Prediction of seasonal runoff in ungauged basins - Runoff Prediction in Ungauged Basins: Synthesis across Processes, Places and Scales - page 108

Figure 6.6. Seasonal water balance
component of the Havel River
(northern Germany). Adapted from
Wundt ( 1953 ).

100

Precipitation

Evaporation

50

Runof f

0

Depletion in spring and
summer

Soil water
storage

-50

J F MA M J J A S O N D

peaks during spring but is sustained through summer by
loss of water from the soil storage, and consequently
exhibits much less relative variation than the climatic
drivers (evaporation and precipitation).
The many examples presented above (see Figures 6.4 ,
6.5 and 6.6 ) show that the effects of storage in snow
versus storage in soils and groundwater can produce
very different results in terms of the seasonality of run-
off. In essence, the effect of storage on seasonality
of runoff is largely a product of the dynamics of storage.
When storage dynamics are out-of-phase with the
precipitation

demonstrate that a spring increase in potential evaporation
(primarily driven by energy availability) preceded the
observed increase in transpiration by approximately one
month. The difference appears to be primarily associated
with vegetation activity. In deciduous sites, such as the
Morgan Monroe State Forest shown in Figure 6.7 ,
increases in evaporation (E) lagged behind increases in
leaf area, suggesting that full canopy activity was not
achieved during the early part of the growing season,
despite increasing leaf area. Temperature limitations were
correlated with the lag, suggesting that reduced stomatal
functioning due to low temperatures may have been
responsible. Applying a correction for soil temperature
(Jolly et al., 2005 ) allowed estimation of the correct sea-
sonality in evaporation (Thompson et al., 2011a ).
The consequences of variability in vegetation activity,
beyond measurements of LAI alone, appear to be relevant
in attempts to model the properties of the seasonal flow
regime curve. Seasonality in the flow regime may occur
even when seasonality in precipitation is weak (e.g., in
south-eastern USA). Temperature corrections that mimic
phenological variation are needed in order to reproduce
the seasonality in the flow regime curves in Appalachia
and the eastern USA (Ye et al., 2012 ), suggesting that
accounting for vegetation function while estimating evap-
oration is an outstanding challenge for seasonal runoff
prediction.
The role of vegetation activity in changing runoff behav-
iour can be identified by a direct analysis of runoff dynamics
during the spring period as vegetation activity increases.
Czikowsky and Fitzjarrald ( 2004 ) explored whether

evaporation (P-E) seasonality (as in the
Havel River), storage will tend to buffer the seasonal
variation imposed by the relative availability of water
and energy. Conversely, if storage dynamics are in phase
with P-E dynamics, as in the case of snow accumulation
and melt, storage acts to exaggerate the underlying
seasonality.

-

Land surface processes and vegetation phenology
Vegetation dynamics provide a throttle on transpiration
losses from catchments. When vegetation is actively tran-
spiring, catchment storage tends to be depleted, runoff
declines, and larger precipitation events are needed to
generate runoff. Vegetation can thus play a major role in
regional runoff seasonality, especially when vegetation
activity is itself seasonal or variable. An example of such
rapid vegetation change, in terms of both leaf area index
(LAI) and photosynthetic capacity, is shown by work on
the deciduous forests in north-eastern USA (Thompson
et al., 2011a ). Studies of eddy-covariance records