Hydrological drought sensitivity to land use changes in a Northwestern Mexico River Basin

1

• Hydrological drought sensitivity to land use
  changes in a Northwestern Mexico River Basin
• Francisco Munoz-Arriola1 Chunmei Zhu1, Andrea
  Ray2 and Dennis P. Lettenmaier2
• Department of Civil and Environmental
  Engineering, Box 352700, University of
  Washington, Seattle, WA 98195
• NOAA Earth System Research Laboratory, 325
  Broadway, Boulder, CA 80303-3328
• Regional Climate Forum for Northwest Mexico
• and the Southwestern U. S. , Ensenada, Baja
  California, México April 1011, 2008

Evaluation of Simulated Streamflow

• VIC Features
• Macro-scale and semi-distributed model
• Subgrid representation of the spatial variability
  in
• Land surface vegetation classes
• Soil moisture storage capacity as a spatial
  probability function (Xinanjiang Model)
• Drainage from the lower soil moisture zone
  (baseflow) as a non linear recession (ARNO model)
• Evapotranspiration based on the Penman-Moneith
  equation (Mid- and bottom soil layers)
• The routed runoff transport is linear and time
  invariant
• Spatial Resolution
• 1/8 o (Zhu and Lettenmaier, 2006)
• Temporal Resolution
• 3-hourly
• 1949 to 1999
• Temporal Resolution
• Drought analysis uses the percentile method with
  the lower 20 threshold (Andreadis et al 2006)

Drought Analysis
ABSTRACT Land-use changes can strongly affect
streamflow generation. In most cases,
deforestation and agricultural intensification
increase streamflow while afforestation reduces
it. In semi-arid basins, such as the Yaqui River
basin (YRB) in Northwestern Mexico, changes in
streamflow generation may impact the
sustainability of the region's agricultural
practices. Water resources in the YRB are
influenced by different climate-phenomena, such
as the North American Monsoon, El Niño Southern
Oscillation (ENSO), and the Pacific Decadal
Oscillation (PDO), and by anthropogenic
activities. Here we evaluate the sensitivity of
drought events in the YRB to land use changes
under different climatological conditions.
Periods were identified where the ENSO and the
PDO influence the temporal variability of
streamflow from 1949 to 1999, using wavelet
analysis. The drought events, which were
influenced by these oscillations, were identified
using runoff-percentile anomalies. Increments of
the agricultural and afforestation practices were
implemented as boundary conditions, and used by
the Variable Infiltration Capacity model (VIC) to
simulate the hydrological surface components of
the YRB. Droughts during the non-monsoon months
of La Niña and the warm phase of the PDO years
were sensitive to increase in crop land-use in
the northernmost part of the YRB. The rest of the
basin observed an increase in runoff, which
reduced the drought occurrence. Drought
sensitivity to afforestation was higher during
the monsoon months. In neutral years followed by
El Niño and in years during the cold phase of the
PDO, drought sensitivity to afforestation was the
highest during the monsoon months.

• An increment of 3 in the crop surface produces
• Streamflow decrease in Angostura
• Paso Nacori and Oviachi streamflow increases
• Changes in streamflow are associate to
• Deficit in soil moisture and the consequent
  inhibition of evapotranspiration previous to the
  monsoon and the first stage of monsoon
  (June-July)
• Soil moisture increment after the monsoon onset
  favors evapotranspiration

• Under current conditions short draught events (5
  days) dominate in all sub-basins
• Paso Nacori experiment the highest number of
  draught events
• The largest drought event occurred in Angostura
  during 1999 (79 days)

• Streamflow deficit produced by the monsoon
  droughts is higher than those in the pre-monsoon
  and post monsoon months.
• Post-monsoon droughts are consistently smaller in
  number of days and events
• The strongest drought events and occurred at
  Angostura during the pre-monsoon months
• The largest drought event occurred in Angostura
  during 1999 (79 days)

Stream Flow (106 cfs)
Angostura
Oviachi
Paso Nacori
Pre-Monsoon
Monsoon
Post-Monsoon
STATEMENT OF PURPOSE To assess the streamflow
drought sensitivity to land use changes under
different climatic regimes in the Yaqui river
basin
Simulated (naturalized
Study Domain
Observed
Three sub basins of the Yaqui River basin were
evaluated, Angostura and Paso Nacori, considered
natural streamflows and Oviachi a streamflow
diverted for water storage (hydropower generation
and irrigation).

• Calibration Munoz-Arriola et al. (to be
  submitted)
• Climate conditions (see below)
• El Niño Souther Oscillation Index (SOI)
• Pacific Decadal Oscillation (PDO)
• Identification and Evaluation
• Wavelet Analysis
• Composites (from Larkin et al 2002 and Mantua et
  al. 1997)

• Angostura Case
• Pre Monsoon
• Drought days were reduced during EN and LN years
• During N years draught days increases
• Changes in the number of draught days occurred
  during the warm phase of PDO
• Monsoon
• During all climate conditions the number of
  drought days increases
• Increases in the number of drought days were
  higher during N, followed by LN, and EN with the
  smallest changes
• During the Warm PDO occurred 92 of the changes
• Non-Monsoon
• All years experimentd a reduction in the number
  of drought days
• Warm PDO experimented a reduction in the number
  of days of drought while during cold PDO occurred
  the opposite

Bavispe River
Angostura
Paso Nacori
NAMS
Sonoran Desert
Moctezuma River
Gulf of Mexico
Sierra Mafre Occidental
Gulf of California
Aros-Papigochi River
Oviachi
Pacific Ocean
Yaqui River
Yaqui Valley
Gulf of California
Warm PDO (1947-1976) Cold PDO
(1977-1995) Pre-Monsoon (January, February,
March, April, May) Monsoon (June, July, August,
and September) Post-Monsoon (Octuber, November,
and December)

• CONCLUDING REMARKS
• Intensification in agriculture is reflected in
  changes in the number of drought days.
  Munoz-Arriola (2007) observed a reduction in the
  streamflow produced by increments in the crop
  surface at Angostura (opposite to what occurred
  at Oviachi and Paso Nacori). This is reflected in
  the increment of drought days during the monsoon
  months. During non-monsoon months the streamflow
  generation is influenced by events such as snow
  melting (may be important in Angostura during the
  spring months) and low evapotranspiration
  producing a reduction in the drought events and
  its duration. This shows the importance of land
  surface processes such ass soil moisture dynamics
  and consequently the inhibition of
  evapotranspiration due to availability of water
  in the first soil layers.

Small- mid- (El Niño EN, La Niña LN, and
neutral N years), and large-scale clime
conditions (warm and cold PDO)were identified
using the squared wavelet coherency (Grinsted et
al. 2004) and information in literature, observed
in the climatological composites of simulated
streamflow (Larkin et al 2004 and Mantua et al
1997). Monsoon months were assumed June, July,
August, and September, the rest of the months are
non-monsoon months.
Andreadis, K.M. and D.P. Lettenmaier, 2006
Trends in 20th century drought over the
continental United States, Geophys. Res. Lett. ,
33, L10403, doi10.1029/2006GL025711 Larkin, N.
K., and D. E. Harrison (2002), ENSO warm (El
Nino) and cold (La Nina) event life cycles Ocean
surface anomaly patterns, their symmetries,
asymmetries, and implications, J. Clim., 15,
1118-1140. MunozArriola, F., D. Lettenmaier, C.
Zhu..and R. Avissar. Hydrological Response to
Land Use Change in the Yaqui River Basin
(submitted, Water Resources Research)
Munoz-Arriola, F. (2007). Hydrological Response
to Precipitation Discrepancy and Land Use Changes
in the Yaqui River Basin. Civil and Environmental
Engineering Department, Duke University. Ph. D.
degree inCivil and Environmental Engineering. Zhu
C.M. and D.P. Lettenmaier, 2007 Long-term
climate and derived surface hydrology and energy
flux data for Mexico,1925-2004, Journal of
Climate, 20, 1936-1946.