PerspectiveGeochemistry and Geophysics

The Rise of Plants and Their Effect on Weathering and Atmospheric CO2

See allHide authors and affiliations

Science  25 Apr 1997:
Vol. 276, Issue 5312, pp. 544-546
DOI: 10.1126/science.276.5312.544

The spread of rooted vascular plants to upland areas during the Devonian Period (400 to 360 million years ago) most likely had an important effect on many Earth processes (1, 2). As reported on page 583 of this issue by Retallack (1), the first well-differentiated forest soils appeared in the Devonian. Along with this, there is additional evidence of a progressive increase in the size and depth of roots from the Silurian to the late Devonian (1, 3), with relatively deep rooting appearing as early as the late Early Devonian (390 million years ago) (4). Chemical differentiation, combined with deep rooting, suggests that the dissolution of bed rocks by weathering at this time was accelerated by the growth of plants. This enhanced chemical weathering would have resulted in enhanced removal of CO2 from the atmosphere, because the net effect of silicate mineral weathering is to convert soil carbon, derived ultimately from photosynthesis, into dissolved HCO3. A representative reaction is 2CO2 + 3H2O + CaAl2Si2O8 → Ca2+ + 2HCO3 + Al2Si2O5(OH)4 (1)

After formation by weathering, the dissolved HCO3 is carried to the ocean by rivers and, if accompanied by dissolved Ca2+ or Mg2+, the carbon is removed from the oceans as Ca-Mg carbonate minerals. In this way, if Ca and Mg silicates are involved in weathering, the overall process results in the removal of CO2 from the atmosphere.

The activities of vascular plants should result in enhanced weathering and, thus, enhanced removal of atmospheric CO2. There are several reasons for this enhancement: (i) Rootlets (plus symbiotic microflora) with high surface area secrete organic acids and chelates, which attack minerals in order to gain nutrients; (ii) organic litter decomposes to H2CO3 and organic acids, providing additional acid for weathering; (iii) on a regional scale, plants recirculate water by means of transpiration followed by rainfall and thereby increase water-mineral contact time; and (iv) plants anchor clay-rich soil, retarding erosion and allowing the retention of water and continued weathering of primary minerals between rainfall events.

It should be kept in mind that enhanced weathering by plants did not simply mean an increase in the global weathering rate. As is often misunderstood, the actual rate of CO2 removal from the atmosphere by weathering, on a multimillion-year time scale, is not an independent parameter and must be essentially equal to the supply of CO2 from degassing (5). Thus, an acceleration of weathering by the rise of vascular land plants in this case must have been matched by some decelerating process because there is no evidence of any equivalently large changes in the rate of degassing at this time. The decelerating process, I believe, was a drop in atmospheric CO2. Lower CO2, by means of the atmospheric greenhouse effect, brought about lower global temperatures and less river runoff, which had a decelerating effect on global weathering rates and helped balance the accelerating effect of the plants.

Estimates of Phanerozoic paleo-CO2 (see figure) are a quantitative demonstration of the importance of the rise of vascular plants on the continents. An outstanding feature is the large drop in CO2 during the Devonian (400 to 360 million years ago). Much of this drop likely was due to the spread of vascular plants to upland areas, where deep roots could bring about enhanced weathering and the development of differentiated forest soils (1). The figure summarizes all published data to date on CO2 derived from the study of the carbon isotopic composition of paleosols. The paleosol isotopic technique has been used by a number of independent investigators (68), and results are in rough semiquantitative agreement, as can be seen from the figure. The paleosol method is based on (i) the determination of the 13C/12C ratio of CaCO3 or the carbonate ion in goethite and (ii) a model for the mixing of soil CO2, derived from organic matter decomposition, with atmospheric CO2. It has been further tested by the analysis of the 13C/12C of soil organic matter included with calcium carbonate (7).

Atmospheric CO2 versus time for the Phanerozoic (past 550 million years). The parameter RCO2 is defined as the ratio of the mass of CO2 in the atmosphere at some time in the past to that at present (with a pre-industrial value of 300 parts per million). The heavier line joining small squares represents the best estimate from GEOCARB II modeling (10), updated to have the effect of land plants on weathering introduced 380 to 350 million years ago. The shaded area encloses the approximate range of error of the modeling based on sensitivity analysis (10). Vertical bars represent independent estimates of CO2 level based on the study of paleosols.

Also shown in the figure are the results of GEOCARB modeling (9, 10) for Phanerozoic carbon dioxide. The GEOCARB model is based on estimates of relative rates of removal of CO2 from the atmosphere over time by means of the burial of organic matter in sediments and the weathering of Ca and Mg silicate rocks followed by the deposition of the Ca and Mg as carbonates in marine sediments. This removal is essentially balanced by the release of CO2 back to the atmosphere by means of the weathering of ancient organic matter in rocks plus degassing resulting from the thermal decomposition of deeply buried sedimentary carbonates and organic matter. There is semiquantitative agreement between the GEOCARB modeling and paleosol results in the general trend of CO2 versus time. Besides the large drop during the Devonian (which continued into the Carboniferous), there are high values in the early Paleozoic (550 to 425 million years ago), moderately high values in the Mesozoic (240 to 100 million years ago), and a general decrease from the late Cretaceous up to the present (80 to 0 million years ago). GEOCARB sensitivity analysis (10) supports Retallack's contention (1) that most of the Devonian-Carboniferous drop in CO2 was the result of enhanced weathering and only in the later stages was it augmented by the enhanced burial of organic matter in swamps.

A recently developed independent method of paleo-CO2 estimation, involving the determination of plant stomatal index (11), is also in good agreement with the results shown in the figure. The stomatal index data are expressed in terms of the ratio of the stomatal index of nearest living equivalents divided by that for the fossil material. Results show a high index ratio to nearest living equivalents (high CO2) for the Early Devonian (400 to 390 million years ago), a much lower ratio (low CO2, close to that at present) for the late Carboniferous and early Permian (310 to 285 million years ago), and intermediate ratios (intermediate CO2) for the mid-Jurassic (150 million years ago).

The results shown in the figure support, on a long geologic time scale, the general correlation between climate and the level of atmospheric CO2 as manifested by the atmospheric greenhouse effect. The large Devonian drop in CO2 was followed by the Permo-Carboniferous glaciation, the most extensive and longest lived glaciation of the entire Phanerozoic. Also, glaciation during the late Cenozoic correlates with a lowering of atmospheric CO2, and the high Mesozoic and early Paleozoic CO2 levels are matched by climates that were distinctly warmer than that at present (12). [A short period of glaciation at the end of the Ordovician represents an exception, but this event has been explained by rather unique paleogeographic circumstances at this time (13).]

In the GEOCARB modeling, the calculated magnitude of the Devonian drop in CO2 is highly sensitive to what is assumed to have been the quantitative effect of vascular plants in the acceleration of silicate weathering. The theoretical curve in the figure is a best estimate based on studies (1416) of the role of plants in present-day weathering, where an approximately sevenfold acceleration has been determined (range 3- to 10-fold). This large measured effect by plants on weathering and CO2 level helps to explain the agreement between theory and the results of the paleosol and stomatal index methods for Siluro-Devonian time. However, much more quantitative data on the role of plants in weathering, as well as further study of paleosols and paleobotany, is needed to better understand how the rise of upland rooted plants and their subsequent evolution (such as the angiosperms) may have affected the evolution of Earth's surface as well as the CO2 content of the atmosphere.

References and Notes

View Abstract

Navigate This Article