Department of Marketing, Transfer of Innovations and Economic Studies,

Institute of Irrigated Agriculture of NAAS, Naddniprianske, Kherson, Ukraine

Corresponding author email: pavel.likhovid@gmail.com

Article Publishing History

INTRODUCTION

Reference evapotranspiration (ET₀) is a valuable index for determination of demands for irrigation water and climate parameters (Hargreaves, 1994). The technique of ET₀ estimation embraces direct (mainly lysimetric) and indirect methods. Indirect methods are represented by a set of formulas, based on empirical description of physical processes, which take place in the system “plant-soil-atmosphere” and are connected with transport of water during physiological processes in plants in the concrete weather conditions. The most wide-spread indirect method, which is generally accepted as the standard worldwide, is the method of Penman-Monteith (Sentelhas et al., 2010). Notwithstanding the fact that this method was proved to be the most reliable in the estimation of ET₀, it is not possible to apply it in every study due to the difficulties, connected with sophisticated computations and huge amount of the input data required to obtain a reliable result.

FAO provides software named ET₀ Calculator, which is aimed to simplify the process of reference evapotranspiration assessment by Penman-Monteith method; however, it is still advisable to provide all the inputs to get the correct estimation, although it is possible to come to nothing more than the maximum and minimum air temperature to perform the rough assessment (Raes & Munoz, 2009). The previous study  of Lykhovyd, (2020) has proved that the idea is not only theoretically interesting, but is of practical value and provides a good chance for agricultural producers to estimate and forecast ET₀ in the concrete zones using fast and convenient formula of K×T (K – coefficient, mm/°C; T – air temperature, °C, (Lykhovyd, 2020).

The goal of the study is to provide an alternative method for local ET₀ computation using the most common weather index – air temperature – as the only input for the reference evapotranspiration estimation in the zone of Ukrainian Polisia.

MATERIAL AND METHODS

Polisia is one of the largest forest zones of Europe and it is located in the south-western part of the Eastern-European Lowland, or in the Polisia Lowland. Ukrainian Polisia is the part of the natural zone, which occupies mainly Northern Ukraine, namely, Volyn (Lutsk), Rivne, Zhytomyr, Kyiv and Chernihiv Oblasts (Internet Encyclopedia of Ukraine). Therefore, the main big cities and meteorological stations situated in the zone are in Lutsk, Rivne, Zhytomyr, Kyiv and Chernihiv (Fig. 1).

Figure 1: Ukrainian Polisia on the map of Nothern Ukraine, Eastern Poland, Southern Belorussia, and Western Russia (Internet Encyclopedia of Ukraine).

The study was based on the long-term meteorological data on monthly maximum, mean, minimum air temperature and wind speed during the warm period (average air temperature ≥0°C), which were obtained from Lutsk, Rivne, Zhytomyr, Kyiv and Chernihiv hydrometeorological stations for the period of January 2010 – October 2020. The mentioned stations were chosen as the key ones to represent the zone of Ukrainian Polisia. The total number of data inputs was: Volyn (Lutsk) oblast – 76; Rivne – 105; Zhytomyr – 102; Kyiv – 105; Chernihiv – 100, Ukrainian Polisia overall – 488. Meteorological indices were used as the inputs for calculation of the reference evapotranspiration using Penman-Monteith equation in ET₀ Calculator software (the developer – FAO UNO). The pairs of “mean air temperature – ET₀” were formed and processed by linear regression within Biostat v7 software at p<0.05. To evaluate the accuracy of the ET₀ computation using the developed regression models the values of mean absolute percentage error (MAPE) and coefficient of determination (Predicted R²) were estimated for each zone (Moreno et al., 2013).

RESULTS AND DISCUSSION

As a result of regression analysis, six models for reference evapotranspiration computation based on the air temperature only were developed, including not only the model for Ukrainian Polisia overall, but for every oblast of this zone (Table 1).

Table 1. Reference evapotranspiration calculation for Ukrainian Polisia and its sub-zones using air temperature (AT), ± MAPE

Name of the zone	Calculation equation (Predicted R2)
Lutsk (Volyn) oblast	0.3551×AT±29.34% (0.9432)
Rivne oblast	0.4122×AT±28.18% (0.9255)
Zhytomyr oblast	0.2799×AT±26.83% (0.9484)
Kyiv oblast	0.3179×AT±26.79% (0.9459)
Chernihiv oblast	0.3508×AT±26.03% (0.9391)
Ukrainian Polisia	0.3410×AT±30.10% (0.9235)

The values of MAPE and Predicted R² testify about acceptable accuracy of the developed regression models for fast and rough estimation of the reference evapotranspiration in the zone of Ukrainian Polisia and some Northern oblasts of Ukraine. The best accuracy was obtained for Zhytomyr, Kyiv, and Chernihiv oblasts, while the overall result for Ukrainian Polisia is slightly worse in accuracy. The highest reference evapotranspiration with accordance to the regression model coefficients is usual for Rivne oblast (mainly owing to strong winds in the region), while the least evapotranspiration is associated with Zhytomyr oblast (light winds and favorable temperature regime is a feature of this region).

Our study is just another option among the available ones for reference evapotranspiration estimation in case of lack of required for full Penman-Monteith equation meteorological data. While some scientists look upon remote sensing technologies as a relevant and most applicable alternative in this case (Cammaleri & Ciraolo, 2013), others try to apply weather forecasts data instead of the data derived directly from meteorological stations (Er-Raki et al., 2010), or are looking for the simpler available method testing the accuracy in comparison with Penman-Monteith one (Tabari et al., 2013), our study was directed on simplification of the original ET₀ calculation equation by the methods of mathematical statistics. The application of mathematical statistics in the form of artificial neural networks has been proved to be quite effective for the purpose (Feng et al., 2016).

Although neural networks are considered to be the best option in such cases, providing the highest accuracy in life sciences (Lykhovyd, 2018), our study was focused on much simpler technique of linear regression analysis, which might be less accurate but more practical as it is possible to get the result in the form of an equation. Besides, it has been proved by the results of our previous study on the reference evapotranspiration estimation in the South of Ukraine that the accuracy of such an approach is sufficient enough to be used in science and practice of irrigated agriculture for rough estimation and forecast of reference evapotranspiration (Lykhovyd, 2020).

CONCLUSION

It is possible to assess reference evapotranspiration in the zone of Ukrainian Polisia by the formula ET₀ = 0.3410×AT. The accuracy of the method is 30.10%. The mentioned model is recommended for reference evapotranspiration estimation in the zone both in practical and scientific purposes. The model is going to be used in the complex research aimed to provide general ET₀ computation formula for Ukraine. Further studies on the enhancement of the model quality are also projected.

Conflict of Interest Statement: There is no conflict of interest regarding the publication of this paper.

REFERENCES

Cammalleri C and Ciraolo G (2013) A simple method to directly retrieve reference evapotranspiration from geostationary satellite images International Journal of Applied Earth Observation and Geoinformation Vol 21 Pages 149-158

Er-Raki S A Chehbouni S Khabba V Simonneaux L Jarlan A Ouldbba JC Rodriguez and Allen R (2010) Assessment of reference evapotranspiration methods in semi-arid regions: can weather forecast data be used as alternate of ground meteorological parameters? Journal of Arid Environments Vol 74 No 12 Pages 1587-1596

Feng Y N Cui L Zhao X Hu and Gong D (2016). Comparison of ELM, GANN, WNN and empirical models for estimating reference evapotranspiration in humid region of Southwest China Journal of Hydrology Vol 536 Pages 376-383.

Hargreaves GH (1994). Defining and using reference evapotranspiration Journal of Irrigation and Drainage Engineering Vol 120 No 6 Pages 1132-1139

Internet Encyclopedia of Ukraine URL: http://www.encyclopediaofukraine.com/display.asp?linkpath=pages%5CP%5CO%5CPolisia.htm

Lykhovyd PV (2018) Prediction of sweet corn yield depending on cultivation technology parameters by using linear regression and artificial neural network methods Biosystems Diversity Vol 26 No 1 Pages 11-15.

Lykhovyd PV (2020) Assessment of reference evapotranspiration in the South of Ukraine by air temperature International Journal on Emerging Technologies Vol 11 No 5 Pages 278-282

Moreno JJM AP Pol AS Abad and Blasco BC (2013). Using the R-MAPE index as a resistant measure of forecast accuracy Psicothema Vol 25 No 4 Pages 500-506

Raes D and Munoz G (2009). The ET₀ Calculator, Reference Manual Version 3.1. Food and Agricultural Organization of the United Nations (FAO) Rome

Sentelhas PC TJ Gillespie and Santos EA (2010) Evaluation of FAO Penman–Monteith and alternative methods for estimating reference evapotranspiration with missing data in Southern Ontario, Canada Agricultural Water Management Vol 97 No 5 Pages 635-644

Tabari H ME Grismer and Trajkovic S (2013) Comparative analysis of 31 reference evapotranspiration methods under humid conditions Irrigation Science Vol 31 No 2 Pages 107-117