DIGITALNA ARHIVA ŠUMARSKOG LISTA
prilagođeno pretraživanje po punom tekstu




ŠUMARSKI LIST 13/2005 str. 229     <-- 229 -->        PDF

P. Kovär: RISK ASSESSMENT OF EXTREME HYDROLOGICAL SITUATIONS CASE STUDY ... Šumarski list SUPLEMENT (2005). 2I9-22S
portancc in water resources management and flood
control (Falkenmark, 1999).


A case study based on water balance computation
has been carried out using data from two small catchments
Vseminka and Drevnice in Moravia. The aim of


EXPERIMENTALBoth Moravian experimental catchments covering
similar areas have suffered at least twice from floods
during recent years (1997, 2001). Fig. 1 gives a general
view of the situation of the catchments. The shape of
the Vseminka catchment is longitudinal, the upper sides
of the central valley are forested down from the
water divide, and land use on the other parts is diversified.
The Drevnice catchment is fan-shaped, highly forested
and only its lower part has a more structured
land use. The soil characteristics of the catchments are
given in Table 1 a, b. The soil groups have been selected
in accordance with methods widely used (U.S.
SCS, 1986). The land use and physiographic factors
are presented in Table 2.


The WBCM model that was implemented with the
aim of quantifying the soil moisture dynamics of the
experimental catchments is a lumped model with probability
parameter distribution over the area. It is based
on the integrated storage approach. Each storage element
represents the natural storage interception, soil
surface and active zone. In this version not all the unsaturated
and ground water zones were used, however only
a recharge to them was considered as an output from
the active zone in the form of deep infiltration. The mo-


RESULTS ANDThe model has been implemented annually since
1992 to compute the water balance components. Thus
the model parameters had already been calibrated. The
aim of this paper is to study two pre-flooding periods in
July 1997 with reference to soil moisture content in the
active zones of both experimental catchments. There
were two flood waves over the territory of Eastern Bohemia
and Moravia (4/7 to 10/7 and 18/7 to 27/7/ 1997).
The major data on these floods is shown in Table 3.
Then, the WBCM-5 model was implemented to compute
soil moisture content fluctuation in the form of
daily moisture deficit values (SMD), as well as the daily
antecedent precipitation index (API). The period of
June 1 to September 3, 1997, shows how dangerous
flooding situation appears when the SMD values drop
below 10 mm and even lower and the catchments have
no more retention capacity to store next rainfall. In particular,
the period 5/7 to 7/8/ 1997 on the Vseminka
and Drevnice was the most hazardous. The result was a
flood close to 100-year return period, which caused
major loss in human life and in property. The second


this study, in a broader sense, is to investigate the effect
of soil, physiographical factors and land use on water
retention capacities in the active zones of the catchments.


CATCHMENTS
del with daily step input/output rates simulates the following
processes (Kovar 1999, Kovar 2000):



potential evapotranspiration, interception and throughfall


direct runoff


active soil moisture zone dynamics
Note: Soil moisture content of the whole unsaturated
zone, ground water dynamics, base flow and total
flow were computed only for the process of parameter
calibration.
The modified Monteith-Penman method was used
for computing potential evapotranspiration. A linear
distribution of local interception capacities was in principle
used for actual interception and throughfall.
For quantifying direct runoff, the US Soil Conservation
Service, SCS method based on runoff curve number
(CN) assessment was used. The recharge of the
active zone and its depletion depend mostly on soil
parameters (field capacity FC, porosity POR, hydraulic
conductivities, KS), on soil moisture content and on
potential evapotranspiration conditions. For this computation,
the one-dimensional Richards equation is
used in the finite difference form (Kovar, 2001).


DISCUSSION
part of this research was to select two extreme periods:
dry and wet ones of about the same duration. As a dry
year with the least precipitation and the highest air
temperature in vegetation period (April - October) in
the last 20 years the year 1992 was selected in the data
record. Similarly, as a wet year was selected the year
1997 with the highest precipitation determined from
the data record from the same stations located appropriately
for both experimental catchments. However,
the next aim of this paper was to select two periods,
one typically dry and the other typically wet, both of
the duration about 6 to 7 weeks when there was either
no rain in dry season or substantial torrential storms
combined with regional rainfalls with their high periodicity
(p< 0.01, N > 100 years) in the wet season. Thus
these daily rain and corresponding daily runoff records
were analysed in both characteristic years 1992 and
1997 and for the Vseminka and Drevnice catchments
the following periods were adequately selected:


Dry period 1992: July 16 to August 31,1992 (47 days)
Wet period 1997: June 26 to August 14,1997 (50 days)