A MODEL OF CORRECTION INDEXES DETERMINATION FOR VIRGINIA TYPE TOBACCO ON THE BASIS OF FIELD MEASURED ETP

Numerous methods of calculating ET0 and ETP offer a great practical possibility for their application, but in order to improve their efficiency it is necessary to check them in particular regions. The experimental study of Virginia type tobacco in irrigation conditions was carried out in the area of village Brzi Brod (Niš). The established potential evapotranspiration of Virginia type tobacco ranged from 498.1-512.7 mm, depending on annual weather conditions. Highly significant correlation coefficients were found between the directly observed ETP of Virginia type tobacco and reference evapotranspiration (ET0) calculated by the following methods: FAO-56 PenmanMonteith (r=0.88), FAO-24 Radiation (0.79), FAO-24 Blaney-Criddle (r=0.85), Tornthwaite (r=0.90) and Hargreaves (r=0.76). The established correction indexes for translation of ET0 calculated by each method were the following: FAO-56 Penman-Monteith 0.91; FAO-24 Radiation 0.76; FAO-24 BlaneyCriddle 0.89; Tornthwaite 0.92; Hargreaves 0.72.

numerous methods of calculating plant demand for water.When evapotranspiration for an area is calculated by different methods, the obtained values show significant differences between each other.UN Food Agriculture Organization (FAO), aiming to find the universal mathematical method for calculation of ET, established the Consultative Group in 1971.As the result of efforts made by the Consultative Group and more than thirty scientific institutes placed in various climatic areas throughout the world, D o o r e n b o s and P r u i t t (1977) presented in the FAO publication four methods for calculation of reference evapotranspiration (ET 0 ).Since the first ETP calculation methods appeared, there was an intention to establish particular methods for global use.Analysis of numerous results of reference evapotranspiration (ET 0 ), obtained by various calculation methods and direct observations, forced FAO experts at Consultation held in May 1990 to recommend FAO-56 Penman-Monteith equation (A l l e n et al., 1998) as the standard method for calculation of ET 0 .
This study was aimed to compare field measured ETP of Virginia type tobacco with the following five ET 0 calculation methods: Tornthwaite (T h o r n t h w a i t e , 1948), FAO-24 Blaney-Criddle, FAO-24 Radiation (D o o r e n b o s and P r u i t t , 1977), Hargreaves (H a r g r e a v e s and S a m a n i , 1985) and FAO-56 Penman-Monteith.The aims of the study also was estimation of possibility to use those methods, as well as establishing of tobacco crop coefficient for the conditions of southern Serbia.

Material and Methods
The field trial of Virginia type tobacco (Hevesi-9) was set in irrigation conditions, in the vicinity of Brzi Brod village, Nišava river's valley, on the alluvium soil type.Irrigation was scheduled and dosed according to soil humidity, which was observed weekly by thermogravimetric method in drying chamber at 105-110°C.Tobacco planting was done within optimal terms, and a modern production technology for this crop was applied.The trial was set in random block design, adapted for irrigation conditions, and the experimental plot area was 35 m 2 .Field water capacity (FWC) was determined at the location of field trial.Irrigation was done by sprinklers, and the term was determined in regard to soil humidity to the depth of 60 cm.
Meteorological elements (air temperature, precipitation, relative air humidity, air humidity deficiency, wind speed, sunshine duration, global radiation) of the trial area were analyzed at Meteorological Station in Niš.Precipitation amount was also measured at the experimental field.Water consumption for evapotranspiration, at the irrigated variant and the control one, was calculated for each month and the whole vegetation period, by making the balance between water consumption from the soil layer to 2 m under the ground, total precipitation amount in the vegetation and water added by irrigation (B o š n j a k , 1999).
Reference evapotranspiration was calculated by the following methods: Tornthwaite, FAO-24 Blaney-Criddle, FAO-24 Radiation, Hargreaves, and FAO-56 Penman-Monteith.Water demand of plants was calculated by the following equation: The values of crop coefficient for tobacco used in calculation of ETP 0 were according to FAO recommendation for 2002 (May --0.5;June -0.8;July -1.1;August -1.0;September -0.85).
Correction index (K) for Virginia type tobacco was determined from the relation ETP/ET 0 =K.

Results and Discussion
In the irrigated variant (pre-irrigation humidity 60% of FWC) the observed average value of evapotranspiration was 504.2 mm (Table 1 Values of reference evapotranspiration (ET 0 ) are calculated by the use of crop coefficients for tobacco (FAO, 2002), and the values of potential evapotranspiration (ETP 0 ) obtained that way, together with field measured ETP of Virginia type tobacco, are given in the Table 4.The calculated values of ETP 0 for Virginia type tobacco were not enough close to the values of ETP measured in the experimental field, which means that crop coefficients for tobacco recommended by FAO could not be used in the investigated area.In order to enable application of the mentioned five methods for calculation of Virginia type tobacco ETP 0 for the conditions of southern Serbia we calculated correction indexes (K=ETP/ET 0 ), which were used for translating reference evapotranspiration (ET 0 ) to potential evapotranspiration (ETP 0 ) of Virginia type tobacco (Table 5).
The three-year mean values of correction indexes for Virginia type tobacco according to respective methods were the following: FAO-56 Penman-Monteith 0.65-1.04,FAO-24 Radiation 0.53-0.88,FAO-24 Blaney-Criddle 0.58-1.05,Tornthwaite 0.7-1.00,and Hargreaves 0.48-0.87;while the values recommended by FAO ranged from 0.50-1.10.Correction indexes for ETP of Virginia type tobacco showed the same tendency as the recommended FAO indexes in regard to ET 0 -they were lower at the start and end of vegetation, and higher in summer months, which was a consequence of weather conditions and of tobacco developmental stages schedule.Regression analysis showed a high level of correlation (r=0.76-0.90) between ET 0 obtained by the mentioned calculation methods and ETP of Virginia type tobacco measured at the experimental field.That highly significant correlation between ETP measured in natural conditions and various methods of ET 0 calculation was also presented in many previous reports (B o š n j a k , 1982; D e v i t t, 1983; K a s s a m and S m i t h, 2001; X u and S i n g h, 2002; The lowest deviation of the correction index mean value for Virginia type tobacco in regard to FAO tobacco coefficient was calculated for the method FAO-24 Blaney-Criddle (4.7%), and the highest one for the method Hargreaves (18.0%).Deviation of the correction index mean value for the method FAO-56 Penman-Monteith was 7%, for Tornthwaite it was 8.2%, and for FAO-24 Radiation it was 11.8%, all in regard to the recommended FAO coefficient for tobacco.
The Previous reports and this study results point to a high probability of expected deviation from the recommended FAO values of K c .For that reason it is necessary to check recommended values of K c , before their use in ETP 0 calculation for particular region, because any error in those calculations can lead to excessive or insufficient irrigation of a crop in question.

C o n c l u s i o n
On the basis of three-year experimental results obtained in field conditions there was a significant correlation between ETP of Virginia type tobacco and the five methods for ET 0 calculation, and the highest level of correlation was reached by the method Tornthwaite.Possibility of using ET 0 calculation methods in determination of plant water demands (ETP) is of a great practical importance, so in this study were determined correction indexes for Virginia type tobacco (FAO-56 Penman-Monteith -0.91; FAO-24 Radiation -0.76; FAO-24 Blaney-Criddle -0.89; Tornthwaite -0.92; Hargreaves -0.72).
The model of correction indexes (ETP/ET 0 =K) presented in this paper is based on the FAO model, which can be used by other professionals.However, the calculated correction indexes for Virginia type tobacco are based on local parameters, so they should be checked for different soil and climatic conditions.
X u and C h i e n, 2005; K u o et al., 2006; V i l l a N o v a et al., 2006; etc).
).At the control variant the average water consumption for ET was 372.7 mm, which was by 26.1% lower than in irrigated variant.Values of field measured daily ETP of Virginia type tobacco, as well as calculated daily values of ET 0 , are given in the Table3.Tab.3.-Field measured daily ETP of Virginia type tobacco and reference ET 0(2001)(2002)(2003) Tab. 4. -Values of field measured daily ETP of Virginia type tobacco and ETP o calcuIated by FAO crop coefficients for tobacco(2001 -2003) Tab. 5. -Correction indexes for calculation ETP 0 of Virginia type tobacco(2001)(2002)(2003) established differences between calculated correction indexes for the mentioned methods and FAO crop coefficients were expected and in accordance with previous reports.B o š n j a k (1982) carried out similar correction of crop coefficients for maize and soybean in the conditions of Vojvodina.L u g o n j a (1983), in agroecological conditions of Kosovo and Metohia, did correct K c for silage maize grown as the second crop.D o o r e n b o s and P r u i t t (1984) stated that values of crop coefficient mainly depend on local climatic conditions.Modification of K c values depending on climatic conditions of the investigated region is also considered as necessary by A l l e n et al. (1998).S a t t i et al. (2004), in their study, found deviation of ± 20% from the recommended FAO K c value for potato.On the basis of their investigation results, K u o et al. (2006) found a significant deviation of K c values for maize, sorghum and soybean, and as the main reason for that they recognized local climatic conditions.