Science rarely proceeds in leaps and bounds. It is better characterised by the gradual accretion of knowledge. As Issac Newton remarked to rival Robert Hooke in 1676: “If I have seen a little further it is by standing on the shoulders of Giants.”

It is appropriate then, on this auspicious day, to illustrate how scientists manage to stand on one another’s shoulders, and how our science evolves. Auspicious, because it is the 150th anniversary of the publication of Charles Darwin’s ‘On the Origin of Species’, and on this very day I am giving a talk at the joint NZ Hydrological and Freshwater Science Society conference entitled ‘A phylogeny of evapotranspiration models’.

Science is to a great extent about models. Not necessarily of the computer variety, but more generally of the narrative. That is, they are explanations of observed phenomena.

One hydrological phenomenon that is of great importance to many scientific disciplines is evapotranspiration. Due to its cumbersome size, we generally shorten it to ET – not to be confused with extra-terrestrials. ET is the sum of water that evaporates from the Earth’s land and water surfaces or transpires from plant’s leaves. Globally, roughly 60% of all rainfall that falls on land returns to the atmosphere as ET. Regionally and locally, this can vary from nearly 0 to 100%.

Models of ET have been around since at least the Classical Period. Greek philosophers attributed the evaporation of water from the seas to the sun and wind. Very little changed for millennia, until a veritable Cambrian explosion of ideas in the 20th century. While the evolution of ET models in recent history is a continuous but slightly bumpy road, several milestones can be identified.

In the 1920s, two Germans wrote two papers suggesting that the movement of water from soil to atmosphere via plants be described by Ohm’s law: water moves from high concentration to low, with the plant providing resistance. This was popularised in 1948 by van der Honert, who notably formulated it in mathematical terms. Over the decades, the idea of a plant as a resistance to water movement has developed from a single resistor to many resistors in series and parallel, representing roots, xylem, leaves and so on.

In the meantime, and very much separated from the Ohm’s analogy advances, a Russian climatologist by the name of Mikhail Budyko described large swaths of the landscape as simple buckets. This was in 1956. These buckets stored water up to a maximum amount, and below some specified threshold, evaporation declined linearly with the amount of water stored.

Advance to 1965 and a meteorologist in Britain, Howard Penman, advanced his theory of evaporation based on radiation balance. In another nod to Ohm’s law, he realised that both the plant and the air around the plant provide resistance to water movement.

An early sign of hybridisation took place in 1969 when Japanese climatologist Syukuro Manabe combined Penman’s basic theory with Budyko’s bucket to create a global climate model (GCM). This sparked a new branch on the evolutionary tree for the climatology community that led to the models used in climate change research. In 1978, James Deardorff notably treated the ground surface differently from the plant canopy, and in 1991 James Collatz and company drew on research by plant physiologists to account for carbon assimilation. After all, plants essentially only lose water in the act of absorbing carbon.

In 1970, however, another branch had started growing. Two engineers at Stanford University, by the names of Molz and Remson, took what was previously a description of water flow in soil (called Richards equation) and added plant water uptake. They went on to suggest that water is not taken up uniformly down the soil depth, but in some way that reflects root activity – more uptake near the top where roots are most abundant. They also provided a framework, also still in use today, linking water uptake to the amount of water present. How water uptake activity varies with depth is still unresolved. I even added my two cents several years back while at MIT, but research rightly continues.

A Dutch soil scientist by the name of Reinder Feddes was also very instrumental in this story, drawing from the work of Molz and Remson, and explicitly discarding the detailed Ohm’s-based approaches. In the 1970s he considered how water availability affected plant water uptake: too little water, and stomata close or xylem stop working, and plant water uptake declines; too much water, and oxygen is depleted, and plant water uptake declines. Feddes’ framework is still very much in use today, though again there is not yet a consensus.

By and large, then, these are the milestones of the evolution of ET models in the 20th century. They illustrate how ideas grow from one another, inspired by new observations or new questions. They show how ideas propagate from one discipline to another, slowly building a more robust depiction of the world. They are still only models of reality, and as such they are all somewhat wrong, but they are still useful and also somewhat right. As more observations become available, new techniques, and new ideas, models of ET will continue to evolve. Some branches will die off, others will fuse together, and yet more will sprout. So goes the evolution of scientific thought by natural selection.