Isotope code-breaker shows how groundwater travels through time and space

The weight of a water molecule’s parts can tell how it has travelled through a river basin. Other isotopes in a water sample can tell its age. A trip to South Africa’s Calitzdorp and Warmwaterberg hot springs reveals the secrets of long-haul groundwater travel.

Written by: Leonie Joubert
Photographs by: Sam Reinders

You first need to understand rocks if you want to understand groundwater, says hydrogeologist Dr Roger Diamond.

Dr Roger Diamond first trained in geology before specialising in the study of the relationship between water and rocks. His specific discipline uses isotope analysis to track how water flows through geological layers and how this recharges groundwater or enters river systems.

South Africa doesn’t have many hot springs, but where they do occur, they can tell a lot about what lies beneath, and how deep the water source is.

The water running from the mouth of the historic Warmwaterberg hot spring and spa is 43.5°C. This means it has percolated up from a depth of about 2 km to 2.5 km below the Earth’s surface. 

A small water sample, analysed using a mass spectrometer can help map the rainfall isotope patterns across a geographic area as vast as the Little Karoo basin and work out what the origins are of the water in the Olifants River.

1)     By comparing the hydrogen and oxygen isotopes from water samples taken in rain gauges on three mountain ranges, with those from a river water sample, Diamond was able to show that the origin of this spring’s water was from rain that fell on the top of the Swartberg Mountains, about 30km north-west of the Calitzdorp spring. The water then travels underneath the Little Karoo basin and eventually surfaces here.

To an untrained eye, the mouth of the Calitzdorp hot spring might look like a cauldron that’s been left to cook a bit too long. The water is a rusty opaque, and there are bits of scum and shredded vegetation on the surface which pool along the edge of the sunken reservoir wall that closes in the spring’s source. Every now and then a bubble plops up from the depths, issuing a breathy exhale of steam into the winter afternoon air.

Messy though this water may look — the site is in disarray after a recent flood — there’s a secret in its molecules, and hydrogeologist Dr Roger Diamond is the codebreaker who knows how to decipher it.

‘This water is at least a few hundred years old, maybe even a few thousand,’ he says. Scooping some to his lips, he describes it as tinny on the tongue.

‘You could be drinking water that fell as rain at the time of the pharaohs, or during the Dark Ages, or the Industrial Revolution.’

To understand how water flows through a system as vast as this, it’s useful to understand how old it is, or where it entered the system as rain. Dating of groundwater can be done several ways, including carbon-14, or other radioactive isotopes. These measurements are sometimes hard to perform in very pure waters such as those in the Cape mountains, and also require expensive analysis not available in South Africa.

It’s the precise weight of the hydrogen and oxygen isotopes in a water molecule that can help pin down where it fell in the water catchment, before trickling into the network of geological features that funnelled it deep into the groundwater system before releasing to this spring.

After studying water samples taken from this very spring in 2012, Diamond was able to show that the water had fallen on the Swartberg Mountains, whose peaks are visible in the far distance, before percolating down into a mysterious labyrinth of deep geological layers, and then bubbling back up just here.

There are many calculations that help unravel the secret.

First: the temperature of the spring water as it surfaces, which at this source is about 52°C. The groundwater here in the Little Karoo generally warms by 20°C for every kilometre of depth closer to the Earth’s hot mantle. By knowing this geological temperature gradient, and the surface water temperature, and factoring in that the water will cool a bit on its journey to the surface, Diamond can calculate that this spring gets its heat at a depth of between 2 km and 3km below the surface.

Finding the source: isotopes show where the rain fell

The next is to find the source of the rainfall that recharges the groundwater.

The Calitzdorp hot springs are on the banks of the Olifants River, tucked up against the base of the Gamkaberg Mountains and about 60 km inland of the coast as the crow flies. It’s 30-odd kilometres to the highest peaks of the Swartberg Mountains, a range north-west of here that puckers up to divide the Little and Great Karoo basins.

How does someone work out if the spring water is being recharged from rain that’s fallen in the nearby valley and flowed into the river from tributary streams, or if it’s from much further away, maybe even outside of the valley?

To answer this, Diamond took samples from rain gauges on all the nearby mountain peaks — the Outeniqua, the Gamkaberg, and the Swartberg — and compared their hydrogen and oxygen molecules with rainfall collected down in the valley.

‘The isotopes in the spring water were very different to those from local rainfall, but similar to those from the rainfall gauges on the mountaintops, but specifically those on the Swartberg,’ explains Diamond.

Bubble’s breath: what the carbon isotopes say

The bubbles erupting from the surface of the Calitzdorp hot spring are doing exactly what champagne does when the cork pops from the bottle: the pressure releases, and the dissolved gases escape in a dizzying fizz.

When the water is 3 km deep, the pressure will hold any carbon dioxide gas in a dissolved state, but as the water flows up towards the surface, the pressure releases it as bubbles of gas. Capturing the bubble’s breath from the spring’s gases is what Diamond did in 1996. By finding out the ratio of carbon-13 to carbon-12, he could show if that water had picked up its carbon from an organic source, such as a peat bog up on the surface, or from being in contact with rocks deep in the mantle, which release inorganic carbon dioxide into the water. 

Carbon dioxide that’s trickled through organic matter, like the kind of peat bog found on top of the mountains surrounding the Little Karoo, has a carbon isotope ratio rich in carbon-12, whereas CO2 that’s picked up its isotopes from deep in the mantle will have relatively more carbon-13.

Diamond’s analysis found the carbon isotope ratios to be typical of organic matter. Since the mountains surrounding the Little Karoo have peat bogs, but the hot, dry valleys don’t, this was further evidence of high mountain recharge, says Diamond.

How old is it, where did it fall, and why does this knowledge matter?

Dating water using the carbon-14 isotope can only measure the age of a substance that may be between a few thousand years old, up to 25,000 years old. Beyond that, the isotope has decayed and is no longer present in the water.

In a case like this, hydrogeologists need to lean on other pieces of evidence in order to calculate the water’s age.

‘We know the water coming up here is old because we’ve been able to work out the long flow path from the mountains to this spring. The spring flow rate and temperature are extremely stable, and that allows us to calculate that the water is at least hundreds of years old,’ says Diamond.

The work his team has done here shows that the hydrogen and oxygen isotopes in the water molecules fell on top of the Swartberg Mountains. Then, looking at the temperature of the water at the spring, and its flow rate — both of which remain constant over time — they could work out the approximate age of the water.  

This kind of information allows for better catchment management in protected mountain areas.

‘If we picked up any changes in water temperature or discharge rates in the Calitzdorp spring, for instance, would mean that changes are happening in very deep, regional flow systems,’ explains Diamond. ‘This would be an alarm bell for the overall aquifer.’

This isotopic method shows its usefulness, and how it can be applied to other systems in order to understand how and where a system is recharged, and how to manage and protect its springs and aquifers.

Studying the isotopic makeup of river water

The banks of the Olifants River took a bruising during heavy flooding in July 2024. Most of the damage caused during a flood event like this happens when water flow peaks, something which often lasts only a few minutes. Being able to anticipate precisely when a flow peak will sweep through a point on a river can help put effective early warning systems in place.

‘From a water management and agriculture perspective, it’s important to know how much water flows in a river system,’ Diamond says. ‘But in a storm event, you need to be able to anticipate the timing of peak flow, because that’s when infrastructure gets damaged.’

Traditionally, river flow is monitored with on-site instrumentation, like the kind that would have been in place in this derelict station at the river crossing at the Calitzdorp hot springs. This kind of information can help build up a historic record that allows the modelling of future flood events. Rain gauge data also helps understand the nature of a particular river system, and how it responds to storm events. More recent advances in satellite imagery adds another layer of data to understanding a river’s flow patterns.

Isotope analysis can add further detail and assist with anticipating peak events.  

By studying the isotopic makeup of the river water, it’s possible to map precisely where the water comes from during a storm event — from high up in the mountains, or nearby on the plains — and calculate what the lag time is between a storm releasing its water, and when that will arrive at any point on the river’s system.