 7th February 2012 – started at 11.04am
  6. Group 1, 11 people at 11.30am.
  7. Group 2, 28 people at 1pm (no evidence this group passed the datalogger).
  8. Group 3, 12 people at 01.30pm.
  9. GC3 – Thursday 23rd February 2012 – started at 10.15am
  10. Group 1, 9 people at 10.45 am.
  11. GC4 – Wednesday 7th March 2012 – started at 9.40am
  12. Group 1, 45 in total, over 6 groups.

Showcaves – Wookey Hole and Goughs Cave.

For both Wookey Hole and Goughs Caves, The dataloggers were positioned in areas of the cave where they could use mains electricity and not be disturbed. Both units were set to record from 6 am (on the Saturday of the experiment) for forty-eight hours.

Fig 4.1.0 shows the temperatures throughout the tests with peaks starting at 9am and ending at about 5pm coinciding with the times the lights are on in the caves. Both temperatures are continually higher than seen in Waterwheel where a mean temperature of 9.12°C was recorded (fig 3.7). The trend lines for these results shows that the temperature fell relatively over the course of the experiment. It could be presumed that this is cyclical and will rise again when environmental conditions change.

show temps.png
Fig 4.1.0 – Comparative Temperatures in Showcaves

Fig 4.1.1 shows the showcave humidity levels that appear to mirror those (both in shape and timescales) of the temperatures in fig 4.1.0. This would suggest an inverse relationship.

show humid.png
Fig 4.1.1 – Comparative Humidity Levels in Showcaves

Fig 4.1.2 shows that the peaks and troughs in the CO2 levels appear during the same time period as for those in the temperature and humidity graphs. The COlevels are noticeably higher than those observed in Waterwheel (fig 3.9 and 3.9.1).

show co2.png
Fig 4.1.2 – Comparative CO2 Levels in Showcaves

Comparisons in relationships (Showcaves)

Figures 4.1.3 and 4.1.4 show the relationships between humidity levels and temperatures in both Wookey Hole and Goughs respectively. The regression lines show R2 values of 0.9646 and 0.96056 respectively. The P value for both sets of results is 0.000. These values show a significant negative relationship between temperature and humidity in the showcave surveys. 

wh humid  temp.png
Fig 4.1.3 – The Relationship between Humidity and Temperature in Wookey Hole
wookey humid temp.png
Fig 4.1.4 – The Relationship between Humidity and Temperature in Goughs Cave.

The relationship has also been calculated between COagainst humidity (fig 4.1.5) and temperature (fig 4.1.6). With R2values of 0.02333 and 0.04923 the relationship is not significant. This test was only done for Goughs cave as it was deemed un-necessary to repeat for Wookey Hole.

gough co2 humid.png
Fig 4.1.5 – The Relationship between CO2 and Humidity in Goughs Cave 
gough co2 temp.png
Fig 4.1.6 – The Relationship between CO2 and Temperature in Goughs Cave

Goatchurch Cavern

It is realised, that as the groups in this cave are relatively large, but quite brief and with the data logger only recording every two minutes, that some information will be missed and graphs not as accurate as hoped.

Fig 4.2.1 shows clearly the sound levels used as an indicator to he presence of people. It is noted that the noise is recorded at a maximum of 82.5dBA[12] The minimum recorded for all sessions was 40bDA[13].

Goat Sound.png
Fig 4.2.1 – Comparative Sound Levels in Goatchurch Cavern

Fig 4.2.2 shows how the CO2 levels rise dramatically when people are present in the cave. They gradually fall towards their normal level[14]

Goat co2.png
Fig 4.2.2- Comparative CO2 levels in Goatchurch Cavern

During the time of the experiments, all of the humidity readings gradually rose[15] with a spike each time at roughly the time people were in the caves. It would be safe to assume that human presence here has minimal effect and is acting in relation to other external factors.

The peaks in Sound and CO2 levels correspond to the number of visitors to the cave during each test, with the highest results coming from GC2 where there were both the most visitors and the largest number in an individual group.

Goat Humid.png
Fig 4.2.3 – Comparative humidity levels in Goatchurch Cavern

It can be seen in fig 4.2.4 that the temperatures fall at the beginning of the experiments[16] and then remain largely constant. There are small spikes at the same time as those in the CO2 and sound level graphs (figs 4.2.1 and 4.2.2). This shows that the presence of people in this situation had little effect. One can presume this effect will be more noticeable with larger number groups in high season. The highest temperature spike was seen in test GC2 as was the case with CO2and sound levels.

Goat temps.png
Fig 4.2.4 – Comparative temperatures in Goatchurch Cavern.

Comparative Results between Showcaves,

Goatchurch Cavern and Waterwheel

To enable the different types of caves to be compared against each other and significances investigated, the data categories have been complied into comparitive charts.


Fig 4.2.5 the error bars show that whilst the readings for the two showcaves are significantly similar, they are statistically significantly higher than those found in Waterwheel. At the same time, they are significantly similar to all of the readings for Goatchurch except for GC3, which is lower. All of the readings for Goatchurch are significantly similar to each other.

combined co2 bar.png
Fig 4.2.5 – Comparison of CO2 levels from all studies

Humidity Levels

Fig 4.2.6 shows that the readings taken on all of the surveys are significantly similar to each other. The only anomaly is Waterwheel shows to be different to GC1. 

combined humid bar.png
Fig 4.2.6 – Comparison of humidity levels from all study sites


Fig 4.2.7 shows that the temperatures in the showcaves (Wookey Hole and Goughs) were significantly higher than those recorded in the other caves. It is suggested that this higher level is due to the artificial lighting in these caves and possibly the greatly increased number of visitors.

Readings taken in Waterwheel and Goatchurch are statistically the same in relation to the error bar readings.

combined temp bar.png
Fig 4.2.7 – A comparison of temperatures recorded from all study sites


This series of experiments has shown that (as would be expected) that all factors relating to the environment in caves are linked either directly of indirectly.

It is known that high CO2 levels have damaging effects on the actual structure of the cave. Levels above 2400ppm corrode the limestone and calcite formations (Sargent 2006). Although levels this high were not experienced during this series of experiments, levels of 1797ppm were experienced in test GC2 with relatively low visitor numbers. It could therefore be assumed that on busy days the levels will reach higher levels and probably exceed the 2400pm level and thus be causing corrosion damage.

In caves such as Goatchurch Cavern, these high levels will not be constant throughout the day, week, year where as, they are more likely to be in the showcaves at Wookey Hole and Goughs (Cheddar). It is known that Goughs has recorded levels in excess of 3800ppm (Sargent 2006) and although this was on an exceptionally busy day in peak season, it can be assumed that high levels will be present whenever the cave is busy and therefore damage is constantly being done to the cave.

The high CO2 levels found in some caves are seemingly solely the effect of human presence in the caves. The results from Waterwheel Swallet (fig43689), where no people were present show that even though there is fluctuation in the CO2levels the levels do not rise greatly compared to their mean and the results from the other caves.

In the Goatchurch surveys, it is evident that the more people visit the area near the datalogger, the larger the variation from the base CO2 levels. This is the case with all the other factors recorded[17]. This shows that human presence in normal caves effects the environment across the whole spectrum. These changes are seemingly short lived and quickly revert back to their base levels. Goatchurch cavern is inevitably suffering as a result of these levels, but the view within the caving community is that this venue is a sacrificial cave (and most commercial caving activity focused there), so any damage done here avoids damage being done elsewhere. 

The situation in the two showcaves is much different. 

Carbon Dioxide levels are much higher, which can be almost certainly attributed to the high number of people in the cave. These levels have serious implications for the structure of the cave, but most importantly for the health and safety of staff and visitors[18] especially those with respiratory problems such as asthma.

The high COlevels in combination with the humidity of the cave and the general dampness will be causing the once pristine formations to corrode. 

A highly significant factor in the showcave environment is the huge increase in the ambient light levels[19]. Throughout both showcaves, many different ferns, lichens and other plants can be found. These rely on light to survive. No plants would be naturally found in any cave past the twighlight zone[20], but many are present here.

Hartstongue GC.jpg
Fig 5.0 – Heartstongue (Phylitis scolopendrium) fern in Goughs Cave 

The Heartstongue Fern (Phylitis scolopendrium) is found throughout Goughs Cave and is pictured in fig 5.0 growing in an area known as the cascade, where there would be no natural light. These ferns need very little natural light, with the Heartstongue requiring only 10 – 18lx thus being able to thrive in the damp showcave, usually near the light installations[21].

It is difficult to assess with the results gained during these tests whether nay of the cave fauna is being disturbed. During the time of the tests in Goatchurch Cavern, a population of Lesser Horseshoe Bats (Rhinophus hipposideros) were observed in the same are as the data logger. It was noted that on most occasions, these had not moved between the datalogger being placed and collected. It could therefore be suggested that the presence of the groups in this cave do not disturb or displace the bats although Chapman (1993, p127) suggests that peoples perception of bat colony size is relative. If you were caving fifty years ago and observed colonies of one hundred or more, you would perceive correctly that the numbers have fallen, and therefore question why, but if you had only visited in the 1980s, then the chances are that the colonies are possibly larger today. It is for this reason that local bat groups are undertaking long-term observations.

It is known that Goughs Cave has one of the most significant bat colonies in the area[22], but during the tests and after discussions with many of the caves staff, it is noted that they are very rarely observed in the showcave areas. This would suggest that they are being disturbed by human factors and are moving in other less accessible parts of the cave system. Work is needed to be done to find out the factors that disturb these bats and find ways to minimise them.

The results from this series of tests shows that humans are having serious impacts on the speleological environment and the people that own, use and exploit them need to be made aware that their actions are irreversible and that they have a duty to protect these natural assets for future generations.


This project was limited in it’s outcome by the following factors:

  • The surveys were undertaken in the low season for both commercial caving and showcaves. Repeating over the summer month would attain more significant results.
  • Spreading the experiment over many sites limited the possibility to compare results to visitor numbers.
  • Accurate visitor numbers were impossible to obtain for Goatchurch Cavern. To be accurate, the cave would need monitoring for the duration of the experiment.
  • Cheddar caves declined to give accurate visitor numbers, so the number used could not be guaranteed.
  • The battery life of the datalogger was not as expected therefore greatly shortening the experiment. Alternative equipment would be sourced if the experiment were to be repeated.
  • It was not possible to survey Swildons Hole (an active river cave) due to the equipment not being waterproof.
  • Due to the short timescale and licensing issues, it was not possible to monitor the bat populations other than to make observations.

Changes Made Since Proposal

  • Grebe Swallet was not surveyed due to lack of access.
  • Swildons Hole was not surveyed due to the equipment not being waterproof.
  • Kick sampling and soil testing was not undertaken as all caves surveyed were dry. Also, it is not permitted to remove soil.

Further Studies

Taking this research forward, the next step would be to monitor the situation in active / wet caves where the situation could be more serious. Here, the ph level of water could be monitored throughout the cave along with monitoring speleothems[23] that are actively being dissolutioned.

Another direction of research would be monitor an individual cave to monitor the levels at low and high season to see recovery rates and the maximum levels of the various factors 


This research project has shown that simply by being in a cave, people have a damaging effect on the ecosystem.

Human presence not only raises CO2 levels, but also temperature, whilst at the same time bringing noise and light into an environment where hey would otherwise not be found.

Showcave owners need to be made aware of the significant damage they are doing to the asset of which they are merely a guardian. Some operators seem to view the caves as simply an asset from which to make as much money as possible with little or no regard for how it is to be preserved into the future. Simple, cheap changes to the infrastructure will make huge changes, both to how the cave survives and the running costs.

Commercial caving operators could maybe adapt the way they operate to minimise their impact

Caves are extremely vulnerable and we need to do everything we can to preserve them.

Appendix A – Locations of Caves Studied

Locations of Caves Studied

This study is based in and around the caves of the Mendip Hill. This is an area of (mainly) carboniferous limestone in the north of Somerset. They rise steeply from barely above sea level to 325 metres at their highest point, Beacon Batch on the area known as Blackdown.

cave locations.jpg
Fig 6.0 – OS 1:50,000 map showing cave locations

For this Project, four caves were chosen based on their characteristics, usage and access restrictions, these are:

Goatchurch Cavern – A “fossil[24]” cave in the Burrington Combe area. This cave as extremely high usage from novice cavers and outdoor pursuits providers. It is common during the height of the season for there to be in excess of 200 visitors in this cave during a day.

In Goatchurch Cavern, the instruments were placed in the large chamber at the bottom of the passage known as “Coal Shute”. It was attached to the cave wall using a series of cable ties[25]

goatchurch cavern.jpg
Fig 6.1 – Goatchurch Cavern showing locations of dataloggers (survey by MCG 1959)

Wookey Hole[26] – This is a medium sized showcave complex in part of a large (mainly underwater and inaccessible by all except for cave divers) cave system. It is unusual in as far as the showcave area is a fossil cave, but adjoining is a major system for cave diving exploration[27]. The River Axe runs directly through this cave. Wookey Hole has both artificial light and audio effects. This cave has two entrances (although one is a man made blasted passage).

wh blasting pic.jpg
Fig 6.2 – blasting the showcave entrance in the mid 70’s
wookey hole.jpg
Fig 6.3 – 2011 CDG survey of Wookey Hole showing locations of the showcave in relation to the cave as a whole and the position of the datalogger. Survey CDG 2012.

Gough’s Cave, Cheddar – This is a showcave complex based on fossil passages. There is a flooded section (currently being explored) but this is distinctly separate from the show cave section being surveyed. Goughs cave has artificial (tungsten lighting throughout). And can witness a large number of visitors. Over 1300 were recorded on some days during 2005 (Sargent 2006)[28].

Gough's Cave survey plan2.jpg
Fig 6.4 – Survey of Goughs Cave showing datalogger position ref

Waterwheel Swallet – This cave intersected and accessed by lengths of Victorian lead mine drainage passages (tailings). The upper sections of this cave are dry enough to be considered fossil, although there is a series of semi-flooded passages lower in the caves that are used as a “sporting” introduction to wet caving. Waterwheel is capped and locked for both safety and the protection of the cave[29]. It was therefore possible to guarantee through having possession of the key that it was not possible for anyone to access the cave during the survey. This location was used as the “control”

The locations for the position of the datalogger in each cave were chosen so as to keep the CO2 sensor dry, allow all sensors to record the required information and so as they did not get damaged by cave visitors, or themselves pose a hazard.

[1] It is accepted that people visiting and travelling through caves causes irreparable erosion of, damage to and dirtying of caves and cave formations. As soon as a cave is discovered this damage begins. Significant cave discoveries today are heavily protected with highly controlled access and stringent rules for any visitors, hopefully to keep damage to a minimum and wherever possible eliminate altogether. It is now common for significant discoveries to be photographed and recorded before being locked for all bar the most essential visits.

[2] Students and school children had been at the forefront of caving during the war years. They are credited with many discoveries and contained amongst their number William (Willy) Stanton who was to become one of the greats of British caving with many a discovery and paper to his name.

[3] The CRG over the years has been amalgamated into other organisations, notably, the National caving association and now the British Cave Research Association (BCRA) (BCRA 2012)

[4] We now know this to be one of the main causes of “Bad Air” where on warm days, the CO2 levels in parts of some caves can get dangerously high.

[5] A mesocavern crack is a rock fissure that will allow air and possibly water to flow into or out of a cave but is not large enough to allow access and be therefore classed as another entrance.

[6] This figure was attained during an experiment to correlate CO2 levels with depth in a cave, not as part of her main experiment.

[7] Light and sound levels, although recorded are not relative to the experiment. They are simply used to indicate the presence of people in the cave. The light sensor in the datalogger used records a minimum of 40dBA. This is considered human silence and is therefore used to indicate silence in this experiment.

[8] The showcave dates were chosen either end of the school half term to ensure high visitor numbers, whereas the Goatchurch dates were chosen to coincide with large groups from local outdoor activity providers.

[9] Many vulnerable caves are locked, either for their own protection or for various safety and access restrictions. Responsible clubs, organisations and individuals hold these keys. Waterwheel Swallet is itself a SSSI along with being withing the Blackmoor (Charterhouse on Mendip) SAM.

[10] During the time Hannah Sargent (2006) undertook her recording of CO2 Levels in Goughs Cave, they had between one thousand and thirteen hundred visitors per day. After the experiment for this project was undertaken, Cheddar Caves were unwilling to release official visitor numbers as agreed. The umbers quoted were obtained from an unofficial source at the caves.

[11] GC1 was recorded using the data logger set on it’s “Easylog” function, recording for as far as the batteries would allow. GC2 – GC4 were recorded using the datalogger set to record every two minutes for twenty-four hours. This increased battery life moderately.

[12] This is the equivalent of moderate talking. It can be assumed that higher readings were missed due to the recording intervals.

[13] 40dBA is the equivalent of “human silence” and is the lowest figure the Easysense datalogger will record.

[14] The minimum level recorded in Goatchurch was 249ppm compared to a minimum of 407 and 455ppm in Wookey hole and Goughs respectively (Appendix D). It can be assumed (comparing fig 4.2.2) that the readings will return to around this reading most nights when the cave is not visited.

[15] This is roughly inverse to the temperature as shown in fig 4.2.4. This effect is considered normal and was seen in the Waterwheel control experiment (Page 13)

[16] This is most likely due to residual heat in the datalogger from its transportation to the cave. This was not a problem in the show cave surveys as the dataloggers were set on a timer to come on at 6am, therefore having twenty two hours to acclimatise in the cave beforehand.

[17] Factors recorded in Goatchurch Cavern were CO2, temperature, humidity, light and sound.

[18] Safety wise, ( 2012) state that levels in between 1000 and 2000ppm will cause complaints of drowsiness and poor air whilst levels between 2000 and 5000ppm will cause “headaches, sleepiness and stagnant air along with poor concentration, loss of attention, increased heart rate and nausea. Groups entering this cave regularly experience these levels. It would be suggested that leaders are made aware of these facts and act accordingly when symptoms are witnessed.

[19] Natural cave light levels are 0lx. Wookey Hole had a maximum recorded level of 72lx

[20] The “twighlight zone” is the area of the cave beyond the entrance which witnesses the limits of natural light entering from outside.

[21] Wookey Hole recorded a maximum of 72lx, whereas Goughs didn’t record any reading during the test. This is believed to be due to the datalogger being obscured from the light by belongings of the cave guide during the test.

[22] Lesser Horseshoe (Rhinolophus hipposideros), Greater Horseshoe (Rhinolophus ferrumequinum), Serotine (Eptesicus serotinus), Noctule (Nyctalis noctua), Long Eared (Plectotus auritus), Daubenton’s (Myotis daubentoni) and Natterer’s Bats (Myotis nattereri) are known to be present in and around Cheddar Gorge and Goughs Cave (Altringham 2003).

[23] A “speleothem” is a term encompassing all cave formations such as Stalactites, Stalagmites and Helectites.

[24] A fossil cave is one devoid of flowing water and actively growing formations. These caves are in a fixed state and are particularly vulnerable to the effects of human presence

[25] Due to the nature of this site and the groups that use it, there was a high chance of the datalogger either being tampered with or removed altogether. Thankfully this did not happen

[26] Contrary to popular belief, Wookey Hole (the cave) gave it’s name to the village in which it sits, no the other way round. The first recorded mention of the cave was in Polyolbion (Drayton 1612) where it was referred to as “Ochy’s dreadful hole”. 

[27] Both Cheddar (Goughs) and Wookey are currently being pushed (explored at their limits and new passage discovered) by members of the CDG (cave diving group). The “Holy Grail” of Mendip caving is through this sort of exploration to link various systems of Mendip together into the fabled “Mendip Master System”.

[28] This cave shows very bad visual signs of human impact. These include highly eroded paths (which have throughout the showcave been concreted over, damage to formations and a large amount of alien plant growth such as ferns, lichens and moss, none of which are present in a natural cave past the “twighlight zone” as they all need light to grow. It would be safe to presume that a higher level of CO2 in this cave is also helping the plant matter thrive.

[29] Nearly all caves are registered as SSSI’s for their geological and biological importance. Some are locked (with keys controlled by responsible individuals or organisations). A few have other designations. Waterwheel is in an area registered a SAM (Scheduled Ancient Monument) due to the local Roman lead mining activity and is accordingly protected.

Leave a Reply