Predictability of evapotranspiration patterns using remotely sensed vegetation dynamics during the North American Monsoon

Qiuhong Tang, Enrique Vivoni, Francisco MuñOz-Arriola, Dennis P. Lettenmaier

Research output: Contribution to journalArticle

40 Citations (Scopus)

Abstract

The links between vegetation, evapotranspiration (ET), and soil moisture (SM) are prominent in western Mexico-a region characterized by an abrupt increase in rainfall and ecosystem greenup during the North American monsoon (NAM). Most regional-scale land surface models use climatological vegetation and are therefore unable to capture fully the spatiotemporal changes in these linkages. Interannually varying and climatological leaf area index (LAI) were prescribed, both inferred from the space-borne Moderate Resolution Imaging Spectroradiometer (MODIS), as the source of vegetation parameter inputs to the Variable Infiltration Capacity (VIC) model applied over the NAM region for 2001-08. Results at two eddy covariance tower sites for three summer periods were compared and evaluated. Results show that both vegetation greening onset and dormancy dates vary substantially from year to year with a range of more than half a month. The model using climatological LAI tends to predict lower (higher) ET than the model using observed LAI when vegetation greening occurs earlier (later) than the mean greening date. These discrepancies were especially large during approximately two weeks at the beginning of the monsoon. The effect of LAI on ET estimates was about 10% in the Sierra Madre Occidental and 30% in the continental interior. VICestimated ET based on interannually varying LAI had high interannual variability at the greening onset and dormancy periods corresponding to the vegetation dynamics. The greening onset date was highly related to ET early in themonsoon season, indicating the potential usefulness of LAI anomalies for predicting early seasonET.

Original languageEnglish (US)
Pages (from-to)103-121
Number of pages19
JournalJournal of Hydrometeorology
Volume13
Issue number1
DOIs
StatePublished - Feb 2012

Fingerprint

vegetation dynamics
leaf area index
evapotranspiration
monsoon
vegetation
dormancy
continental interior
eddy covariance
MODIS
land surface
infiltration
soil moisture
anomaly
rainfall
ecosystem
summer

Keywords

  • Interannual variability
  • Land surface model
  • Operational forecasting
  • Remote sensing

ASJC Scopus subject areas

  • Atmospheric Science

Cite this

Predictability of evapotranspiration patterns using remotely sensed vegetation dynamics during the North American Monsoon. / Tang, Qiuhong; Vivoni, Enrique; MuñOz-Arriola, Francisco; Lettenmaier, Dennis P.

In: Journal of Hydrometeorology, Vol. 13, No. 1, 02.2012, p. 103-121.

Research output: Contribution to journalArticle

Tang, Qiuhong ; Vivoni, Enrique ; MuñOz-Arriola, Francisco ; Lettenmaier, Dennis P. / Predictability of evapotranspiration patterns using remotely sensed vegetation dynamics during the North American Monsoon. In: Journal of Hydrometeorology. 2012 ; Vol. 13, No. 1. pp. 103-121.
@article{a7ae4a66a4c04689b9bcde80eb260850,
title = "Predictability of evapotranspiration patterns using remotely sensed vegetation dynamics during the North American Monsoon",
abstract = "The links between vegetation, evapotranspiration (ET), and soil moisture (SM) are prominent in western Mexico-a region characterized by an abrupt increase in rainfall and ecosystem greenup during the North American monsoon (NAM). Most regional-scale land surface models use climatological vegetation and are therefore unable to capture fully the spatiotemporal changes in these linkages. Interannually varying and climatological leaf area index (LAI) were prescribed, both inferred from the space-borne Moderate Resolution Imaging Spectroradiometer (MODIS), as the source of vegetation parameter inputs to the Variable Infiltration Capacity (VIC) model applied over the NAM region for 2001-08. Results at two eddy covariance tower sites for three summer periods were compared and evaluated. Results show that both vegetation greening onset and dormancy dates vary substantially from year to year with a range of more than half a month. The model using climatological LAI tends to predict lower (higher) ET than the model using observed LAI when vegetation greening occurs earlier (later) than the mean greening date. These discrepancies were especially large during approximately two weeks at the beginning of the monsoon. The effect of LAI on ET estimates was about 10{\%} in the Sierra Madre Occidental and 30{\%} in the continental interior. VICestimated ET based on interannually varying LAI had high interannual variability at the greening onset and dormancy periods corresponding to the vegetation dynamics. The greening onset date was highly related to ET early in themonsoon season, indicating the potential usefulness of LAI anomalies for predicting early seasonET.",
keywords = "Interannual variability, Land surface model, Operational forecasting, Remote sensing",
author = "Qiuhong Tang and Enrique Vivoni and Francisco Mu{\~n}Oz-Arriola and Lettenmaier, {Dennis P.}",
year = "2012",
month = "2",
doi = "10.1175/JHM-D-11-032.1",
language = "English (US)",
volume = "13",
pages = "103--121",
journal = "Journal of Hydrometeorology",
issn = "1525-755X",
publisher = "American Meteorological Society",
number = "1",

}

TY - JOUR

T1 - Predictability of evapotranspiration patterns using remotely sensed vegetation dynamics during the North American Monsoon

AU - Tang, Qiuhong

AU - Vivoni, Enrique

AU - MuñOz-Arriola, Francisco

AU - Lettenmaier, Dennis P.

PY - 2012/2

Y1 - 2012/2

N2 - The links between vegetation, evapotranspiration (ET), and soil moisture (SM) are prominent in western Mexico-a region characterized by an abrupt increase in rainfall and ecosystem greenup during the North American monsoon (NAM). Most regional-scale land surface models use climatological vegetation and are therefore unable to capture fully the spatiotemporal changes in these linkages. Interannually varying and climatological leaf area index (LAI) were prescribed, both inferred from the space-borne Moderate Resolution Imaging Spectroradiometer (MODIS), as the source of vegetation parameter inputs to the Variable Infiltration Capacity (VIC) model applied over the NAM region for 2001-08. Results at two eddy covariance tower sites for three summer periods were compared and evaluated. Results show that both vegetation greening onset and dormancy dates vary substantially from year to year with a range of more than half a month. The model using climatological LAI tends to predict lower (higher) ET than the model using observed LAI when vegetation greening occurs earlier (later) than the mean greening date. These discrepancies were especially large during approximately two weeks at the beginning of the monsoon. The effect of LAI on ET estimates was about 10% in the Sierra Madre Occidental and 30% in the continental interior. VICestimated ET based on interannually varying LAI had high interannual variability at the greening onset and dormancy periods corresponding to the vegetation dynamics. The greening onset date was highly related to ET early in themonsoon season, indicating the potential usefulness of LAI anomalies for predicting early seasonET.

AB - The links between vegetation, evapotranspiration (ET), and soil moisture (SM) are prominent in western Mexico-a region characterized by an abrupt increase in rainfall and ecosystem greenup during the North American monsoon (NAM). Most regional-scale land surface models use climatological vegetation and are therefore unable to capture fully the spatiotemporal changes in these linkages. Interannually varying and climatological leaf area index (LAI) were prescribed, both inferred from the space-borne Moderate Resolution Imaging Spectroradiometer (MODIS), as the source of vegetation parameter inputs to the Variable Infiltration Capacity (VIC) model applied over the NAM region for 2001-08. Results at two eddy covariance tower sites for three summer periods were compared and evaluated. Results show that both vegetation greening onset and dormancy dates vary substantially from year to year with a range of more than half a month. The model using climatological LAI tends to predict lower (higher) ET than the model using observed LAI when vegetation greening occurs earlier (later) than the mean greening date. These discrepancies were especially large during approximately two weeks at the beginning of the monsoon. The effect of LAI on ET estimates was about 10% in the Sierra Madre Occidental and 30% in the continental interior. VICestimated ET based on interannually varying LAI had high interannual variability at the greening onset and dormancy periods corresponding to the vegetation dynamics. The greening onset date was highly related to ET early in themonsoon season, indicating the potential usefulness of LAI anomalies for predicting early seasonET.

KW - Interannual variability

KW - Land surface model

KW - Operational forecasting

KW - Remote sensing

UR - http://www.scopus.com/inward/record.url?scp=84858305494&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84858305494&partnerID=8YFLogxK

U2 - 10.1175/JHM-D-11-032.1

DO - 10.1175/JHM-D-11-032.1

M3 - Article

AN - SCOPUS:84858305494

VL - 13

SP - 103

EP - 121

JO - Journal of Hydrometeorology

JF - Journal of Hydrometeorology

SN - 1525-755X

IS - 1

ER -