Orbital forcing of Early Jurassic wildfires

How does wildfire activity evolve in a steady state ‘background’ climate in deep time? We discuss the role of orbital forcing on wildfire activity in the Early Jurassic. Linking fire and hydrology, we found that fluctuations in fire activity occurred on precessional and eccentricity time scales.

Like Comment
Read the paper

In the palaeo-community, as researchers we tend to like to look at big ‘exciting’ events, that are often linked with catastrophic changes rather than times in Earth’s past where it appears nothing extraordinary happened. Yet it can be these seemingly mundane periods of time that can yield some very important and exciting information.

It all started with a preliminary study as part of Sarah’s PhD project, looking into changes in wildfire activity by analysing changes in charcoal abundance, a by-product of wildfire, during marked events in Earth’s past. One of these events occurred in the Jurassic period, about 183 million years ago, and is recorded in the Mochras borehole. 

The Mochras borehole is a unique sedimentary archive, which is twice to three times as thick when compared with other UK and European successions from the same age. To compare the amount of charcoal found around the major climatic event, Sarah sampled an interval of Mochras core that contained no event. Here, she took 18 samples covering 10 m of rock core, in a bid to look at a brief ‘background’ fire pattern of the Early Jurassic.

Interestingly, there appeared to be clear cyclic changes in the amount of charcoal found within the core sediments. These cyclic changes appeared to match with subtle changes in the rock core consisting of alternating light and dark sediment beds, and that also occurred over a very specific timescale of every 20,000 years. Excited by the unexpected results, we started to look further into these cyclic changes. 

Cyclic changes in Earth’s climate occur over tens to hundreds of thousands years. This natural ‘beat’ of Earth’s climate is governed by the spatial distribution of solar energy the Earth’s surface receives, determined by the ~20,000 yr cycle of precession (spin of the axis), ~40,000 yr cycle of obliquity (tilt of the axis) and the 100,000 and 405,000 yr cycles of eccentricity (shape of orbit). Collectively, these orbital cycles are called Milankovitch cycles. We can recognise these periodicities in the sedimentary deposits as far back as the banded iron formation (older than 1.8 billion years).

 Orbital processes influence global temperature, ice sheet growth, monsoonal strength and associated storm activity.  Excited that these natural orbital cycles clearly appeared to have a defined effect on wildfire activity some 188 million years ago, we thought it would be interesting to look deeper at just how fire activity fluctuates under ‘normal’ environmental and climatic circumstances to the natural beat’ of Earth’s climate on a medium-long time scale.

 So, what was just a small preliminary study soon took off and evolved into a PhD project of its own, taken on by lead author Teuntje Hollaar. Teuntje set about the monumental task of extending this record within the Mochras borehole, looking at a (geological) high resolution of every ~2000 years spanning over 350,000 years of fire activity. She extended the record from just 5 precession cycles to 18, identifying and counting over 46,000 individual macrocharcoal particles and creating projected counts of more than 15 million microcharcoal particles. It is safe to say, we were not deprived of charcoal. 

Large (~ 2 cm) long fossilized charcoal particle from the Upper Pliensbachian of the Mochras borehole, preserved on the plane of a core slab. Photo credit Teuntje Hollaar.
Large (~ 2 cm) long fossilized charcoal particle from the Upper Pliensbachian of the Mochras borehole,
preserved on the plane of a core slab. Photo credit Teuntje Hollaar.


Teuntje’s extended charcoal record showed the same metre-scale peaks as observed in the preliminary study, but also a major shift in charcoal abundance spanning the length of the study interval. Because of the high resolution (~10 samples per precession cycle), we were now able to run power spectral analysis on the charcoal record, which showed the presence of the ~20,000 yr precession cycle in the charcoal record at a 99% confidence interval.


 Power spectrum of the macrocharcoal record showing a strong peak at the precession frequency1
 Power spectrum of the macrocharcoal record showing a strong peak at the precession frequency1

Next to charcoal, our fire proxy, we also looked at hydrological changes. Hydrological changes can be detected in the clay mineral record, with transported clays forming in soils developed under a hydrolysing climate. In the present day, kaolinite is commonly found in tropical soils, with high humidity, high temperatures and a year-round wet climate. Smectite is found in more temperate regions, formed in soils in a warm climate with contrasting seasonal humidity2. This relation between smectite and kaolinite often holds up in the geological record and therefore forms a powerful tool in assessing the hydrological cycle in combination with charcoal abundance. 

Metre-scale and record long fluctuations in charcoal abundance and smectite (S/I) and kaolinite (K/I) proportions1

Using this relation between smectite and kaolinite, we found that there was a shift from a year-round wet climate (kaolinite) in the lower part of our study record, followed by a seasonal climate (smectite). We know from previously established age models, that the length of this record is approximately one long (405 kyr) eccentricity cycle, with the clay record indicating we have both a minimum and a maximum eccentricity phase in our interval. In the Jurassic, maximum eccentricity is often linked to a stronger monsoonal climate, which is in agreement with our high smectite recordings for part of the section. Simultaneously with high smectite, we have a high charcoal abundance. This indicates that charcoal, and inferred wildfire activity, was high during a period with a strong monsoonal climate. A strong monsoonal climate would have had a strong seasonal contrast, including a humid season and a dry season. A humid season would then have allowed for biomass to build up, and the following dry season would have lowered the moisture levels of the fuel (plants) leading to easy ignition.

The vegetation 188 million years ago was very different to today’s. Here, the landscape was dominated by a now extinct conifer, which produced the pollen grain Classopollis spp., with an understory of ferns. Other plants that would be part of the sub-canopy/understory were likely cycads, tree ferns, horsetails and mosses. From experimental burns, we know that conifers are extremely flammable, because of chemical and biological properties, which causes them to burn in a live state. Ferns on the other hand, rarely ignite when live, but are known to cause intense fires in a dry state. In a nearby location to our study site, evidence was found that the conifer group expanded into the hinterland on an orbital timescale3. This would have led to a higher fuel load and flammability for wildfire.

We found that fire activity is strongly influenced by orbital precession forcing, with highest fire activity at times of maximal forcing during the eccentricity maximum. These times of extreme seasonal contrast are also observed in our hydrological record too. During these periods, a humid season allows for fuel (vegetation) to build up, which is easily ignited during the subsequent dry season, independent from vegetation type. During the year-round wet season, linked to a minimum in the precessional forcing and eccentricity phase, high fuel (vegetation) moisture levels would have prevented rapid ignition and therefore frequent wildfires.

In parallel with this research, the overarching Early Jurassic Earth System and Timescale (JET) project, which aims to reconstruct a complete Early Jurassic astrochronologically tuned time scale, started drilling in the midst of the Covid-19 pandemic. The JET project (PI Stephen Hesselbo) cored at Prees, Shropshire, UK, and a stratigraphically complete and fossiliferous core was recovered. This will provide a great opportunity in the future to compare our results from the Mochras borehole. 


On the left: Drilling of Prees-2 in operation. On the right: Aragonitic ammonite from Prees-2. Photo credit Dominika Szucs.

For more check out our paper here and follow us on Twitter @HollaarTeuntje and @SarahBaker189.


  1. Hollaar, T.P., Baker, S.J., Hesselbo, S.P. et al. Wildfire activity enhanced during phases of maximum orbital eccentricity and precessional forcing in the Early Jurassic. Commun Earth Environ 2, 247 (2021).
  2. Chamley, H. Clay sedimentology. Springer, Berlin, Heidelberg, 623 (1989).
  3. Bonis, N. R., Ruhl, M., & Kürschner, W. M. Milankovitch-scale palynological turnover across the Triassic–Jurassic transition at St. Audrie's Bay, SW UK. Journal of the Geological Society, London, 167(5), 877-888 (2010).


Teuntje Hollaar

PhD, University of Exeter