- Climate Science Special Report: The Fourth National Climate Assessment: Volume I
- Featured Publications
- All Publications
finding 8.3 : key-finding-8-3
Future decreases in surface (top 10 cm) soil moisture from anthropogenic forcing over most of the United States are likely as the climate warms under the higher scenarios. (Medium confidence)
This finding is from chapter 8 of Climate Science Special Report: The Fourth National Climate Assessment: Volume I.
Process for developing key messages: In the northern United States, surface soil moisture (top 10 cm) is likely to decrease as evaporation outpaces increases in precipitation. In the Southwest, the combination of temperature increases and precipitation decreases causes surface soil moisture decreases to be very likely. In this region, decreases in soil moisture at the root depth are likely.
Description of evidence base: First principles establish that evaporation is at least linearly dependent on temperatures and accounts for much of the surface moisture decrease as temperature increases. Plant transpiration for many non-desert species controls plant temperature and responds to increased temperature by opening stomata to release more water vapor. This water comes from the soil at root depth as the plant exhausts its stored water supply (very high confidence). Furthermore, nearly all CMIP5 models exhibit U.S. surface soil moisture drying at the end of the century under the higher scenario (RCP8.5), and the multimodel average exhibits no significant annual soil moisture increases anywhere on the planet.b3bbc7b5-067e-4c23-8d9b-59faee21e58e
New information and remaining uncertainties: While both evaporation and transpiration changes are of the same sign as temperature increases, the relative importance of each as a function of depth is less well quantified. The amount of transpiration varies considerably among plant species, and these are treated with widely varying degrees of sophistication in the land surface components of contemporary climate models. Uncertainty in the sign of the anthropogenic change of root depth soil moisture is low in regions and seasons of projected precipitation decreases (Ch. 7: Precipitation Changes). There is moderate to high uncertainty in the magnitude of the change in soil moisture at all depths and all regions and seasons. This key finding is a “projection without attribution” statement as such a drying is not part of the observed record. Projections of summertime mean CONUS precipitation exhibit no significant change. However, recent summertime precipitation trends are positive, leading to reduced agricultural drought conditions overall.1d369e9a-53fa-4829-9bfe-9b380c83aae9 While statistically significant increases in precipitation have been identified over parts of the United States, these trends have not been clearly attributed to anthropogenic forcing (Ch. 7: Precipitation Change). Furthermore, North American summer temperature increases under the higher scenario (RCP8.5) at the end of the century are projected to be substantially more than the current observed (and modeled) temperature increase. Because of the response of evapotranspiration to temperature increases, the CMIP5 multimodel average projection is for drier surface soils even in those high latitude regions (Alaska and Canada) that are confidently projected to experience increases in precipitation. Hence, in the CONUS region, with little or no projected summertime changes in precipitation, we conclude that surface soil moisture will likely decrease.
Assessment of confidence based on evidence: CMIP5 and regional models support the surface soil moisture key finding. Confidence is assessed as “medium” as this key finding—despite the high level of agreement among model projections—because of difficulties in observing long-term changes in this metric, and because, at present, there is no published evidence of detectable long-term decreases in surface soil moisture across the United States.
ProvenanceThis finding was derived from figure -.2: Confidence / Likelihood
Alternatives : JSON YAML Turtle N-Triples JSON Triples RDF+XML RDF+JSON Graphviz SVG