Climate Change Natural History

Evaporating History

Horton, T.W., W.F. Defliese, A.K. Tripati, C. Oze, Evaporation inducted 18O and 13C enrichment in lake systems: A global perspective on hydrologic balance effects, Quaternary science reviews (2015),


Population growth is a global challenge. It increases the demand for food, energy, and fresh water. Managing sustainable water resources on a global scale requires a concrete understanding of regional hydrological cycles.

Mono Lake Tufa, a carbonate deposit;  source:
Figure 1: Mono Lake Tufa, a carbonate deposit;

Stable isotopic records in lacustrine (lake) carbonates (figure 1) are powerful recorders of shifts in the hydrologic balance. They are ideal because their chemistry is directly related to the balance of the hydrologic cycle. For example, oxygen isotopes experience  kinetic fractionation during evaporation. Kinetic fractionation is the preferential reaction of one isotope over another. There are two isotopes for oxygen: one is heavier than the other and their ratio makes a unique signal that scientists can measure.   The signal changes if the ratio of the two isotopes changes.  When evaporation occurs the heavier isotopes does not move as quickly as the light one, so it is left behind.   As evaporation continues, heavy isotopes accumulate relative to the light ones, forcing a shift in the ratio; scientists measure the shifts and then interpret what caused them. In a similar fashion, the degassing of CO2 during evaporation results in systematic fractionation that enriches the bicarbonate ions in the heavier carbon isotope.

A challenge with interpreting hydrologic shifts solely from carbonate stable isotopes is separating the effects of evaporation from influences distinct from the hydrologic cycle. For instance, the co-variation of d13C and d18O in lake carbonates has been previously documented in the Dead Sea, epithermal systems, and in the lab, although few specifically address the effect of evaporation on both C and O isotopes. Therefore, there is a need to develop a multi-proxy approach. A group of researchers worked at the University of Canterbury in Christchurch, New Zealand to determine how a shift in the hydrologic balance influence the co-variation of carbonate isotope records. The findings will enable scientists to better constrain interpretations of hydrologic history made from carbonate records. They may also improve predictions of changes expected from future hydrologic shifts. Additionally, enhanced understanding of paleo-hydrologic balances is advantageous to interpretations of paleoclimate and paleotopography.


Researchers sought to develop a multi-proxy approach to interpret paleo-hydroclimatic conditions from ancient lake systems. They used laboratory experiments, analysis of multiple available data sets, and sample analysis to accomplish their goal.

The effect of evaporation on d18O and d13C in the water was analyzed in the laboratory. Natural water samples from a river, ground water, and coastal low land, were left to evaporate for six days at 21 degrees Celsius. The progression of d18Owater and d13CDIC was assessed daily. d18Owater was measured with a liquid water isotope analyzer. d13CDIC was measured by analyzing the CO2 gas released after acidification on a mass spectrometer

Quaternary lake carbonate dual isotope records from the Western U.S. were analyzed for co-variation that may be attributed to evaporation.   Fifty-seven records ranging more than 100 degrees in latitude and 4800 meters in altitude were used. Aridity (figure 3) values from ArcGIS, and modern meteoric, river, and lake hydrogen and oxygen isotope values were also used in analyses.

Global Climate map.  Relevant climate discussed in the Horton et al. are arid, semi-arid, and humid; source:
Global Climate map. Relevant climate discussed in the Horton et al. are arid, semi-arid, and humid; source:

Researchers also completed isotopic analysis on lake carbonate samples from the Quaternary (last 2.5 may) from Mono lake, CA, U.S., and middle Miocene (~23-5 mya) from CA, USA and Otago, NZ.   Measurements were made via acidification and mass spectrometry.


A major shift was observed in both d13C and d18O of greater than 10 permil with respect to source water (meteoric water) in more than 70% of the records analyzed. Interpretation of the Western U.S. lake records revealed that river data plot near the global meteoric water line (follow doi link to view article figure 2). There are some deviations attributed to dampening from the mixing of source waters, and/or evaporation along flow paths. Lake water data is the most unique from the global meteoric water line; especially those in arid regions, an affect attributed to evaporation.  Investigators were surprised by some of the results; it is against intuition that the d18O are heavier in some lakes of the Western U.S. because it is expected that the recharge of meteoric water would counteract the enrichment.   However, the effects of evaporation are so strong that the opposite is observed. Even lakes in humid regions may have enrichment up to 12 permil.

A similar shift in isotopic values was observed in laboratory experiments. The shifts co-vary linearly with slopes equal to .95 +/- .25. The observations support the idea that evaporation in lakes has a major co-varying effect on d18O and d13C due to the enrichment of heavy oxygen and carbon isotopes, respectively.


Models were compared to emphasize the importance of using multiple proxies. In one model the focus was on d18O in water expected from summer rain in an average summer.  Model output was compared  to the d18O from the 57 records. The model assumes the local meteoric water is unaltered and the values output were lighter (less enriched in heavy isotopes) than the data from the records. The model is limited by assumptions; for example, had winter temperatures or annual precipitation been used the d18O modeled would have been more positive and more negative, respectively. The second model used paired carbonate isotope records with modern lake and source water oxygen and carbon isotope data; in this case the model was able to output the d13C and d18O expected if carbonate formed in equilibrium with source and lake water.   The point made is that multi-proxies will yield better-constrained interpretations.


The scientists were successful in reaffirming that evaporation impacts d18O and d13C recorded in lake carbonates. They demonstrated that using a single element analysis limits the ability to interpret evaporative effects because they are so sensitive to other environmental factors. When coupled with additional proxies, however, it is possible to more accurately reconstruct ancient hydrologic systems, paleoclimate, and paleotopography.

The consequences of evaporation lay in the chemistry. For instance, when water is removed via evaporation the other constituents become more concentrated; the change may result in precipitation of mineral phases. Also, 13C enrichment of dissolved inorganic carbon (DIC) can increase alkalinity. The effects of changes in the hydrologic cycles can be extreme and it is important to understand them so that precautionary measures to minimize their impact can be made.


Leave a Reply

Your email address will not be published.