**3. Sensitivity analysis**

If the model in certain areas has not been used, then it is necessary to carry out sensitivity analysis. Sensitivity analysis limits the number of parameters that need optimization to achieve good correlation between simulated and measured data. The method of analysis in the SWAT model called PARASOL is based on the method of Latin Hypercube One-factorat-a-Time (LH-OAT). LH-OAT combines the advantages of global and local sensitivity analysis (van Griensven et al., 2006). This method performs LH sampling of data at first, followed by OAT sampling. The new scheme allows the LH-OAT to unmistakably link the changes in the output data of each model to the modified parameter (van Griensven et al., 2006). For the sensitivity analysis and calibration a special tool called SWAT-CUP is available which includes all important algorithms (GLUE, PSO, MCMC, PARASOL and SUFI2) of which Sequential Uncertainty Method (SUFI2) was shown to be very effective in identifying sensitive parameters (Abbaspour et al., 2007).

Tool within the model can automatically carry out the sensitivity analysis without the measured data or with the measured data. The tool varies values of each model parameter within a range of (MIN, MAX). Parameters can be multiplied by a value (%), part of the value can be added to the base value, or the parameter value can be replaced by a new value. The final result of the sensitivity analysis are parameters arranged in the ranks, where the parameter with a maximum effect obtains rank 1, and parameter with a minimum effect obtains rank which corresponds to the number of all analyzed parameters. Parameter that has a global rank 1, is categorized as "very important", rank 2 − 6 as "important", rank 7 − 41 (i.e. the number of parameters in the analysis − i.e. flow 7 - 26) as "slightly important" and rank 42 (i.e. flow 27) as "not important" because the model is not sensitive to change in parameter (van Griensven et al., 2006).

Sensitivity analysis was performed using the measured data of the river Reka tributary Kožbanjšček (subcatchment 5) and the river Dragonja (subcatchment 14). The analysis was performed for an average daily flow, sediments, TP and NO3 -. Table 3 represents for each model the first 10 parameters that have the greatest impact on the model when they are changed. The sensitivity analyses demonstrated great importance of the hydrological parameters that are associated with surface and subsurface runoff.

Alpha\_Bf factor determines the share between the base and surface flow contribution to the total river flow. Cn2 curve runoff determines the ratio between the water drained by the surface and subsurface runoff in moist conditions. Ch\_K2 describes the effective hydraulic conductivity of the alluvial river bottom (water losing and gaining). Surlag represents the

*n*

=± − ⋅ (6)

2 ( 1) *<sup>s</sup> txtn*

If the model in certain areas has not been used, then it is necessary to carry out sensitivity analysis. Sensitivity analysis limits the number of parameters that need optimization to achieve good correlation between simulated and measured data. The method of analysis in the SWAT model called PARASOL is based on the method of Latin Hypercube One-factorat-a-Time (LH-OAT). LH-OAT combines the advantages of global and local sensitivity analysis (van Griensven et al., 2006). This method performs LH sampling of data at first, followed by OAT sampling. The new scheme allows the LH-OAT to unmistakably link the changes in the output data of each model to the modified parameter (van Griensven et al., 2006). For the sensitivity analysis and calibration a special tool called SWAT-CUP is available which includes all important algorithms (GLUE, PSO, MCMC, PARASOL and SUFI2) of which Sequential Uncertainty Method (SUFI2) was shown to be very effective in

Tool within the model can automatically carry out the sensitivity analysis without the measured data or with the measured data. The tool varies values of each model parameter within a range of (MIN, MAX). Parameters can be multiplied by a value (%), part of the value can be added to the base value, or the parameter value can be replaced by a new value. The final result of the sensitivity analysis are parameters arranged in the ranks, where the parameter with a maximum effect obtains rank 1, and parameter with a minimum effect obtains rank which corresponds to the number of all analyzed parameters. Parameter that has a global rank 1, is categorized as "very important", rank 2 − 6 as "important", rank 7 − 41 (i.e. the number of parameters in the analysis − i.e. flow 7 - 26) as "slightly important" and rank 42 (i.e. flow 27) as "not important" because the model is not sensitive to change in

Sensitivity analysis was performed using the measured data of the river Reka tributary Kožbanjšček (subcatchment 5) and the river Dragonja (subcatchment 14). The analysis was performed for an average daily flow, sediments, TP and NO3-. Table 3 represents for each model the first 10 parameters that have the greatest impact on the model when they are changed. The sensitivity analyses demonstrated great importance of the hydrological

Alpha\_Bf factor determines the share between the base and surface flow contribution to the total river flow. Cn2 curve runoff determines the ratio between the water drained by the surface and subsurface runoff in moist conditions. Ch\_K2 describes the effective hydraulic conductivity of the alluvial river bottom (water losing and gaining). Surlag represents the

*x* sample arithmetic mean (alternative scenario)

*s* sample standard deviation (alternative scenario)

identifying sensitive parameters (Abbaspour et al., 2007).

parameters that are associated with surface and subsurface runoff.

parameter (van Griensven et al., 2006).

*n* number of pairs (alternative scenario)

Student distribution

**3. Sensitivity analysis** 

2 *t*α

*μ* average of the corresponding random variables (base scenario)

α

surface runoff velocity of the river and Esco describes evaporation from the soil. For the sediment modelling the most important parameters are Spcon and Spexp that affect the movement and separation of the sediment fractions in the channel. Ch\_N − Manning coefficient for channel, determines the sediment transport based on the shape of the channel and type of the river bed material. Ch\_Cov − Channel cover factor and Ch\_Erod − Channel erodibillity factor proved to be important for the Dragonja catchment. Soil erosion is closely related to the surface runoff hydrological processes (Surlag, Cn2). The analysis showed importance of the hydrological parameters that are associated with surface and subsurface runoff (Cn2, Canmx, Sol\_Awc), evaporation (Revapmin, Esco, Blai), base flow (Alpha\_Bf) and groundwater (Rchrg\_Dp, Gwqmn), suggesting numerous routes by which sediment nitrate nitrogen (NO3-N) and TP are transported (Table 3). We noticed that the amount of N is also influenced by other parameters that are not included in the sensitivity analysis tool like Rate factor for humus mineralization of organic nutrients active N and P (CMN.bsn), half-life of nitrates and the shallow aquifer (HLIFE\_NGW.gw), fraction of algal biomass that is N (Al1.wwq). TP results are significantly affected by the parameters that control surface runoff (Cn2, Canmx, Usle\_P). Usle\_P factor adjusts the USLE value for a particular land management. This means that the soil loss from the terraced land is different, from non terraced slopes. Parameters which have a significant impact on P, but not included in the sensitivity analysis tool are: fraction of algal biomass that is P (Al2.wwq), P availability index (PSP.bsn), P enrichment ratio for loading with sediment (ERORGP.hru), BC4.swq, benthic sediment source rate for dissolved P in the reach (RS2.swq), organic P settling rate (RS5.swq).


Table 3. SWAT parameters ranked by the sensitivity analysis for the Reka subcatchment 5 and Dragonja subcatchment 14 (1998 - 2005)
