Evapotranspiration Partitioning in Surface and Subsurface Drip Irrigation Systems

Water transfer from the soil-plant system to the atmosphere occurs through evapotranspiration, which includes evaporation of water from the soil and other surfaces and transpiration through plant stomata. Evaporation is the process whereby liquid water is converted to water vapor (vaporization) and removed from the evaporating surface (vapor removal). Water evaporates from a variety of surfaces, such as lakes, rivers, pavements, soils and wet vegetation. Transpiration consists of the vaporization of liquid water contained in plant tissues and the vapor removal to the atmosphere. Transpiration, like direct evaporation, depends on the environmental factors including energy supply, vapor pressure gradient and wind. Hence, radiation, air temperature, air humidity and wind terms should be considered when assessing transpiration. The soil water content and the ability of the soil to conduct water to the roots also determine the transpiration rate. The transpiration rate is also influenced by crop characteristics, environmental aspects and cultivation practices. Evaporation and transpiration occur simultaneously and there is no easy way of distinguishing between the two processes (Allen et al., 1998). Where the evaporating surface is the soil surface, the amount of water available at the soil surface is the main sources of evaporation. Accordingly any irrigation method which decreases the water availability in the soil surface will decrease the evaporation considerably. Apart from the water availability in the topsoil, the evaporation from a cropped soil is mainly determined by the fraction of the solar radiation reaching the soil surface. The degree of shading of the crop canopy is other factors that affect the evaporation process. This fraction decreases over the growing period as the crop develops and the crop canopy shades more and more of the ground area. When the crop is small, water is predominately lost by soil evaporation, but once the crop is well developed and completely covers the soil, transpiration becomes the main process. Evapotranspiration (ET) partitioning into soil surface evaporation (Es) and crop transpiration (Tc) is fundamental to many irrigation management studies. In many cases such as design of irrigation system, measurement of the whole ET is sufficient but when the research on water consumed by crop become more precise measurement or estimation of its


Introduction
Water transfer from the soil-plant system to the atmosphere occurs through evapotranspiration, which includes evaporation of water from the soil and other surfaces and transpiration through plant stomata.Evaporation is the process whereby liquid water is converted to water vapor (vaporization) and removed from the evaporating surface (vapor removal).Water evaporates from a variety of surfaces, such as lakes, rivers, pavements, soils and wet vegetation.Transpiration consists of the vaporization of liquid water contained in plant tissues and the vapor removal to the atmosphere.Transpiration, like direct evaporation, depends on the environmental factors including energy supply, vapor pressure gradient and wind.Hence, radiation, air temperature, air humidity and wind terms should be considered when assessing transpiration.The soil water content and the ability of the soil to conduct water to the roots also determine the transpiration rate.The transpiration rate is also influenced by crop characteristics, environmental aspects and cultivation practices.Evaporation and transpiration occur simultaneously and there is no easy way of distinguishing between the two processes (Allen et al., 1998).
Where the evaporating surface is the soil surface, the amount of water available at the soil surface is the main sources of evaporation.Accordingly any irrigation method which decreases the water availability in the soil surface will decrease the evaporation considerably.Apart from the water availability in the topsoil, the evaporation from a cropped soil is mainly determined by the fraction of the solar radiation reaching the soil surface.The degree of shading of the crop canopy is other factors that affect the evaporation process.This fraction decreases over the growing period as the crop develops and the crop canopy shades more and more of the ground area.When the crop is small, water is predominately lost by soil evaporation, but once the crop is well developed and completely covers the soil, transpiration becomes the main process.Evapotranspiration (ET) partitioning into soil surface evaporation (E s ) and crop transpiration (T c ) is fundamental to many irrigation management studies.In many cases such as design of irrigation system, measurement of the whole ET is sufficient but when the research on water consumed by crop become more precise measurement or estimation of its two components, E s and T c , will be valuable.Prime attempts to partition ET include methods covering the ground surface of a plot to eliminate E s and measure water loss by crop (Tc) and compare it with water loss by uncovered plot (ET) to reach E s (Harrold et al., 1959;Peters and Rassel, 1959;Shaw, 1959).The researches showed, applying ground cover to partition ET changes the field and soil surface energy balance and does not estimate crop transpiration accurately in natural condition ( (Fritschen and Shaw, 1961).Developing microlysimeter (Boast and Robertson., 1982), provided E s measurement directly without drastic changes in soil and field condition caused by surface covers (Shawcroft and Gardner., 1983).Applying this method simultaneously with ET measurement at the same place provides ET components separately (Ham et al., 1990;Jara et al., 1998;Sepaskhah and Ilampour, 1995).Some results showed there are some limitations with using micro-lysimeter especially when E s consisted small portion of ET (Ham et al., 1990, Jara et al., 1998).Additional researches applied ET measurement simultaneously with transpiration measurement frequently using sap flow gauges.Sakuratani (1987) was first who reported ET components separately in this way (Ham et al., 1990).In some of those researches micro-lysimeter was used for evaluating the accuracy of measured E s with the calculated ones.Ashktorab et al. (1989) measured E s by Bowen ratio energy balance from bare soil.Then they applied it with ET measurement using weighing lysimeter to partition ET components.Their results suggest an accurate method to measure E s under the crop canopy (Ashktorab et al., 1994).The latest work on partitioning ET was method applying Bowen ratio energy balance (BREB) to measure both ET and E s (Zeggaf et al., 2008).Their results showed this technique can provide a framework for partitioning ET at maize field simply and economically to previous methods (Zeggaf et al., 2008).Applied and precise methods of ET partitioning provide useful data for farm irrigation management and water use efficiency improvement.This knowledge particularly for modern irrigation systems implementing with high costs is more important, where E s reduction is one of the advantages of modern irrigation systems such as surface drip irrigation (DI) and subsurface drip irrigation (SDI).Accurate and efficient management should be applied to reach such advantages of these systems.Subsurface drip irrigation (SDI) is an alternative to conventional drip irrigation, which would become an attractive option to most of the farmers in arid and semi arid regions like Iran.The advantages of SDI compared to surface drip irrigation include direct application of water to the root zone, less E s , potentially greater water use efficiency and fewer weed and disease problems (Phene et al, 1991).SDI reduced tillage using semi-permanent beds (Senn and Cornish, 2000) and removed the need for deep cultivation between the crops.SDI has been found to increase yield over surface drip (Sakellarious-Makrantonaki et al., 2002); furrow irrigation (Hanson et al., 1997); and sprinkler irrigation (De Tar et al., 2004), providing the SDI system receives good irrigation scheduling (Haman and Smajstrla, 2002).Soil and canopy energy balances have some interactions in crop environment and irrigation systems change this environment significantly which may has influences on ET.Sprinkle irrigation increase the air humidity, surface irrigation keep the soil surface wet for at least one day after irrigation and, drip irrigation decrease the crop water stress by short irrigation interval.Toward a precise irrigation management, measuring ET component is required to have confidence on development of new and precise irrigation systems such as SDI.Besides, relation between the effective factors in ET could provide us valuable information for better farm irrigation management and water use efficiency improvement.Based on our knowledge, there is no information on proportion of E s or T c component under SDI, where soil surface kept dry and that may increase the soil surface temperature during the day time.
In this chapter we are going to show how we could partition ET for SDI and DI systems in a maize field using a BREB method and discuss energy balance elements variation under these two irrigation systems.

Energy balances theories 2.1 Energy balance theory at maize field
Energy balance at field level can be expressed as: Where R n is net radiation reaching the field, above the maize canopy, λE is latent heat flux, H is sensible heat flux and G is soil heat flux (all units of W/m 2 ).In equation ( 1) the convention used for the signs of the energy fluxes is R n positive downward and G is positive when it is conducted downward from the surface, λE and H are positive upward.
Partitioning of energy between λE and H is determined by the BREB (Bowen., 1926, Perez et al., 1999) by the following equation: where is the Bowen ratio.By solving equation ( 1) and ( 2) at the same time the following expressions for λE and H are obtained: Assuming equality of eddy transfer coefficients for sensible heat and water vapor in the averaging period and measuring air temperature and vapor pressure gradients between the two levels, the Bowen ratio ( ) is calculated by: Where ∆T and ∆e are air temperature and vapor pressure differences between the two measurement levels and is psychrometric constant which is calculated by the following equation: Where C p is the specific heat of air at constant pressure (1.01 kJ/kg ˚C), P is atmospheric pressure (kPa), ε is the ratio between the molecular weights of water vapor and air (0.622), and L v is latent heat of vaporization (kJ/kg).Psychrometric constant for the experiment site was determined 0.058 (kPa/˚C).

Energy balance theory at soil surface
Energy balance at soil surface can be expressed as: where, R ns is the net radiation reaching the soil surface, λE s is the soil surface latent heat flux, H s is sensible heat flux from the soil surface (all units of W/m 2 ).R ns was determined by the empirical equation ( 8) with R n and LAI which has been used previously by some other authors (Gardiol et al., 2003;Kato et al., 2004).
Bowen ratio at soil surface was calculated similar to the energy balance computation at field level with the following equation: which using equations ( 5) and ( 6) and measurement of air temperature and vapor pressure gradients by ventilated psychrometers near the soil surface, Bowen ratio at soil surface was determined.By solving equation ( 7) and ( 9) simultaneously, latent heat flux from the soil surface was determined by equation ( 10).

Energy balance theory at crop canopy
Energy balance at maize canopy can be expressed as below (Zeggaf et al., 2008, Ham et al., 1991): where R nc is net radiation absorbed by crop canopy, λE c is crop canopy latent heat flux and H c is crop canopy sensible heat flux (all units of W/m 2 ).Applying the principle of continuity and the definition of R n , it can be shown that R nc , equation ( 12), is the difference between R n measured above and that below the maize canopy (Ham et al., 1991).
Canopy latent heat flux was calculated from equation ( 13): Then H c was calculated as a residual from equation (11).

Methodology development
The research was conducted in summer 2009 at experimental station of agricultural engineering research institute (AERI), Karaj-Iran (35˚ 21′ N, 51˚ 38′ E, 1312.5 m above sea level).The field soil was prepared for planting in spring.Results from soil experiments up to 80 cm below surface showed the soil type was loam texture (47 % sand, 44 % silt, 9 % clay) with ECe=1.7.Irrigation water were supplied from underground well with an quality which had no negative impact for maize (EC=0.8dS/m and pH=7.8).
The experimental field was defined in an area of 40×60 m 2 in selected site (Fig. 1).A day before planting 50 kg potash fertilizer was added to soil and maize (Double Cross 370) was planted on 15 June 2009.The crop was planted with 0.75 m row width and north-south orientation.
The field was bordered by irrigated maize field except in western side which was unplanted.Irrigation water was supplied from the well and chemical quality analysis showed water in this region has good quality.A subsurface drip-tape irrigation system with 0.30 m dripper distance was used to apply irrigation water.Drip tapes were placed 15 cm below the soil surface in nearest place to the plant rows.Special attention was paid while positioning drip tapes to transfer water correctly.Crop water requirement was estimated based on long time meteorological data (averaging from 1988 to 2008) and calculation of crop ET by method recommended in FAO 56, Penman Monteith Method (P-M) (Allen et al., 1998).The P-M method is recommended for the Karaj area by Dehghanisanij et al. (2004).From the early crop growth period 20% over irrigation based on 3 day intervals was applied to prevent water stress.Recommended nitrate fertilizer (according to soil experiments it was 400 kg/ha) was distributed during the crop growth period and closely to crop establishment place by irrigation system (fertigation).At the period of this experiment 41-44 and 59-62 day after emergence (DAE), leaf area index (LAI) was measured in 41, 44, 59, 62 DAE.Each time 3-5 plants were selected randomly and the whole leaf area of a plant was measured with leaf area meter (Area Measurement system, DELA-T Devices, ENGLAND) in the laboratory.
Then LAI was calculated from multiplying the average plant leaf area by plant density.LAI values for the days between the days of measurement obtained by linear interpolation (Gardiol et al., 2003).Automatic weather station was established in the field simultaneously with start of experiment period and hourly average values of solar radiation (Rs), air temperature, relative humidity and wind speed were measured and logged continuously.
From 41-44 and 59-62 DAE, ET and Es were determined simultaneously by measuring all energy fluxes at maize field and soil surface using two independent measurement systems.
Then by subtracting the latent heat flux at soil surface from the latent heat flux at maize field (ET), transpiration was obtained.Energy balance equipments consisted of a net radiometer (CNR1, Kipp & Zonen), two soil heat flux plates (MF-180M, EKO Japan) and four hand made thermocouple ventilated psychrometers for Bowen ratio measurement in both field level and soil surface.The details of constructed psychrometers have been described in Kosari (2010).Two independent measuring systems separated by 5 m distance, were placed 9 m from the east edge of the field as the system number 2 was positioned 5 m from the east edge to maximize fetch to height ratio when prevailing wind (north-western to southeastern) were present (Fig. 1).That was greater than minimum adequate ratio reported by Heilman et al. (1989) for measuring Bowen ratio during our experiment period.Measurement equipments in each measurement system were installed on a tall rod.Two ventilated psychrometers used for measuring temperature and water vapor gradients at field level above the crop.These two psychrometers were installed 1 m apart as the lowest one was positioned 0.2 m above the crop canopy (Ham et al., 1991;Jara et al., 1998).The remaining two psychrometers used for measuring temperature and water vapor gradients at soil surface.These two psychrometers were fixed 0.1 m apart on the rod as the lowest one was positioned 0.05 m above the soil surface (Ashktorab et al., 1989).Net radiation at field level was measured with net radiometer installed 1 m above crop canopy.Soil heat flux was calculated as an average of two soil heat flux plates positioned 0.02 m below the soil surface.All data were measured every minute by a CR23X data logger connected to an AM16/32 multiplexer (Campbell Scientific, Inc., UT) and averaged 30 min intervals.www.intechopen.com

Evapotranspiration analysis
Diurnal trend of evapotranspiration measurement by BREB method compare to evapotranspiration estimation by P-M method for 60 and 61 DAE are shown in Fig. 2.These days were selected because they are representative of cloudy and clear sky condition respectively which is believed they show all sky condition during measurement period.As it is shown in Fig. 2 both methods have the same trend and a good correlation (R 2 =0.92 and 0.95 for DAE 60 and 61, respectively).Therefore evapotranspiration by BREB showed 9% variations compare to P-M method which can be acceptable.Positive value of MBE parameter shows overestimation of P-M method compare to BREB.These differences can be caused by different measurements of effective and required parameters in both methods.In other word, BREB evapotranspiration was obtained by direct measurement of required parameters at field level and soil surface while in P-M method maize evapotranspiration was obtained by estimation of reference evapotranspiration and crop coefficient.Ortega et al. (1995) found a good correlation between refernce evapotranspiration by BREB and Penman method on irrigated grass.

Energy balance at farm level
Daytime average of energy balance measurements in terms of (W/m 2 ) at Maize field, soil surface and crop canopy for DI and SDI are presented in tables 2 to 4. In the measurement period net radiation values ranged from 304 to 333 (W/m 2 ) resulted by the minimum and maximum solar radiation in corresponding days respectively (Table 2).Latent heat flux (λE) From the Maize field ranged from 207 to 267 (W/m2) for SDI and 197 to 296 (W/m2) for DI.
According to the results λE ranging from 62 to 83 % of R n under SDI and 61 to 94 % for DI, which shows maize cropping system is under non-stressed conditions (Ham et al., 1991).Herein G/R n was ranging 6 to 8 % for SDI and 8-15 % for DI, which is close to 10 % reported by Yunusa et al. (2004).Soil surface energy balance measurements showed R ns values decreased with crop growth and canopy cover increment.Furthermore it showed R ns partitioned primarily between soil heat flux and latent heat flux from the soil surface and there was very little sensible heat flux during experiment period.The G variation under SDI was 17 to 25, and it was 25 to 47 W/m 2 in under DI.The λEs accounted for about 40 to 73 (W/m 2 ) in SDI, where maximum value of λE s was 60 W/m 2 for surface DI.Accordingly, λE s /R ns and G/R ns ratios ranged between 56 to 71 % and 18 to 31 % respectively (Table 3).The G under SDI was much less compared to that under DI during crop developing stage than mid-season stage.It is while, λE s was larger under SDI compared to that for DI.These results contributed to higher possible potential for T under SDI during crop developing stage (  Available energy (R n -G) and λE from maize field for the 60 th DAE are shown in Fig. 4. The linear regression lines between λE and R n -G were obtained with high values of r 2 =0.99 for both SDI and DI.Based on the slope of the trade lines in Fig. 4, there was no a significant reduction in available energy to the maize field and soil surface between two irrigation system.Accordingly, SDI aimed at reducing soil evaporation compared to DI, is not effective when soil surface covered by canopy (LAI=3.5).
Available energy (R ns -G) and λE s from soil surface for 41 th DAE are shown in Fig. 5.The 41 th DAE presenting crop developing stage, when LAI was about 2.20.Accordingly, there was a wide scattering for available energy under DI which might be because soil surface is not uniformly wet under DI.However, the condition under SDI was uniformly and a linear regression lines between λE s and R ns -G were obtained.

Conclusion
The Bowen ratio method could be used for partitioning ET under surface (DI) and subsurface drip irrigation system (SDI).Partitioning ET for advance irrigation system could provide us useful information for better irrigation management during crop growth stages and development of new irrigation technique.Partitioning ET and measurement of energy balance over maize field, canopy and soil by Bowen ratio showed that soil had major impact on the energy balance between the soil and canopy when soil surface is not covered fully by crop canopy.In crop developing stage, energy balance of maize field was different under DI and SDI.This result could be contributed to more difference between the systems in early crop development stage, when soil surface is not covered fully by crop canopy.
As it was shown daytime soil heat flux values were greater under DI (25-47 W/m 2 ) compared to that under SDI (17-25 W/m 2 ).It may caused by heat convection in DI while moving down the water from the surface and higher temperature of water when drip tapes were positioned on the ground.Therefore available energy for soil evaporation, R ns -G, was lower in DI.As it was shown λE s accounted for about 41 to 63% of R ns in DI while it was about 56 to 71% in SDI.It was observed the ground in both DI and SDI became wet but reverse direction of moving water in subsurface system, as may contributed to more evaporation in SDI.According to the results, more consideration should be applied using SDI systems on depth of lateral line which carrying the emitters, canopy size, crop type, and plant water stress affect soil and canopy energy balances.Those data will be useful for validation of ET models.

Fig. 1 .
Fig. 1.Schematic diagram showing field position and location of energy balance measurement systems.

Fig. 3 .
Fig. 3. Diurnal trend of energy balance components at soil surface of maize field in 60th DAE under surface and subsurface drip irrigation.

Fig. 4 .
Fig. 4. Available energy (R n -G) and latent heat flux (λE) from maize field and soil (R -G and λE s ) in 60 th DAE for surface (DI) and subsurface (SDI) drip irrigation.

Table 1 .
Daytime average values of meteorological parameters measured by the automatic weather station in the experiment period are shown in Table1.Plant in days 41-44 DAE was in developing stage and in days 59-62 DAE was in mid-season stage.Irrigation has been done on 41, 44, 59 and 63 DAE, which exceptionally because of some problems the last one irrigated with 4 days interval.In the experiment period the 42 and 60 DAE received maximum and minimum solar radiation respectively.Daytime average values of meteorological values in experiment period.

Table 2 .
Daytime average energy fluxes at maize field.