top of page



Chronicling the imperceptible, seldom-experienced hazards lurking in unsuspecting landscapes – recording the historical memory of infrequent natural disaster zones that will still pose a rare but perennial risk to future inhabitants of these active geological areas across intervals of deep time.




The generally-accepted definition for a ‘risk assessment’ is a systematic process of evaluating the potential risks that may be involved within a projected activity or undertaking – including the very act of simply residing within an affected environmental setting. Consequently, these studious investigations and safety appraisals often attempt to mitigate [or sometimes prevent] the identifiable hazards from impacting the wellbeing of those who may, one unfortunate day, encounter such prospects. Many of these appraisals, however, are often temporally situated; usually factoring in the narrow swathes of time that are mostly beneficial for ameliorating contemporary exposure incidents – measured in decadal intervals or the limited lifespan of present-day generations. Examples of these time-limited assessments may be seen in various short-term pollution-prevention policies – isolating hazardous substances for today, with little practical plans for deeper time intervals. Such imperative surveys often fail to factor in these longer-term prospective risks for those who may, one day, inhabit these hazardous settings; prospective timeframes that will likely suffer from a loss of generational folk memory, the failed retention of adequate scientific documentation, or even the slow erosion of local mnemonic devices designed to forewarn future generations of hidden and enduring hazards.

On 15th August 1984 and 21st August 1986, two unpredictable and catastrophic events occurred around remote deep-water lakes situated in the Northwest regions of Cameroon. In the case of Lake Monoun (1984), local residents reported a loud noise emanating from the lakeside at around 22:30, followed by accounts of a ‘cloud’ emanating from the direction of this crater lake. By morning, 34 people lay dead with little evidence for an actual explanation. Survivors of this fatal mystery spoke of a whitish smoke-like mist that tasted bitter and acidic sweeping across their communities, suffocating people and wildlife trapped in its advance. Reports indicate a truck engine stalled after entering a mist, before killing

passengers who unfortunately disembarked from the vehicle to investigate. Two of the truck passengers survived however; by sitting on top of the vehicle cabin until the deadly mists retreated, leaving behind scenes of horror strewn across their landscape. Many theories, including state-sponsored terrorism were proposed, yet none could explain what had occurred that lone night.


Two years later, a similar but far more lethal incident occurred in the nearby landscape surrounding Lake Nyos (1986). The lake itself has long held a specific cultural association with the nuance of death in local mythology, with nearby inhabitants frequently attributing the varying small die-offs of wildlife [such as amphibians and birds] to ancestral spirits or other malevolent entities. While eyewitness accounts of a cause vary, what is known is that a rumbling noise was heard, followed by a strange white cloud which reportedly manifested on 21st August 1986 before closely hugging the ground as it hastily swept through a nearby valley at 20–50 kilometres per hour towards multiple low-lying villages. The eerie mist freely travelled downwards from the lake for over 23 kilometres, silently suffocating victims as they lay asleep in the countryside and villages of Nyos, Kam, Cha, and Subum. The lake waters turned red, aside from the copious patches of floating debris now scattered across its tranquil surface. By the time the ‘white cloud’ had apparently dissipated, it had killed 1,746 people and over 3,500 livestock living in the vicinity of the now silent lake, along with an incalculable loss of local wildlife which had also failed to respond in time to the lethal embrace of these creeping mists.

A local farmer from the village of Subum was the first person to raise the alarm after he awoke from a black out to find his deceased daughter laying still in her bed. Racing to the nearby town Wum to desperately check on his other family, he passed body after body scattered around intact homes and farms before reaching his relatives who knew nothing of what had transpired at Subum. Cameroon authorities initially hesitated to enter the area after encountering the first corpses, fearing whatever had killed these people may still be silently lurking in the locale, leaving a single, stubborn priest named Father TenHorn to push ahead alone looking for survivors of the strange catastrophe. Medical teams were again puzzled by the cause of this calamity, relinquishing ample space for conspiracy theories to fester (such as the Neutron Bomb claim) within public debates. In the aftermath of these mists, around 4,000 inhabitants also fled the area in fear of the next foreboding embrace, but many of these exposed individuals subsequently still developed respiratory problems, lesions, and paralysis as a result of their brief encounters with this imperceptible force. Communities built farther uphill seemed to survive unscathed, but those in the lower valley fell silent, along with birdsong and other wildlife; ‘On that day there were no flies on the dead’. Scientific investigations ensued for answers, while survivors suffered from social stigmas as a result of the inexplicable calamity that befell them.


The events at Lake Monoun and Lake Nyos, ‘Cameroon’s Killer Lakes’, are now understood to be the result of a phenomenon called ‘Limnic eruptions’; a very rare type of natural disaster that only occurs when dissolved carbon dioxide (CO2) suddenly erupts from crater lake waters, forming an invisible shroud capable of suffocating wildlife, livestock, and humans unfortunate enough to be engulfed in its latent advance. As the erupting cloud of CO2 is denser than the surrounding air, it closely clings to the landscape while displacing the breathable atmosphere; asphyxiating anything living caught in the cloud while leaving nearby vegetation virtually untouched. Sometimes [but not exclusively] referred to as a ‘Mazuka’ (Swahili; ‘evil winds’), such eruptions are exceedingly rare due to the relatively uncommon factors that cumulatively manifest limnically-active or ‘exploding’ lakes. For a lake to even undergo a limnic eruption, the various layers of water must rarely if ever mix (known as a meromictic lake); hydrological stratification which facilitates a lethal build up of CO2 or equivalent concentrations of methane gas in deeper waters from natural stores (like active volcanic activity, or through the mass decomposition of organic matter). Finally, the unstable deeper waters need a simple trigger event (such as a landslide or small earthquake, as is believed to be the case for the above incidents) to propel their dangerous gaseous payload outwards to much lethal effect.


Since these well documented catastrophes transpired, mechanical mitigation efforts at Lake Monoun and Lake Nyos have been employed to desperately syphon the buildup of dissolved gases from deeper waters in a controlled manner. Essentially, these intensive operations use a pipeline to induce a natural, restrained eruption which may prevent future limnic eruptions from occurring. By 2012, Lake Monoun was declared ‘safe’ through these degassing operations, while Lake Nyos experienced a 40% drop in CO2 water saturation after an additional two pipes were installed. Such imperative pilot schemes to degasify the lakes were only pioneered after these disasters had asphyxiated the landscape. Yet the long-term future of these vital operations is still tentative due to varying human-dependent factors such as funding obstacles, routine equipment maintenance, and also the persistent need to monitor these deeper waters (which continue to absorb asphyxiating gases) as the population surrounding the lake also continues to grow. One prevailing concern already identified is the higher saturation of CO2 settling in the atmosphere above the pipelines, causing further health problems for wildlife and maintenance engineers. Despite these two modern calamities, no effigies or memorial stand along the afflicted shores to signify the disasters, or the lurking threat that may, one day, breach the tranquil surfaces again.

Today, the global ratio between meromictic and holomictic (layer-mixing) lakes is around 1:1000, presenting an underexplored repository of potentially hazardous candidates which may be at risk of future limnic eruptions; lakes now warranting crucial scientific investigation, monitoring and long-term awareness for the possible dangers secreted beneath perceptibly silent waters, in addition to assessments of potential seismic triggering events. Such endeavours can be neatly classified as long-term experiments; enduring projects which are documented elsewhere in the After the Horizon initiative. Since the discovery of limnic eruptions as a rare class of natural phenomena, another such lake has now also been identified lying along the borders of the Democratic Republic of Congo and Rwanda. Lake Kivu is about 1,700 times the size of Lake Nyos, and supports a population of over two million human inhabitants, in addition to an incalculable quantity of livestock and wildlife that all depend upon its waters for survival. Inconveniently, there is already geological evidence that this Great African Lake possesses an extended history of local mass extinctions (presumably from limnic eruptions) transpiring nearly every 1,000 years. Troublingly, the waters presently store about 2.8 gigatons of CO2 (by comparison, the Lake Nyos limnic eruption was between 100,000–300,000 tons), in addition to hefty reserves of methane. Suspicions for Quilotoa Lake (Ecuador) have also been asserted, but there is insufficient evidence to determine whether this isolated water body has (or even will) experience an abrupt eruption.

Outside of the limnic phenomenon, other types of ‘Mazuka’ sites have also been identified such as Mammoth Mountain in California. This Mountain – an active lava dome and great ski resort – possesses a long history of spewing out toxic, asphyxiating volcanic emissions; gases which are accredited with causing the deaths of three members of a Ski Patrol in 2006, in addition to large patches of tree-kills across the entire region. Other sites that discharge lethal, noxious gases can be observed within the iconic Grotta del Cane (literally ‘Cave of Dogs’) situated in Naples Italy; a cavern which was renowned for ‘fainting dogs’. In traditional scientific literature, this cavern was used in textbooks as a basis to illustrate the toxicity and density of CO2 in comparison to the surrounding breathable atmosphere. It is also worth noting that the nearby Lake Agnano was once suspected to be another limnic lake. Such identified sites serve as frequent reminders for the unsuspecting yet foreboding (and often invisible) hazards beneath landscapes we inhabit.

To a related extent, there are a multitude of other actively monitored zones scattered across our planet that also pose an analogous threat to continuous [safe] human inhabitation, in addition to challenging the shifting generational tolerances for an ‘acceptable risk’ to future human life in unpredictable terrain. Perhaps the most salient example of this tentative inhabitation dilemma may be observed across vast stretches of the modern-day Japanese coastline. Given the unpredictable tectonic instability across the entire region, Japan encounters magnitudes of earthquakes on a frequent basis, with many of these tremors also giving rise to the devastating phenomena of tsunami; massive displaced volumes of water which destructively wash over the low-lying shoreline, sweeping away anything – or anyone – directly caught in its path. Moreover, the nation possesses an extensive historical record of encounters with severe tsunami and associated powerful earthquakes over centuries; accounts stretching as far back as the Hakuhō earthquake (29th November 684CE). Nowadays, the nation operates several interdependent tsunami monitoring and warning services across the region, in addition to incorporating tremor mitigation technologies to reinforce and protect structures along the coastline. Despite these adaptation and mitigation efforts, instances of severe ‘big one’ catastrophes breaching defences have already occurred. While not a phenomenon unique to this specific region, Japan's frequent encounters with tsunamis of varying magnitudes has subsequently shaped the social lifestyles, and psychologies of coastal communities over generations, for the benefit of their descendants.

It is within these coastal communities across Japan that we can see the identifiable beginnings of the principles we now deem as ‘risk assessments’. The Hakuhō earthquake and tsunami devastated the region and caused widespread flooding and loss of life, but the catastrophe also interlinked the delayed appearance of tsunami with experienced seismic activity. Moreover, the material record of these subsequent, recurring natural disasters (some of the notable severe instances occurring ‘once in a century’) filtered downwards through the living memory of the local populace. Survivors, having witnessed a devastating calamity, began to proactively evaluate the perennial risks posed by re-inhabiting lower coastal altitudes. These risk appraisals gave rise to the many mnemonic lithic devices, created over successive generations, that now dot the precarious coastline of some long-ravaged municipalities. Known collectively as ‘tsunami stones’, each relic consists of a monolith [or stele] inscribed with a forewarning for successive generations; ‘messages from the past’ cautioning those who are born long after a tsunami of the risks still posed by building homes at lower altitudes along the coastline. Many of these stones record the historical disasters and loss of life that have proliferated over the past two centuries alone; standing guard at each witnessed 'inundation line'. Some of these silent sentinels are even older, standing as a visible testament to the hardship of inhabiting these landscapes; tales captured within long-eroded inscriptions that symbolise the conventional wisdom common to all these relics – it is only a matter of when, not if, another tsunami sweeps beyond the fragile shoreline.

On 11th March 2011, a massive tsunami caused by the Tōhoku earthquake, swept inland through Aneyoshi bay in Honshū, washing away entire communities that had long called this coast home. The surge of seawater swept inland, wiping away several tsunami stones, before the water eventually ceased climbing uphill at 40 metres above standard sea level. The waters failed however to reach the ‘Aneyoshi tsunami stone’; erected by survivors of the 1896 and 1933 tsunamis, located at 55 metres altitude. The simple inscription roughly translates to; ‘Homes on higher ground will guarantee the comforts of descendants. This is a reminder for the horror of tsunamis, do not build homes below this point. We suffered tsunamis in 1896 and also in 1933, only 2 villagers in the former disaster and 4 in the latter survived. Keep on guard even as years pass by.’ Everyone who duly followed this warning inscription survived the 2011 Tōhoku tsunami. Here, we see a crucial link between maintaining the living memory of historical events, and the survival of generations now residing within that area. This example also raises the multitude of challenges we encounter in navigating the narrow swatches of time which govern scientific observation programs, safety policies, technological mitigation efforts, or indeed the successive lifespans of several familial generations and governments who reside across these precarious landscapes.

Considering the unpredictable and potential catastrophic risks associated with some rare types of natural phenomenon, and the variability associated with the frequencies of these potential disasters over intervals of deep time, it is necessary to preserve the memory of past incidents as a crucial part of any future risk-management or disaster prevention schemes. Coupled with this position, there is also a need to incorporate contemporary studies concerning other historical ‘places at risk’, and promote engagement with ongoing programs developed by scientists and governmental authorities to safeguard communities from future disasters. For instance, due to the identified phenomenon of ‘Lahar’ (debris-laden mudflows triggered by volcanic activity), volcanologists have established educational programs in vulnerable communities to communicate realistic hazard probabilities, scenarios for unfolding events, and strategies local leaders should adopt to safeguard these peoples from calamity. An example of lahar-prone zones in the world today include Mount Rainier in Washington, and another stratovolcano Mount Galunggung in Indonesia. Given the past geological evidence for lahar taking place centuries ago at the former mountain, much of these endeavours depend upon modelling disasters based on unobserved calamities that now threaten future prospects, as opposed to lived ‘survivor mentality’ (i.e., the nuance for tsunami stones). Meanwhile, Mount Galunggung experienced violent lahars during an 1822 eruption, causing 4,011 deaths. Perhaps, precaution guidance and informed scientific consensus may, one day, form the backbone of cultural traditions and local lore to forewarn posterity of the perennial hazards lurking on these slopes?

Nevertheless, maintaining awareness of past calamities, and supplying crucial documentation for posterity who continue to reside within these regions to commit informed decisions on behalf of their inherited homes, is perhaps the strongest form of risk appraisal we may furnish over unpredictable intervals of deep time. After all, the temporary measures we employ today to adapt, mitigate or prevent human tragedy are only truly beneficial if continually investigated, maintained, and properly funded across generational timescales – hurdles in ‘cathedral thinking’ which we often fail to incorporate. Tsunami stones standing guard over these precarious landscapes are only successful if their messages are known, understood, and contemplated by those now at risk of the same prospects. Securing this knowledge, while also supplementing local and institutional memory with retained documentation, is essential for any long-now observatory within planetary custodianship to successfully safeguard those who inhabit these foreboding landscapes across generations.


As part of the After the Horizon initiative, the Beyond the Earth foundation is committed to preserving vital records of these identified sites and associated repositories of peer-reviewed scientific literature (concerning both these instances and the phenomenon in general) for the benefit of posterity stewardship applications, while raising awareness of our need to, at least, safeguard some baseline knowledge of these enduring threats for the worlds of tomorrow. This goal adheres with our stated ethos to furnish posterity nations and research consortia with a factual opportunity to make informed decisions on behalf of their inherited life-world, through the conservation of significant data from sources that may not be readily available for further scientific scrutiny. This material archive will essentially function as a modern rendition of the ‘tsunami stone’; a catalogue of information which will hopefully forewarn our descendants of the often unseen and unpredictable threats that may be secreted beneath their very feet; natural forces which may resurface again given enough time. In addition, this repository will also hopefully aid with durational monitoring efforts as part of an established program in long-term experiment praxis; effectively, ensuring that slow science is logistically supported as it continues to document and record these unfolding phenomena for the benefit of future scientific scrutiny.

Notes to the catalogue: Given the many unpredictable properties and factors affecting sufficient investigations of hazardous phenomena beforehand, this broad index instead chronologically lists natural phenomena according to the date of first instance and subsequent recurrence(s), while loosley grouping these events into their defined disaster categories for easier comprehension (i.e., chronologies of lahars, lake tsunami etc.). The corresponding repository of peer-reviewed literature faithfully preserves the authoring agents' own observations, theories and, if available, original datasets to enable posterity to generate their own independent conclusions about this preserved information, while comparing this data against any contemporary chapters of human ‘calamity history’. Apart from the scientific value of these interred studies, this catalogue will hopefully provide a crucial resource for educators working in the realms of long-term experimentation, and provide a fruitful avenue for inspiring the development of additional deep time studies, that may only prove insightful for the generations which may learn from these investigations.

Page last updated: 05 May 2023

bottom of page