Based in Mahone Bay, nova scotia, agile libre is an independent publisher of technical books about the subsurface.  

Trust your observation skills

Becoming a scientist means that you must develop observation skills. With rare exceptions, observation skills are not taught explicitly either in secondary school or university. Most teachers spend little or no time on the importance of careful and methodical observation, on what that actually entails, on how existing knowledge and preconceived notions influence our observations, or on the trapdoor that lies between observation and interpretation. Students are generally only given instruction in how to record observations, usually in the context of a lab in which those recording methods are practised.

For example, in my first year, I had to draw the exquisite fossil specimens of our university’s historical geology collection in a sketchbook, to be handed in for a mark. Instruction: draw the specimens in such a way that all the different body parts are clearly visible and label them. I did realize that this lab helped me memorize the fossils for the inevitable lab test, but that drawing taught me observation was lost on me.

I wasn’t introduced to the idea that ‘what you don’t see, doesn’t exist’ until I was a graduate student. A charismatic professor casually introduced the notion as I was trying to record and understand bewildering sedimentary structures on a tidal flat. It was a revelation and I have been very conscious of it ever since. I think this awareness helped me improve my observation and recording skills in subsequent years.

At some point I had learned to observe the essential features and to record these using accepted codes. That meant that my observations could be checked and replicated, a crucial ingredient of scientific practice. I loved spending long hours in splendid isolation in a core lab, quietly working my way through metres and metres of core, slowly and carefully logging my way up through time. I learned to distance myself from interpreting, instead focusing on observing all features and letting the story tell itself. I loved it.

In the late 1980s I taught for a few years at the Institute of Earth Sciences of Utrecht University. Several faculty members had just participated in a scientific expedition in the seas around Indonesia. One of their sampling areas was Kau Bay, a dysoxic silled basin within the island of Halmahera, from where they had
retrieved a number of cores. The sediments of Kau Bay are mostly extremely fine grained. All cores had been logged lithologically, but the paleoceanographers and geochemists were in need of more detailed insight of the sedimentological processes that filled the basin throughout the Holocene.

We picked a crucial 10 metre core, stored in 1 metre sections. We subjected the entire core to X-ray radiography, analyzed grain sizes, and painstakingly compiled everything in a log, one for each metre section. Anticipation was high when we put the 10 individual sheets together.

And then… no story. The log didn’t tell us anything, there was no pattern, no trends. I refused to believe that this was the reality and asked about how the core sections were numbered. Sure enough, what I thought was the top metre was really the bottom one, the second from the top was the second from the bottom, etc. A misunderstanding — because I had made sure to ask beforehand what the labelling convention was.

We took the individual logs apart and taped them together in the proper order. The story instantly presented itself to us: three distinct facies were distinguished and they reflected clearly the different processes that were responsible for filling Kau Bay under changing climatic conditions. It was not only a huge relief, but also a terrific life lesson: I could trust my own observations.


Barmawidjaja, D, A de Jong, K van der Borg, W van der Kaars, W van der Linden, and W Zachariasse (1989). The timing of the postglacial marine invasion of Kau Bay, Halmahera, Indonesia. Radiocarbon 31, 948–956, DOI 10.1017/S0033822200012571.

Kosters, E, W Zachariasse, and D Barmawidjaja (1989). Sea-level controlled facies variability in deep-marine Holocene sediments of a dysoxic sea, Kau Bay, Indonesia. Annual Meeting British Sedimentological Research Group.


Why you care about Hashin–Shtrikman bounds

Where did the data come from?