 |
| Titel |
Historical land-use-induced evapotranspiration changes estimated from present-day observations and reconstructed land-cover maps |
| VerfasserIn |
J. P. Boisier, N. de Noblet-Ducoudré, P. Ciais |
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| ISSN |
1027-5606
|
| Digitales Dokument |
URL |
| Erschienen |
In: Hydrology and Earth System Sciences ; 18, no. 9 ; Nr. 18, no. 9 (2014-09-11), S.3571-3590 |
| Datensatznummer |
250120467
|
| Publikation (Nr.) |
 copernicus.org/hess-18-3571-2014.pdf |
|
|
|
|
|
| Zusammenfassung |
| Recent results from the LUCID model intercomparison project have revealed
large discrepancies in the simulated evapotranspiration (ET) response to the
historical land-use change. Distinct land-surface parameterizations are
behind those discrepancies, but understanding those differences rely on
evaluations using still very limited measurements. Model benchmarking studies
with observed ET are required in order to reduce the current uncertainties in
the impacts of land use in terrestrial water flows. Here we present new
estimates of historical land-use-induced ET changes based on three
observation-driven products of ET. These products are used to derive
empirical models of ET as a function of land-cover properties and
environmental variables. An ensemble of reconstructions of past ET changes
are derived with the same set of land-cover maps used in LUCID, with which we
obtain an average decrease in global terrestrial ET of
1260 ± 850 km3 yr−1 between the preindustrial period and the
present-day. This estimate is larger in magnitude than the mean ET change
simulated within LUCID with process-based models, and substantially weaker
than other estimates based on observations. Although decreases in annual ET
dominate in deforested regions, large summertime increases in ET are
diagnosed over areas of large cropland expansion. The multiple ET
reconstructions carried out here show a large spread that we attribute
principally to the different land-cover maps adopted and to the crops' ET
rates deduced from the various products assessed. We therefore conclude that
the current uncertainties of past ET changes could be reduced efficiently
with improved historical land-cover reconstructions and better estimates of
cropland ET. |
| |
|
| Teil von |
|
|
|
|
|
|
|
|