During the last decades there is a growing concern about climate change and its devastating effects on water availability. The aim of this study was to assess surface runoff changes in the mountainous catchment of the Klinovitikos torrent (Central Greece) under climate change. To this end, precipitation and temperature data were derived from a high-resolution (25×25 km) RegCM3 regional climate model for the baseline period 1974-2000 and future period 2074-2100. Subsequently, Thornthwaite and Mather water balance model was applied to quantify the effects of precipitation and temperature changes on surface runoff. The results showed a decrease in surface runoff until the end of the 21st century, approximately -35.5%, -14.4%, -49.8% and -43.7% for winter, autumn, spring, and summer, respectively. It is also noted that the reduction will be greater in the dry season. To this end, adaption of mountainous catchment management to climate change is crucial to avoid water scarcity.
The existence of life depends on the availability of freshwater on earth's surface.
During the last decades, the world's rapid population growth, urbanization and
industrialization have increased the demand for water worldwide
The climate regime in Mediterranean regions favors the development of drought
phenomena due to the pre-long dry period and the uneven distribution of rainfall
The aforementioned changes will highly influence catchments hydrologic response. Therefore, water resources management has become a crucial challenge needs to be addressed. To this end, stakeholders should also quantify surface runoff under future climate conditions for infrastructure projects scheduling.
General Climate Models (GCMs) are very helpful in understanding the future
evolution of the global climate, but have a spatial resolution (100–300 km) too coarse
for assessing regional or local changes
During the last years, several studies assessed the effect of climate change on
streamflow using high resolution RCMs
The main object of the current research is to quantify the effect of climate change on surface runoff in a mountainous catchment of Central Greece using the Thornthwaite-Mather water balance model and climate simulation of RegCM3 regional climate model. 2
The study was conducted in the mountainous catchment of Klinovitikos torrent (Fig.1). It is located in Thessaly Regional Unit (Central Greece) over the mountain range of Pindus and is a tributary of Pinios River. It covers an area of 171.1 km2 and the relief is rather intense. The mean elevation is 1112 m.a.sl (maximum 2204 m.a.s.l and minimum 320 m.a.s.l) whereas mean catchment slope is 48.4% and main stream slope 6.5%. Moreover, the region has great environmental importance as belongs to the European nature conservation network Natura 2000 according to the criteria of Directive 92/43/EEC and specifically includes the Site of Community Importance (SCI) with code GR1440002.
In the frame of the EU funding project ENSEMBLE
(http://ensembleseu.metoffice.com/) a set of multi-model RCM simulations to characterize climate
change in Europe with high spatial resolution (25Î25 Km) were produced. Data
from the RCMs used as input in hydrological models, so their ability to represent
climate conditions should be examined prior to their use in impact assessment studies
To this end, a recent study by Stefanidis (2018) evaluates the ability of seven
RCMs of ENSEMBLE project, under A1B SRES emission scenario, to represent
temperature and precipitation conditions over the mountainous Central Pindus (study
area) for the baseline period 1974-2000. The results concluded that best simulations
were made by the International Centre for Theoretical Physics Regional Climate
Model (RegCM3)
In order to estimate surface runoff, the Thornthwaite and Mather (1957) water balance model was chosen. This model uses as parameter the maximum soil-moisture holding capacity (K) and having as input precipitation and temperature data, given as result the surface runoff and the actual evapotranspiration. The mathematical description of the model given by Equation 1: where ΔS is change in soil moisture, P is the annual rainfall, Ε is annual actual evapotranspiration, and Q is water surplus (percolation και surface runoff). The corresponding equations of the monthly water balance described below: § §
If Pn>ETp If Pn< ETp
Sn = min(Sn-1 + P - ETp ), max(K )
n Qn = max(Sn-1 + Pn - ETp - K ), min(0)
P - ETP ) Sn = Sn-1 * exp( n
K
Qn = 0 ETa = (Sn-1 - Sn ) + Pn - Qn = Pn - DSn - QAn - Dn
DS = Sn - Sn-1
QAn = a(Qn + QAn-1)
For the month n, Pn is mean monthly rainfall, ΕΤp is mean monthly potential evapotranspiration, ΕΤa is mean monthly actual evapotranspiration, Sn: maximum soil retention, Qn: water surplus, QAn: surface runoff, Dn is infiltration and α: runoff coefficient.
In order to calculate the amount of precipitation that evaporates or transpires back to the atmosphere the generally accepted method proposed by Thornthwaite (1948) were used. The monthly potential evapotranspiration (PET) is estimated from average monthly temperature (T).
ETp = 16(10tn )a * Ld
J a = 0.0016J + 0.5
12 t J = å ( n )a n=1 5 (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) where, ETP is the potential evapotranspiration, tn is mean monthly temperature and Ld is coefficient depending on the latitude (in decimal degrees) of the location of the study area.
Additionally, to take into account the snow accumulation a revised version of the model was used as proposed by Dingman (2002). In this method, mean monthly precipitation (Pm) divided into rainfall (RAIN) and snowfall (SNOW) according the snow melt factor Fm based on mean monthly temperatures (Τ). The equations calculate these factors were given below:
Therefore:
T £ 0O C : Fm = 0 0o C < T < 6O C : Fm = 0.167 * T
T ³ 6O C : Fm = 1 RAIN = Fm * Pm SNOW = (1 - Fm ) * Pm (12) (13) (14) (15) (16) (17) (18) (19) (20) The monthly melting amount of snow (MELT) was given by equation: MELT = F * (SNOW+ PACKm-1)
m where PACKm, the snow accumulation for each month m, was estimated from the above equation:
PACKm = (1 - Fm )2 * P+ (1 - Fm ) * PACK m-1
Monthly precipitation considering both rainfalls and snowfall was estimated as described below:
Subsequently, the known equations were applied if Pn> ETP and if Pn< ETP.
Moreover, the potential maximum soil moisture storage after runoff begins (K), related to curve number and was computed by Equestion 20:
Pn = RAIN + MELT
K = 25400
CN - 254
Curve number (CN) is an index that represents the combination of a hydrologic
soil group and land use and management class and has a range of 30 to 100. SCS
developed a soil classification system that consists of four groups, which are
identified by the letters A, B, C, and D according to their infiltration, retention and
evaporation capacity. There are tables from USDA Soil Conservation Service that
indicate CN for characteristic land cover descriptions and a hydrologic soil
group
Finally, infiltration process was assessed using empirical coefficients according
the literature
Based on the values of the above mention table (Table 1) and the equation 21, a mean infiltration coefficient was estimated for the study catchment.
w = w1 * F1 + w2 * F2 + ...wn * Fn
FSum (21) where w1…wn is the infiltration coefficient for each petrographic formation, F1…Fn is the area of each petrographic formation and Fsum is the total catchment area.
Regarding the data of RegCM3 model for the baseline period (1974-2000) and future period (2074-2100) it was highlighted a decrease (-20%) of annual precipitation (mm) and increase (+3.7 oC) of mean annual temperature until the end of the 21th century. The decrease of precipitation will be higher in spring, while the increase of the temperature will be higher in summer (Fig 2).
In order to determine the effect of climate change on surface runoff it was necessary to estimate the input parameters of the water balance model. To this end, geological subsoil, land cover and soil data layers were generated using Geographical Information Systems (GIS) (Fig. 3). Taking into account the above mention parameters mean CN values for Klinovitikos catchment was estimated to 54.2, the potential maximum soil moisture storage (K) equal to 214.29 and mean infiltration rate (w) equal to 0.33.
Based on the above mention parameters, the water balance model was applied and surface runoff was estimated for both baseline (1974-2000) and the future (20742100) period. The results showed that the decrease in rainfalls and increase in temperatures lead to decrease of surface runoff.
0 -5 -10 -15 -20 -25 f f o n u R e ca ) fr m Su (m s e g n a h C
Additionally, the results between the two aforementioned periods were compared so as to quantify the effects on climate change of surface runoff. It was noticed a decrease in surface runoff until the end of the 21st century, approximately -35.5%, 14.4%, -49.8% and -43.7% for winter, autumn, spring, and summer, respectively (Fig. 4).
During the last decades there is a growing concern about the effects of climate change on water resources, especially in Mediterranean region where climate expected to be warmer and dryer. In this study, the effect of climate change on surface runoff in a mountainous catchment of Central Greece was assessed using the Thornthwaite and Mather water balance model and climate simulation of RegCM3 regional climate model.
Regarding future climate conditions in the study area, a decrease of precipitation (mm) and increase of temperature is expected until the end of the 21th century. Additionally, the higher decrease of monthly rainfall was recorded in spring, while the higher increase is occurred in summer. Taking into account the reported changes of climatic condition significant decrease of surface runoff changes was estimated especially in spring and summer months.
The need for rational water resources management nowadays is a necessity. Despite the construction of classical water saving dams (reservoir), appropriate adaption measures must be applied in the mountainous forested regions where surface runoff generates. These measures include appropriate silvicultural treatments in order to achieve all-aged forest stands structure which can increase water production and stream regulation using check dams as after their siltation, water infiltrated through the deposits that act as ideal artificial aquifers.