FLOOD00113

• County: Statewide
• Title: Changes in the Hydrological Response of a Small Mediterranean Basin a Year After a Wildfire
• Date: 6/8/1992
• Prepared For: Journal of Hydrology
• Prepared By: Elsevier Science Publishers
• Floodplain - Doc Type: Educational/Technical/Reference Information

292

J. LAVABRE ET AL.

HYDROLOGICAL RESPONSE OF A MEOITERRANEAN BASIN

FLOOD ON DECEMBER THE 9th 1990

293

TABLE 8

Flood events in 1990

250

3S0

450

5S0

Date

Peak Dow
(m)s-l)

Total water yield
(mm)

Total rainfall
(mm)

o

I h 12h I h 12h
) October 1990 2.0 2.3 2.6 26 26
3 October 1990 6.2 1.9 2.2 15 23
25 November 1990 6.0 3.4 15.7 17 65
9 December 1990 8.0 11.8 32.0 28 78
CONCLUSIONS

4

~ 8

ci

.. 12
"
..
a:

16

We have used a 23 year period of data before the fire as a reference period
providing the normal hydrological behaviour of a small Mediterranean basin
(1.4km').
Some linear regressions and a lumped conceptual hydrological model were
calibrated in this reference period. They were used to assess what the response
of the basin would have been if the fire had not taken place. The comparison
between these results and the data observed have served to point out the
changes in hydrological behaviour of the basin due to the consequences of fire.
Two main aspects were checked: (1) the changes on the annual runoff yield
and (2) the changes in the flood regime.
(I) As for the annual runoff, we used five independent methods, with
different time steps, of assessing the increment produced after the fire. Table
10 shows these evaluations.
To summarise, the increment can be evaluated to be approximately
between 25 and 30% of supplementary runoff(aboutI30mm), which cannot
be explained either by hazard fluctuations or by climate fluctuations.
We suggest that this change is due to the consequences of the fire, and

9000

6750

~
;; 4500
g
,
a:

2250

o
2S0

350

450

550 650
Time (minutes)

Fig. 9. Rimbaud flood event on 9 December 1990 at a I min time step.

particularly to the destruction of the vegetation cover. Using the lumped
conceptual model, we tested if a simple change in the evapotranspiration
amounts can lead to a better reproduction of the observed monthly runoff.
The results showed that the reduction of losses is hydrologically equivalent to
applying the model in a new environment where the potential evapotranspira-
tion is reduced by 50% (without changing either the model conceptualisation
or the previously calibrated parameters).
It shows that the main hydrological change induced by the fire, at least at

TABLE 9

Flood events in the reference period (1967-1989)

Date

Peak flow
(mls-I)

I h
rainfall (mm)

12h
rainfall (mm)

September 1968
October 1973
April 1976
October 1987

i 13.7
9.1
4.8
4.8

158
150
t20
100

89
58
18
4t

OSO

750

8S0

750

850