Dynamics of crop coefficients for citrus orchards of central India using water balance and eddy covariance flux partition techniques

Agricultural Water Management 212 (2019) 68–77

Contents lists available at ScienceDirect

Agricultural Water Management journal homepage: www.elsevier.com/locate/agwat

Dynamics of crop coefficients for citrus orchards of central India using water balance and eddy covariance flux partition techniques

T

⁎

Srinivasa Rao Peddinti , BVN P Kambhammettu Department of Civil Engineering, Indian Institute of Technology Hyderabad, Telangana, India

A R T I C LE I N FO

A B S T R A C T

Keywords: Evapotranspiration SIMDualKc Eddy covariance Flux partition Crop coefficient Citrus Central India

Accurate quantification of crop water demand and characterizing its temporal variability is essential in evaluating the role of water saving strategies for sustainable production. Representing evapotranspiration (ET) as a proxy to crop water needs is often misleading, particularly during the periods of high non-stomatal exchanges. This study is aimed at modeling crop water requirements for citrus orchards of central India using single and dual crop coefficient approaches. ET fluxes derived from eddy-covariance (EC) technique were used to develop single crop coefficient (K c ) curves at daily, weekly, and seasonal scales. Site-specific Kc values for initial, mid, and late season were found to be 0.43, 0.78, and 0.80 respectively. ET partitioning was done by estimating soil evaporation coefficient (K e ), and basal crop coefficient in the presence of water stress (Ks × K cb ) using soil-water balance (SIMDualKc) and EC flux partitioning (EC FP) methods. SIMDualKc model was calibrated against electrical resistivity tomography (ERT) derived soil water contents (R2 = 0.81, RMSE = 0.021 cm3 cm−3). Energy based flux partitioning was done by considering the correlation between high-frequency water vapour and carbon fluxes and applying flux variance similarity principles. Direct measurement of evaporation (E) and transpiration (T) at four citrus trees (using micro-lysimeters and sap flow meters) was used to assess the performance of two models. Three-stage basal crop coefficients from SIMDualKc and EC FP methods were found to be 0.18, 0.57, 0.63 and 0.26, 0.51, 0.59 respectively. Both methods were effectively partitioning the ET fluxes at daily scale (R2 > 0.5, RMSE < 0.6 mm, NSE > 0.3), however, variability in estimated fluxes between the two methods is high during the initial stage and gradually diminished with time.

1. Introduction Sustainable irrigation practices aim at improving crop productivity with manageable resources under changing the hydrologic environment (Saadi et al., 2015). This requires a critical evaluation of water saving strategies (including deficit irrigation, mulching, partial root zone drying, etc.) for improved crop water productivity (CWP), a measure of crop yield per unit of irrigation water. An accurate assessment of crop water requirements during the growth period can help to i) organize irrigation amount and timing, ii) maximize crop yield, and iii) minimize water losses and crop stress (Anderson et al., 2017). In semi-arid developing countries like India, optimal irrigation strategies will also help in the reduction of energy consumption, non-exploitation of groundwater resources, reduced weeding and associated labor costs (Priyan and Panchal, 2017). In eco-hydrology, crop water requirements are interchangeably used with evapotranspiration (ET), a measurable or estimated hydrological parameter. Reference ET (ET0) is the maximum possible ET that can occur from a well-watered, standard vegetation

⁎

surface for given meteorological conditions. United Nations Food and Agriculture Organization irrigation and drainage paper guidelines (FAO-56) are widely adopted in estimating the crop water requirements (Allen et al., 1998; Kashyap and Panda, 2001; Liu and Luo, 2010). FAO56 relates crop evapotranspiration (ETa ) with ET0 through crop coefficient (K c ), given by (Allen et al., 1998):

ETa = K c ∙ETo

(1)

Methods for estimating ET0 can be grouped into: i) mass transfer models such as Dalton, Meyer, Penman (Djaman et al., 2017; Valipour et al., 2017), ii) radiation models such as Priestley Taylor, Ritchie (Tabari et al., 2013), iii) temperature based models such as HargreavesSmani, Blaney-Criddle (Bormann, 2011; Muniandy et al., 2016; Valipour et al., 2017), and iv) pan evaporation models (Grismer et al., 2002; Zuo et al., 2016). Nevertheless, each method has advantages and application limitations in terms of data requirement, accuracy, simulation period, and global validity. Numerous studies across the world have concluded that FAO-56 Penman-Monteith method (Allen et al.,

Corresponding author. E-mail address: [email protected] (S.R. Peddinti).

https://doi.org/10.1016/j.agwat.2018.08.027 Received 6 April 2018; Received in revised form 19 August 2018; Accepted 20 August 2018 0378-3774/ © 2018 Elsevier B.V. All rights reserved.

Agricultural Water Management 212 (2019) 68–77

S.R. Peddinti, B.P. Kambhammettu

techniques. The water balance was performed separately for soil evaporation and root zone layers using SIMDualKc model. EC flux partitioning was done by cross-correlating stomatal (photosynthesis and transpiration) and non-stomatal (respiration and evaporation) exchanges of high-frequency CO2 and H2O concentrations using fluxvariance similarity theory. Intermodel comparison for dual crop coefficients was performed using residual statistical indicators. Applicability of the two models to partition ETa is further validated using the measured evaporation and transpiration data sets.

1998) that combines energy and mass balance approaches is the most accurate, reliable and standard method in estimating ET0 (Djaman et al., 2015; Pandey et al., 2016; Sentelhas et al., 2010; Tabari et al., 2011). Following Eq. 1, ETa can be estimated from ET0 using the scaling factor (i.e., crop coefficient, K c ) that accounts for soil, crop phenological and management conditions. FAO-56 uses two approaches to define this scaling factor, viz. i) single crop coefficient (K c ), that considers the integrated effect of soil evaporation (E) and plant transpiration (T) averaged over each growing season, and ii) dual crop coefficient that is further portioned into soil evaporation coefficient (K e ) and basal crop coefficient in the presence of water stress (K cb × K e ) (Allen et al., 1988). In spite of readily available crop coefficient curves defined for a number of crops considering standard and non-standard conditions, ETa estimation from FAO-56 guidelines suffers from the following drawbacks: i) season specific ETa data need to be adjusted and validated considering local agro-climatic conditions, ii) crop coefficients are averaged over each growth stage and hence neglect intra-seasonal variability, iii) influence of water saving strategies on the dynamics of ETa cannot be readily evaluated, and iv) ignorance of health of the tree can hamper management strategies when applied to diseased trees. Commonly used methods for quantification of ETa include: soil water balance (Bodner et al., 2007; Senay et al., 2011), bowen ratio energy balance (Inman-Bamber and McGlinchey, 2003; Malek, 1993), lysimeters (Liu and Luo, 2010; Miranda et al., 2006), and micro-meteorological techniques including EC (Er-Raki et al., 2009; Peddinti et al., 2018). Even though diurnal ETa variations are obtained, these methods can not partition ETa and hence limit their applications mostly to single crop coefficient analysis (Gowda et al., 2008). A number of studies have concluded that, dual crop coefficient technique outperforms its counterpart in estimating ETa for a wide range of cropping systems like maize, wheat, citrus, pecans (Er-Raki et al., 2009; Zhao et al., 2013; Ibraimo et al., 2016) under different water conservating strategies including drip irrigation, plastic or organic mulching, partial root zone drying, and deficit irrigation (Anderson et al., 2017). Direct measurement of E and T fluxes using micro-lysimetry (Liu et al., 2002) and sap flow techniques (Rana et al., 2005) are expensive and timeconsuming, hence limit their role to performance evaluation and validation of estimation methods. Of the available indirect methods for partitioning ET fluxes, methods based on micro-meteorological measurements and hydrological balance are more common (Zhao et al., 2015). Globally, India ranks fourth in orange production, accounting for about 11% of worlds’ tonnage. However, India ranks 64th in orange crop productivity (yield per unit area), accounting for 9.23 tons/ha. Vidarbha region in Maharashtra, central India is the leading producer of mandarin oranges (Citrus reticulata) accounting for 40% of country’s production with a yield of 6 tons/ha, far below the nation's average (Peddinti et al., 2018). Low crop yield in Vidarbha region is chiefly attributed to: i) improper management activities assisted with erratic rainfall patterns, and ii) propagation of a water mold disease ‘root rot gummosis’ (Phytophthora Spp.). The disease propagation has a strong correlation with soil moisture and temperature (Choudhari et al., 2018; Savita and Avinash, 2012). Irrigation in excess of crop water demand increases soil moisture within the rhizosphere, a condition favorable for further growth of disease-causing bacteria. Sub-optimal irrigation strategies result in increased water stress, thereby reducing root water uptake and crop yield. Hence, an accurate assessment of citrus crop water requirements is not only important from irrigation scheduling viewpoint, but also in managing and controlling the disease (Choudhari et al., 2018; Peddinti et al., 2018). This research is aimed at understanding the dynamics of crop water requirements in citrus orchards of central India. Tower based EC flux measurements were collected at high frequency for one crop cycle to develop diurnal and seasonal K c curves, and adjust the FAO-56 specified crop coefficients applicable to the region. Dual crop coefficients (K e and K cb × Ks ) were analyzed using water balance and flux partitioning

2. Materials and methods 2.1. Study area This study was conducted in the citrus orchards of Goregone village (latitudes: 21° 25′ 30.7″ to 21° 26′ 2.4″ E, longitudes: 78° 9′ 30.2″ to 78° 10′ 5.6″ N, elevation: 392 m asl) in the Vidarbha region of central India. As per Köppen-Geiger classification, the area falls under tropical savanna climate zone (Aw) characterized by lengthy dry months followed by short but extremely rainy wet months (Kottek et al., 2006). Mean annual precipitation in the region is about 900 mm with more than 70% occurring during monsoon (Jul-Sep). Mean annual ET0 is estimated to be 1500 mm. Average daily maximum temperatures (32 to 45 °C) occur during summer months (Mar-Jun), while average daily minimum temperatures (15 to 24 °C) occur during winter months (Dec-Feb). The humidity of the region varies from 35% in summer to 73% in monsoon (CGWB, 2013). Mean seasonal wind speed over the region is in the range of 1.5 to 2.7 m/s. Citrus crops in India are generally grown in three cycles: i) Ambia bahar (Feb flowering), ii) Mrig bahar (Jun flowering), and iii) Hast bahar (Oct flowering). Due to limited water supplements, farmers of this region prefer Mrig Bahar that demands less water during summer (initial stage) and more water during monsoon (growth stage). Citrus trees of the experimental fields are matured (8 years old) with 2.5 to 3 m height and 70% ground cover, healthy (free from Phytophthora spp.), and planted at 5 m spacing. Water requirement of citrus trees is generally met through flood system during flowering and early growth stages (for the ease with fertilizer application), and through drip system during late growth stage (due to limited resources). Irrigation is given at a frequency of 10–15 days with a ponding depth ranging from 6 mm to 40 mm depending on crop growth stage and antecedent precipitation/soil moisture conditions. The soil of the region is classified as vertisol having high water holding capability with rich in clay content (Peddinti et al., 2018). Hydro-geologically, the study area forms part of Deccan plateau characterized by multiple layers of solidified flood basalt resulting from volcanic eruptions. Groundwater is available under phreatic conditions with water drawn from upper weathered to fractured aquifers (Peddinti et al., 2016). Depth to groundwater in the study area is ranged from 12 m (premonsoon) to 6 m (post-monsoon) (CGWB, 2013; Central Ground Water Board, 2013). 2.2. Eddy covariance data collection and processing EC flux tower was installed to measure fast and slow response meteorological parameters during the crop cycle (DOY: 62–365, 2017). Carbon (CO2), water (H2O) fluxes were measured at 5 m height using an open path fast response infrared gas analyzer (IRGASON-EB-IC, Campbell Sci. Inc., USA) and a 3D sonic anemometer. Flux data was sampled at 10 Hz frequency and averaged over 30 min interval using a data logger (CR1000, Campbell Sci. Inc., USA). Additionally, slow response meteorological variables including precipitation (TE525-L-PTL, Tipping Bucket, Campbell Sci. Inc., USA), soil heat flux (HFP01SC-L-PTL, Campbell Sci. Inc., USA), solar radiation (CNR 4, Campbell Sci. Inc., USA), and soil moisture (CS616-L-PT-L, Campbell Sci. Inc., USA) were obtained at 30-min interval. Crop characteristics within the flux footprint were observed to be homogeneous (for at-least 500 m in the 69

Agricultural Water Management 212 (2019) 68–77

S.R. Peddinti, B.P. Kambhammettu

estimated following Eq. (1) on daily time step during the crop cycle. The daily K c values were averaged over each growth stage to represent seasonal K c values that are in line with FAO-56 specified coefficients. Additionally, weekly (7-day) averaged K c curves were developed for ease with the management and scheduling operations. 2.4. Dual crop coefficient approach Sustainable irrigation strategies in water-scarce regions aim at improving T at the expense of E and hence demand for the partition of ET fluxes for critical evaluation. Dual crop coefficient approach proposed by Allen et al. (1998) is accepted as a standard method to partition ET fluxes in the absence of measured data (Phogat et al., 2016). Under water-full condition, ETa from dual crop coefficient approach is given by (Allen et al., 1998):

ETa = (K cb + K e ) ETo

(3)

Fig. 1. Assessment of energy balance closure performed at the flux tower location considering day time flux measurements. 30-min. mean turbulent energy fluxes (LE+H) were regressed with available energy (Rn-G) over the citrus orchards for one crop cycle. The deviation of the linear fit from 1:1 line is attributed to the neglegence of energy sinks within the canpoy.

where K e is soil evaporation coefficient [-], and K cb is basal crop coefficient [-]. When the available soil water in the root zone drops below a critical level, crop water stress can occur to reduce ETa , given by:

prevailing wind direction), and hence the measured fluxes are considered to be representative of the study area. Raw fluxes were processed using Eddypro software (V. 6.2.0, LI−COR, USA) to obtain the corrected carbon and water fluxes at half-hour intervals. All standard corrections including data spike, coordinate rotation, tilt on sonic temperature for pressure and humidity, frequency response, and air density fluctuations were applied during the processing (Rodda et al., 2016). Gap filling and uncertainty analysis were performed for missing data periods using REddyProc package (Reichstein et al., 2016). Daily convective flux densities (LE and H) were obtained from the covariance of water vapour, and temperature perturbed with vertical wind velocity, sampled at 10 Hz frequency. Net radiation (Rn) and soil heat flux (G) were obtained at 30 min. interval and averaged for each day. Quality of eddy flux data was tested by performing energy balance closure considering daytime half-hour flux measurements. This is achieved by linearly regressing turbulent energy fluxes (H + LE) with available energy (Rn-G) and solving for the energy balance ratio (EBR). The EC tower had a reasonable energy closure error (R2 = 0.86) with a slope of 0.74 and EBR of 1.22 (Fig. 1). These quality indices are within limits commonly found with FLUXNET sites (Wilson, 2002; Conceição et al., 2017) and hence, the flux measurements are considered to be reliable during the crop period. Lack of perfect energy balance closure can be attributed to the ignorance of canopy heat storage and photosynthetic radiative energy (Er-Raki et al., 2009).

where Ks is crop water stress [-], that can be estimated either by considering root zone moisture depletion (Rosa et al., 2012) or by measuring soil suction at the root zone (Feddes et al., 1978). The intricacy in the estimation of E and T fluxes, and adjustment of individual coefficients to suit local agro-climatic conditions makes dual crop coefficient approach to be computationally intensive.

ETa = (Ks * K cb + K e )* ET0

2.5. (a) SIMDualKc model SIMDualKc model applies FAO-56 version of dual crop coefficient approach to partition ETa fluxes by considering non-standard conditions (Allen et al., 1998, 2005). The model can separate evaporation from the soil wetted by precipitation and irrigation, thus improving the accuracy of computations (Rosa et al., 2012). Soil water balance model in the upper root zone is performed at daily scale using a traditional bulk layer model (Allen et al., 1998, 2005):

Dr , i = Dr , i − 1−(P−RO)i−Ii−CRi + ETc + DPi

The FAO-56 Penman-Monteith combined equation (Allen et al., 1998; Monteith, 1965) based on aerodynamic theory and energy balance considering short (grass) as reference crop was used in estimating daily ET0 (mm/day), given by: 900

0.408Δ (Rn−G ) + γ T + 273 u2 (es−ea) Δ + γ (1 + 0.34u2)

(5)

where Dr is root zone depletion at the end of day ‘i', RO is surface runoff, I is irrigation depth, CR is capillary rise from groundwater table, ETa is actual crop ET, and DP is deep percolation to the bottom of root zone, all in length units. For the study region, return flows are insignificant and depth to water table is much below the root zone, and hence, RO and CR components were neglected. The model initiates with FAO-56 tabulated basal crop coefficients (K cb ), and progressively corrects K cb considering local climatic, soil, crop, and management conditions. Evaporation coefficient (K e ) is computed by applying daily water balance for the top soil layer that is characterized by depth and evaporable water (Pereira et al., 2015). Additionally, soil water stress (Ks) is computed by considering the amount of available water within the effective root zone (Allen et al., 1998, 2005). The model requires climate (Fig. 2), crop (phenology, initial crop coefficients, root zone characteristics, etc.), soil (Table 1) and management (irrigation scheduling, fraction wetted by irrigation) parameters. A number of studies have concluded that, when calibrated against soil water content (SWC) observations in field crops and orchards, the performance of SIMDualKc model can be significantly improved (Pereira et al., 2015; Wu et al., 2015). We conducted time-lapse three dimensional (3D) ERT at selective trees to capture root zone soil moisture profiles for use with model calibration. ERT data quality was improved by considering stacking and reciprocal measurements (Sreeparvathy et al., 2018). For a detailed description of SIMDualKc model description and working, the reader is advised to refer (Rosa et al., 2012, 2016).

2.3. Single crop coefficient approach

ET0 =

(4)

(2)

where, u2 is wind speed at a height of 2 m from the surface (m s−1), T is mean daily air temperature (°C), es is saturation vapour pressure at the surface (kPa), ea is actual vapour pressure of air above the evaporating surface (kPa), Rn is net radiation at the crop surface (MJ m-2 day−1), G is soil heat flux density (MJ m-2 day−1), Δ is slope of vapour pressure curve (kPa °C−1), and γ is psychometric constant (kPa °C−1). The corrected half-hour latent heat (LE) fluxes were averaged over each day to represent diurnal ETa (mm/day). Single crop coefficient (K c ) was 70

Agricultural Water Management 212 (2019) 68–77

S.R. Peddinti, B.P. Kambhammettu

Fig. 2. Seasonal variability in daily meteorological parameters at the experimental site during citrus crop cycle (DOY: 62–365, 2017). All parameters were measured at a height of 5 m using slow response sensors of the EC flux tower.

For a set of physically meaningful range of ρcp, cr (-1 ≤ ρcp, cr ≤ 0),

Table 1 Calibrated values for soil and crop parameters used with SIMDualKc model.

corresponding σc2p can be obtained. Individual components of carbon and water fluxes (w′c′ p , w′c′r , w′q′t , w′q′e ) are then obtained by considering day time EC flux measurements, using the following relations:

Parameter Description

Notation

Value

Units

Source

Depth of soil surface evaporation layer Total evaporable water Readily evaporable water Total available water Readily available water Ground cover fraction Root zone depth

Ze

0.1

m

Measured

TEW REW TAW RAW Fc z

25 7 172 82 0.7 0.6

mm mm mm mm – m

Surface soil field capacity Surface soil wilting point

θFC θWP

0.42 0.19

– –

Allen et al., (1998) Allen et al., (1998) Allen et al., (1998) Allen et al., (1998) Allen et al., (1998) Peddinti et al., (2018) Measured Measured

2

σp2

′ σc2p σc2p ⎛ w′q′e ⎞ ⎛ w q′e ⎞ 1 +2 σc2p + = 2 2 2 ⎜ w′q′ ⎟ WUE ρ WUE WUE 2 ⎜ w′q′ ⎟ cp, cr t ⎠ t ⎠ ⎝ ⎝

σc2 = σc2p +

w′q′t = 2.6. (b) EC FP model

⎜

−2σc2p WUE −2 ± 4σc4p WUE −4 − 4σc2p WUE −2ρc−p2, cr (σc2p WUE −2 − σq2)

w′c′ w′q′

2σc2p WUE −2ρc−p2, cr −2σc2p ± 4σc4p − 4σc2p ρc−p2, cr (σc2p − σc2) 2σc2p ρc−p2, cr

(7b)

w′q′ ′ ′ ⎛w q e ⎞ ′ ′

⎟

⎝wqt ⎠

We adopted flux variance similarity theory proposed by (Scanlon and Sahu, 2008) to convert high-frequency CO2 and H2O concentrations into individual stomatal and non-stomatal fluxes in the presence of leaf-level water use efficiency (WUE) (Scanlon and Sahu, 2008; Scanlon and Kustas, 2010). This theory is based on the fundamental hypothesis that: i) stomatal carbon and water fluxes are negatively correlated (ρcp,qt = -1) with slope equal to WUE, ii) non-stomatal carbon and water fluxes are positively correlated (ρcr,qe = 1), and iii) combined stomatal and non-stomatal fluxes that contribute to EC tower measurements are negatively correlated with slope deviating from WUE. The model first establishes the relation between photosynthetic carbon variance (σc2p ) and cross-correlation between stomatal and nonstomatal carbon fluxes (ρcp, cr ) using the equation:

WUE =

2 ⎛ w′c′r ⎞ ⎛ w′c′r ⎞ σcp + 2σc2p ⎜ w′c′ ⎟ ⎜ w′c′ ⎟ ρ 2 p ⎠ cp, cr p⎠ ⎝ ⎝

(7a)

+1

(8a)

w′c′

w′c′ p =

′ ′

⎛wcr ⎞ + 1 ′ ′ ⎝w c p ⎠ ⎜

⎟

(8b)

′

where, w , q′e , q′t , c′r , and c′ p denotes respectively the perturbation of vertical wind speed, evaporation, transpiration, respiration, and photosynthetic fluxes from their respective means. The overbar signifies the time-average (30 min.). Set of individual fluxes obtained for each combination of σc2p and ρcp, cr were narrowed down subjected to the constraints that are meaningful during daytime measurements. A more detail description of the analytical procedure is given in Scanlon and Sahu (2008). Following (Anderson et al., 2017), half-hourly partitioned fluxes (E and T) were averaged over each day to obtain the daily fractions of transpiration (FT) and evaporation (FE) for use with dual crop coefficient analysis given by:

+1

+1

j

FT =

(6) 71

∑i Ti j ∑i

j

Ei + ∑i Ti

(9a)

Agricultural Water Management 212 (2019) 68–77

S.R. Peddinti, B.P. Kambhammettu j

FE =

∑i Ei j ∑i

j

Ei + ∑i Ti

(9b)

Here, the indices ‘i' and ‘j’ denotes the first and last half-hour periods of a day for which EC FP program finds a solution. A threshold minimum of 10 periods (5 h) is considered in the calculation of daily FE and FT values. 2.7. Evaporation and transpiration measurements Daily evaporation from bare soil was measured using four microlysimeters installed within the flux tower fetch. Lysimeters are made of polyvinyl chloride (PVC) pipes having wall thinness of 2 mm and consist of two concentric open cylinders of length 20 cm with diameters of 15 cm and 18 cm. The inner cylinder is capable of moving in and out for sample collection, whereas the outer cylinder is permanently buried to prevent possible soil collapse (Facchi et al., 2017; Liu et al., 2002). Bottom of the inner cylinder is screened to prevent the loss of soil during sampling (Evett et al., 1995; Facchi et al., 2017). Cumulative daily soil evaporation from each lysimeter is estimated by gravimetric method using high precision electronic balance, and the average of four is reported. Plant transpiration was measured using SFM1 sap flow meters (ICT International, Armidale, NSW, Australia) that are placed in four citrus trees located within the flux tower fetch. Sap flow meter utilizes heat-ratio method (HRM) to measure high, low, and reverse sap flows using a set of three measurement needles having 35 mm stainless steel probes placed at 5 mm spacing (Barron-Gafford et al., 2017). We installed sap flow sensors at tree trunk above the ground surface and monitored the fluxes at 15 min. interval. A cumulative daily average of all four sap flow measurements was reported for use with analysis.

Fig. 3. Daily reference evapotranspiration (ET0 ) considering well-watered short grass at the experimental site during citrus crop cycle (DOY: 62–365, 2017).

(> 2 m/s) during crop development stage is replacing the evaporable layer with dry air thereby increasing VPD and hence ET0 . Temporal patterns of ET0 (Fig. 3) were observed to be similar to that of a semi-arid savanna climate found elsewhere (Er-Raki et al., 2009), characterized by high climatic demand in excess of precipitation. Annual average rainfall (∼ 500 mm) is much lower than average crop water requirement (i.e., K c x ET0 = 792 mm), indicating the necessity to irrigate citrus orchards to avoid water stress and reduced crop productivity (Er-Raki et al., 2009). During the crop period, daily ET0 was varied from 1.2 mm/day (harvest) to 8.1 mm/day (crop development) with an average of 4.3 mm/day and a total of 1300 mm. Maximum ET0 during crop development stage (DOY: 116 to 148) is in congruence with the period of highest radiation and air temperature. Seasonal dynamics conclude that ET0 is rapidly increasing during initial and crop development stages, followed by a gradual decrease towards the harvest.

2.8. Inter-model comparison and performance evaluation Performance of SIMDualKc and EC FP models to bifurcate ET fluxes is evaluated using goodness-of-fit indicators including coefficient of determination (R2), root mean squared error (RMSE), and index of agreement (di) (Qiu et al., 2015). Direct measurement of E and T fluxes using micro-lysimeters and sap flow measurements at four tree locations were used in performance evaluation. Additionally, SIMDualKc derived soil moistures were compared with ERT derived moistures under water unlimited (after irrigation) and water stress (before irrigation) conditions.

3.2. Dynamics of single crop coefficient Single K c values published in FAO-56 for citrus orchards (Allen et al., 1998) need to be adjusted to suit local agro-climatic conditions and account for management activities (Muniandy et al., 2016). Variability in K c at scales ranging from daily to weekly to seasonal are provided in Fig. 4. During initial and crop development stages, FAO-56 has significantly overestimated the crop coefficients. Low K c values during these periods can be due to conscious water stressing of the plants being practiced in the study region to initiate blooming. Despite a few rain spells, crop development stage has recorded lower K c during the development stage, due to high atmospheric demand (ET0 ). During mid and late seasons, FAO-56 has significantly underestimated the crop coefficients. High K c values are due to increased precipitation events and the highest leaf area that translates into maximum water use (Mahohoma, 2016). The percentage difference in seasonal K c ranges from 16% (mid-season) to 62% (initial stage). Also, we observed an increase in citrus crop coefficient during mid season (similar to annual crops), which is contradicting with FAO-56 published values. Specific to Mrig bahar, farmers of the region intentionally water stress the crops during summer months (initial stage), that helps in rapid blooming with fruit development and growth stages coinciding with the monsoon. Hence, the highest discrepancy in seasonal K c from FAO-56 was observed during initial to mid seasons. Overall deviation in K c throughout the crop cycle can be attributed to semi-arid tropical climate conditions, plant physiological characteristics and management activities practiced in the region. As on-farm management activities are being practiced at scales between daily to seasonal, we developed K c curves by considering weekly aggregates (Fig. 4). A wide range of seasonal crop coefficients in citrus trees have been reported by various researchers across the globe. In tropical to dry climates of USA, seasonal crop

3. Results and discussion This section first discusses the role of site-specific meteorological parameters on ET0 estimates. Diurnal and seasonal variations in K c obtained from EC flux data were discussed in relation to crop phenology. Applicability of soil water balance and EC FP models to partition ETa fluxes were evaluated and validated considering direct measurements of E and T. For ease with analysis and implementing management practices, we considered four growth seasons, viz. initial (DOY: 62 to 115), crop development (DOY: 116 to 148), mid-season (DOY: 149 to 325), and late season (DOY: 326 to 365). 3.1. Meteorological conditions and reference ET Weather data was collected at 30 min. interval using slow response sensors mounted at a height of 5 m and averaged for each day. Dynamics of key meteorological parameters contributing to ET0 during the crop period (Mar-Dec) are represented in Fig. 2. At a daily scale, both solar radiation and air temperature patterns are in synchronous with ET0 variability. The gap between RHmax and RHmin is high during the initial and crop development stages leading to high vapour pressure deficit (VPD) and thus the evaporative demand. A high wind speed 72

Agricultural Water Management 212 (2019) 68–77

S.R. Peddinti, B.P. Kambhammettu

Fig. 5. Daily variation in SIMDualKc simulated soil water content (SWC) within the root zone, along with ERT derived (observed) SWC for slective dates used in model calibration.

seasonal averages adjusted to local agro-climatic conditions resulting from SIMDualKc and EC FP models are presented in Fig. 6. Three-stage Kcb values from SIMDualKc and EC FP methods were found to be: 0.18, 0.57, 0.63, and 0.26, 0.51, 0.59 respectively. SIMDualKc simulated mid and late season Kcb curves are well in agreement with FAO-56 tabulated values, as the deviation from the standard conditions is minimal. The discrepancy between the two model results is mainly attributed to the fundamental difference in the methodology adopted and data requirement while partitioning ET fluxes. During initial and crop development stages, FAO-56 specified K cb values were significantly higher than the region-specific coefficients. This is due to the antithetical treatment conditions being followed within the study area. An increase in daily K e at the expense of K cb from SIMDualKc model during initial stage is in congruence with the high water stress experienced by the plants. As SIMDualKc model starts with FAO-56 dataset and progressively updates to suit local conditions, their seasonal means were less departed from FAO-56 tabulated K c values. Accumulated ETa for the entire crop cycle from SIMDualKc and EC FP models are 1008 mm and 725 mm respectively. Low crop water requirement resulting from EC FP is attributed to the inability of model to find a solution to eq 9(a) and 9(b) for a few days of the crop period. In comparison to SIMDualKc, EC FP model has overestimated K cb during the initial stage (by 44%) and underestimated during mid and harvest stages (by about 11 and 7% respectively). The contribution of E and T fluxes to measured daily ETa values from the two methods is represented in Fig. 7. Maximum daily ETa from EC FP method (5.4 mm/day) is slightly lower than from SIMDualKc approach (5.8 mm/day). Partitioning results are deviating between the two methods, particularly during the initial stage. The fraction of E and T to total ET fluxes during the crop cycle using SIMDualKc and EC FP methods are 38%, 62%, and 33%, 66% respectively. The contribution of T to ET fluxes is significantly high during mid and harvest seasons. It can be observed that EC FP method was failed to obtain a solution during mid-season, resulting in few data gaps (DOY: 160 to 170, and 230 to 240). The basal crop coefficients obtained in this study (using SIMDualKc and EC FP methods) are in agreement with the published literature, except for the initial stage. Er-Raki et al., 2009 observed the three-stage basal crop coefficients for the citrus orchards in Morocco under flood and drip irrigation system to be 0.3, 05, 0.4 and 0.35, 0.55, 0.45 respectively. Alves et al., 2007 have been reported that the K cb for young lime orchards (Citrus latifolia Tanaka) were varying from 0.4 to 1.0 in the tropical to temperate climates of Brazil.

Fig. 4. Daily variation (dots) in single crop coefficient (Kc), and their weekly averages (error bars) during citrus crop cycle. FAO-56 specified (solid red line) and locally adjusted (solid blue line) 4-stage linear Kc curves are superimposed on the daily variations. (DOY: day of the year; WOY: week of the year).

coefficients during initial, mid, and late seasons were reported as 0.80 ± 0.22, 0.87 ± 0.14, and 0.91 ± 0.13 (Fares et al., 2008; Hoffman et al., 1982; Rogers et al., 1983; Snyder and O’Connell, 2007; Van Bavel et al., 1967). In tropical to temperate climates of Brazil, seasonal citrus crop coefficients were reported as 0.55 ± 0.24, 0.51 ± 0.18, and 0.62 ± 0.12 (Alves et al., 2007; Marin and Angelocci, 2011). In Mediterranean climates of Spain, seasonal crop coefficients were reported as 0.39 ± 0.13, 0.87 ± 0.17, and 0.91 ± 0.09 (Castel et al., 1987; Romero et al., 2006; Villalobos et al., 2009). Seasonal crop coefficients for citrus orchards obtained in this study (0.43, 0.78, 0.80) were in agreement with the published reports typical to tropical savanna climate and FAO-56 specified values, except for the initial growth stage.

3.3. Dynamics of dual crop coefficient Variation in SIMDulaKc simulated soil water content (SWC) within the root zone is represented in Fig. 5. Scanty rainfall and intentional water stressing of the crop during initial stages has resulted in low soil moisture, just above the wilting point. Following this period, water is readily available for plant growth, due to supplemental irrigation. Abundant rain during mid-season (DOY: 180–230) has resulted in SWC to be slightly more than field capacity, resulting in waterlogged conditions. Even though irrigation was ceased before harvesting, increased rain spells have prevented water stress conditions. Spatial distribution of ERT-derived soil moisture, lumped over the root zone was considered to represent the soil moisture observations (Peddinti et al., 2018). Throughout the crop period, simulated SWC is well in agreement with the observed moistures (R2 = 0.81, RMSE = 0.021 cm3 cm−3, n = 27) with a low bias (-2.13) of estimation. The residuals (Oi – Pi) of the estimated SWC were found to be small and stationary with a mean and variance of -0.008 and 0.0003 respectively. Diurnal variation in basal crop coefficients (K cb × Ks ) and their 73

Agricultural Water Management 212 (2019) 68–77

S.R. Peddinti, B.P. Kambhammettu

Fig. 6. Daily variation (dots) in basal crop coefficients (K cb ), their weekly averages (error bars), along with FAO-56 tabulated (solid black line) and adjusted (solid blue line) seasonal basal crop coefficient curves during citrus crop cycle considering SIMDaulKc approach (left) and EC FP approach (right).

indictors can be found in Pereira et al. (2015). The ability of two models (viz., SIMDualKc and EC FP) to bifurcate ETa fluxes is evaluated by comparing with field measurements (Fig. 8). SIMDualKc has outperformed EC FP model in simulating T fluxes during mid and late growth stages. Both models are well reproducing the measured E and T fluxes with R2 > 0.5, and D > 0.8. Bare soil evaporation and plant

3.4. Performance evaluation Goodness-of-fit indicators play a crucial role in evaluating the performance of various modeling strategies in replicating field conditions. Most of these indicators work on model simulated deviations (known as errors) for a set of observations. More description on goodness-of-

Fig. 7. Diurnal contribution of evapration (E) and transpiration (T) fluxes to crop evapotranspiration (ETc) during citrus crop cycle resulting from SIMDualKc model (top) and EC FP model (bottom). 74

Agricultural Water Management 212 (2019) 68–77

S.R. Peddinti, B.P. Kambhammettu

Fig. 8. Comparison of SimDualKc simulated (row: 1) and EC FP model estimated (row: 2) evaporation (col: 1) and transpiration (col:2) fluxes with field observations (direct measurements using micro-lysimmeter and sapflow meters) for selective dates.

( ) ET

were used to develop daily and seasonal K c = ETa variations. Region 0 specific seasonal K c values (K c _ ini : 0.43, K c _ mid : 0.78, K c _ end : 0.80) were slightly deviating from FAO-56 tabulated values, due to the divergent irrigation practices and varying climatic conditions. For dual crop coefficients, we have tested the applicability of FAO-56 specified soilwater balance (SIMDualKc) and EC FP models. SIMDualKc model simulated SWC was agreed well with ERT observations with an R2 of 0.76 and RMSE of 0.018 cm3 cm−3. Direct measurement of E and T fluxes (using micro-lysimeters and sap flow meter) at selective trees was used to evaluate the performance of two models using six goodness of fit indicators. Seasonal basal crop coefficients from SIMDualKc (K cb _ ini : 0.18, K cb _ mid : 0.57, K cb _ end : 0.63) and EC FP (K cb _ ini : 0.26, K cb _ mid : 0.51, K cb _ end : 0.59) models were marginally different. Conscious water stressing of the crop to initiate blooming during initial stages has resulted in low crop coefficients. Both methods are reliable in partitioning ETa fluxes. However, the ability of SIMDualKc to couple with a crop model is recommended for use in evaluating the role of water saving strategies.

Table 2 Goodness-of-fit indicators in evaluating the performance of SIMDualKc and EC FP models to partition ETa fluxes (Here ME = Mean error, MAE = Mean absolute error, RMSE = Root mean square error, NSE = Nash-Sutcliffe efficiency, D = Index of agreement, R2= Coefficient determination; all in mm/d). Goodness-of-fit indicator

ME MAE RMSE NSE D R2

SIMDualKc model

EC FP model

Evaporation

Transpiration

Evaporation

Transpiration

−0.01 0.26 0.35 0.37 0.85 0.53

0.18 0.25 0.35 0.35 0.83 0.56

−0.25 0.47 0.61 0.33 0.82 0.51

−0.42 0.5 0.62 0.31 0.83 0.65

transpiration were measured using micro-lysimeter and sap flow meter at four trees within the study area for use with model comparison. If the deviation in measured E and T fluxes is within 10% of cluster mean, the data is included in performance evaluation. A total of 6 performance indicators (Table 2) were considered in this study. It can be concluded that both methods are reliable in partitioning the ETa fluxes for the study region. However, ability of SIMDualKc to couple with a crop yield model can aid in evaluating the role of water saving strategies on crop water productivity.

Acknowledgement This research is carried as a part of ITRA-Water project that was supported by Media Lab Asia, Ministry of Electronics and Information Technology (MeitY), Government of India. References

4. Conclusions

Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop Evapotranspiration-Guidelines for Computing Crop Water Requirements-FAO Irrigation and Drainage Paper 56 300. FAO, Rome, pp. D05109 9. Allen, R.G., Pereira, L.S., Smith, M., Raes, D., Wright, J.L., 2005. FAO-56 dual crop

The objective of this study is to model crop water requirements and develop single and dual crop coefficients for the citrus orchards grown in central India. Tower based weather and micro-meteorological fluxes 75

Agricultural Water Management 212 (2019) 68–77

S.R. Peddinti, B.P. Kambhammettu

Mahohoma, W., 2016. Measurement and Modelling of Water Use of Citrus Orchards. Doctoral dissertation. University of Pretoria. Malek, E., 1993. Theoretical and Applied Climatology Comparison of the Bowen RatioEnergy Balance and Stability-Corrected Aerodynamic Methods for Measurement of Evapotranspiration. Theor. Appl. Climatol. 178, 167–178. Marin, F.R., Angelocci, L.R., 2011. Irrigation requirements and transpiration coupling to the atmosphere of a citrus orchard in Southern Brazil. Agric. Water Manage. 98, 1091–1096. https://doi.org/10.1016/j.agwat.2011.02.002. Miranda, F.R., Gondim, R.S., Costa, C.A.G., 2006. Evapotranspiration and crop coefficients for tabasco pepper (Capsicum frutescens L.). Agric. Water Manage. 82, 237–246. https://doi.org/10.1016/j.agwat.2005.07.024. Monteith, J.L., 1965. Evaporation and environment. Symp. Soc. Exp. Biol. https://doi. org/10.1613/jair.301. Muniandy, J.M., Yusop, Z., Askari, M., 2016. Evaluation of reference evapotranspiration models and determination of crop coefficient for Momordica charantia and Capsicum annuum. Agric. Water Manage. 169, 77–89. https://doi.org/10.1016/j.agwat.2016. 02.019. Pandey, P.K., Dabral, P.P., Pandey, V., 2016. Evaluation of reference evapotranspiration methods for the northeastern region of India. Int. Soil Water Conserv. Res. 4, 52–63. https://doi.org/10.1016/j.iswcr.2016.02.003. Peddinti, S.R., KBVN, P., YSN, T., 2016. Efficacy of electrical resistivity tomography in demarcating deccan trap aquifer system methodology site description. Indian Natioanl Groundwater Conference. Peddinti, S.R., Kambhammettu, B.V.N.P., Ranjan, S., Suradhaniwar, S., Badnakhe, M.R., Adinarayana, J., Gade, R.M., 2018. Modeling soil–water–disease interactions of flood-irrigated mandarin orange trees: role of root distribution parameters. Vadose Zone J. 17https://doi.org/10.2136/vzj2017.06.0129. 0. Pereira, L.S., Paredes, P., Rodrigues, G.C., Neves, M., 2015. Modeling malt barley water use and evapotranspiration partitioning in two contrasting rainfall years. Assessing AquaCrop and SIMDualKc models. Agric. Water Manage. 159, 239–254. https://doi. org/10.1016/j.agwat.2015.06.006. Phogat, V., Šimůnek, J., Skewes, M.A., Cox, J.W., McCarthy, M.G., 2016. Improving the estimation of evaporation by the FAO-56 dual crop coefficient approach under subsurface drip irrigation. Agric. Water Manage. 178, 189–200. https://doi.org/10. 1016/j.agwat.2016.09.022. Priyan, K., Panchal, R., 2017. Micro - Irrigation : An EFficient Technology for India’s sustainable Agricultural Growth 1. pp. 398–402. Qiu, R., Du, T., Kang, S., Chen, R., Wu, L., 2015. Assessing the SIMDualKc model for estimating evapotranspiration of hot pepper grown in a solar greenhouse in Northwest China. Agric. Syst. 138, 1–9. https://doi.org/10.1016/j.agsy.2015.05.001. Rana, G., Katerji, N., De Lorenzi, F., 2005. Measurement and modelling of evapotranspiration of irrigated citrus orchard under Mediterranean conditions. Agric. For. Meteorol. 128, 199–209. https://doi.org/10.1016/j.agrformet.2004.11.001. Rodda, S.R., Thumaty, K.C., Jha, C.S., Dadhwal, V.K., 2016. Seasonal variations of carbon dioxide, water vapor and energy fluxes in tropical Indian mangroves. Forests 7, 1–18. https://doi.org/10.3390/f7020035. Rogers, J., Jr, L.A., Calvert, D., 1983. Evapotranspiration from a humid-region developing Citrus grove with a grass cover. Am. Soc. Agric. Eng. 7. Romero, P., Navarro, J.M., Porras, I., Martinez, V., Botía, P., 2006. Deficit Irrigation and Rootstock-their Effects on Water Relations,vegetative Development, Yield, Fruit Quality and Mineral Nutrition of Orange. pp. 1537–1548. Rosa, R.D., Paredes, P., Rodrigues, G.C., Alves, I., Fernando, R.M., Pereira, L.S., Allen, R.G., 2012. Implementing the dual crop coefficient approach in interactive software. 1. Background and computational strategy. Agric. Water Manage. 103, 8–24. https:// doi.org/10.1016/j.agwat.2011.10.013. Rosa, R.D., Ramos, T.B., Pereira, L.S., 2016. The dual Kc approach to assess maize and sweet sorghum transpiration and soil evaporation under saline conditions: application of the SIMDualKc model. Agric. Water Manage. 177, 77–94. https://doi.org/10. 1016/j.agwat.2016.06.028. Saadi, S., Todorovic, M., Tanasijevic, L., Pereira, L.S., Pizzigalli, C., Lionello, P., 2015. Climate change and Mediterranean agriculture: impacts on winter wheat and tomato crop evapotranspiration, irrigation requirements and yield. Agric. Water Manage. 147, 103–115. https://doi.org/10.1016/j.agwat.2014.05.008. Savita, G.S.V., Avinash, N., 2012. Citrus diseases caused by Phytophthora species. GERF Bull. Biosci. 3, 18–27. Scanlon, T.M., Kustas, W.P., 2010. Partitioning carbon dioxide and water vapor fluxes using correlation analysis. Agric. For. Meteorol. 150, 89–99. https://doi.org/10. 1016/j.agrformet.2009.09.005. Scanlon, T.M., Sahu, P., 2008. On the correlation structure of water vapor and carbon dioxide in the atmospheric surface layer: a basis for flux partitioning. Water Resour. Res. 44, 1–15. https://doi.org/10.1029/2008WR006932. Senay, G.B., Leake, S., Nagler, P.L., Artan, G., Dickinson, J., Cordova, J.T., Glenn, E.P., 2011. Estimating basin scale evapotranspiration (ET) by water balance and remote sensing methods. Hydrol. Process. 25, 4037–4049. https://doi.org/10.1002/hyp. 8379. Sentelhas, P.C., Gillespie, T.J., Santos, E.A., 2010. Evaluation of FAO Penman-Monteith and alternative methods for estimating reference evapotranspiration with missing data in Southern Ontario, Canada. Agric. Water Manage. 97, 635–644. https://doi. org/10.1016/j.agwat.2009.12.001. Snyder, R.L., O’Connell, N.V., 2007. Crop coefficients for microsprinkler-irrigated, cleancultivated, mature Citrus in an arid climate. J. Irrig. Drain. Eng. 133, 43–52. https:// doi.org/10.1061/(asce)0733-9437(2007)133:1(43). Sreeparvathy, V., Kambhammettu, B.V.N.P., Peddinti, S.R., Sarada, P.S.L., 2018. Application of ERT, saline tracer and numerical studies to delineate preferential paths in fractured granites. Groundwater. https://doi.org/10.1111/gwat.12663. Tabari, H., Somee, B.S., Zadeh, M.R., 2011. Testing for long-term trends in climatic

coefficient method for estimating evaporation from soil and application extensions. J. Irrig. Drain. Eng. 131, 2–13. https://doi.org/10.1061/(ASCE)0733-9437(2005) 131:1(2). Alves, J., Folegatti, M.V., Parsons, L.R., Bandaranayake, W., Da Silva, C.R., Da Silva, T.J.A., Campeche, L.F.S.M., 2007. Determination of the crop coefficient for grafted “Tahiti” lime trees and soil evaporation coefficient of Rhodic Kandiudalf clay soil in Sao Paulo, Brazil. Irrig. Sci. 25, 419–428. https://doi.org/10.1007/s00271-0060057-5. Anderson, R.G., Alfieri, J.G., Tirado-Corbalá, R., Gartung, J., McKee, L.G., Prueger, J.H., Wang, D., Ayars, J.E., Kustas, W.P., 2017. Assessing FAO-56 dual crop coefficients using eddy covariance flux partitioning. Agric. Water Manage. 179, 92–102. https:// doi.org/10.1016/j.agwat.2016.07.027. Barron-Gafford, G.A., Sanchez-Cañete, E.P., Minor, R.L., Hendryx, S.M., Lee, E., Sutter, L.F., Tran, N., Parra, E., Colella, T., Murphy, P.C., Hamerlynck, E.P., Kumar, P., Scott, R.L., 2017. Impacts of hydraulic redistribution on grass–tree competition vs facilitation in a semi-arid savanna. New Phytol. 215, 1451–1461. https://doi.org/10. 1111/nph.14693. Bodner, G., Loiskandl, W., Kaul, H.P., 2007. Cover crop evapotranspiration under semiarid conditions using FAO dual crop coefficient method with water stress compensation. Agric. Water Manage. 93, 85–98. https://doi.org/10.1016/j.agwat.2007.06. 010. Bormann, H., 2011. Sensitivity analysis of 18 different potential evapotranspiration models to observed climatic change at German climate stations. Clim. Change 104, 729–753. https://doi.org/10.1007/s10584-010-9869-7. Castel, J.R., Bautista, I., Ramos, C., Cruz, G., 1987. Evapotranspiration and irrigation effeciency of mature orange orchards in Valencia (Spain). Irrig. Drain. Syst. Eng. 1, 205–217. https://doi.org/10.1007/BF01102930. Central Ground Water Board, 2013. Ground Water Information, Amaravati District, Maharashtra. CGWB, Faridabad, India. Choudhari, R.J., Gade, R.M., Lad, R.S., Adinarayana, J., Phanindra, K.B.V.N., 2018. Epidemiological Relations to Phytophthora Spp. Causing Citrus Root Rot in Nagpur Mandarin. pp. 406–417. Conceição, N., Tezza, L., Häusler, M., Lourenço, S., Pacheco, C.A., Ferreira, M.I., 2017. Three years of monitoring evapotranspiration components and crop and stress coefficients in a deficit irrigated intensive olive orchard. Agric. Water Manage. 191, 138–152. https://doi.org/10.1016/j.agwat.2017.05.011. Djaman, K., Balde, A.B., Sow, A., Muller, B., Irmak, S., N’Diaye, M.K., Manneh, B., Moukoumbi, Y.D., Futakuchi, K., Saito, K., 2015. Evaluation of sixteen reference evapotranspiration methods under sahelian conditions in the Senegal River Valley. J. Hydrol. Reg. Stud. 3, 139–159. https://doi.org/10.1016/j.ejrh.2015.02.002. Djaman, K., Balde, A.B., Rudnick, D.R., Ndiaye, O., Irmak, S., 2017. Long-term trend analysis in climate variables and agricultural adaptation strategies to climate change in the Senegal River Basin. Int. J. Climatol. 37, 2873–2888. https://doi.org/10.1002/ joc.4885. Er-Raki, S., Chehbouni, A., Guemouria, N., Ezzahar, J., Khabba, S., Boulet, G., Hanich, L., 2009. Citrus orchard evapotranspiration: comparison between eddy covariance measurements and the FAO-56 approach estimates. Plant Biosyst. 143, 201–208. https://doi.org/10.1080/11263500802709897. Evett, S.R., Warrick, a.W., Matthias, a.D., 1995. Wall material and capping effects on microlysimeter temperatures and evaporation. Soil Sci. Soc. Am. J. 59, 329. https:// doi.org/10.2136/sssaj1995.03615995005900020009x. Facchi, A., Masseroni, D., Miniotti, E.F., 2017. Self-made microlysimeters to measure soil evaporation: a test on aerobic rice in northern Italy. Paddy Water Environ. 15, 669–680. https://doi.org/10.1007/s10333-016-0566-7. Fares, A., Dogan, A., Abbas, F., Parsons, L.R., Obreza, T.A., Morgan, K.T., 2008. Water BAlance Components in a Mature 72https://doi.org/10.2136/sssaj2007.0167. Feddes, R.A., Kowalik, P.J., Zaradny, H., 1978. Simulations of Field Water Use and Crop Yield. John Wiley, New York. Gowda, P.H., Chavez, J.L., Colaizzi, P.D., Evett, S.R., Howell, T.A., Tolk, J.A., 2008. ET mapping for agricultural water management: present status and challenges. Irrig. Sci. 26, 223–237. https://doi.org/10.1007/s00271-007-0088-6. Grismer, M.E., Orang, M., Snyder, R., Matyac, R., 2002. Pan evaporation to reference evapotranspiration conversion methods. J. Irrig. Drain. Eng. 128, 180–184. https:// doi.org/10.1061/(ASCE)0733-9437(2002)128:3(180). Hoffman, G.J., Oster, J.D., Alves, W.J., 1982. Evapotranspiration of mature orange trees under drip irrigation in an arid climate (Citrus sinensis). Trans. - Am. Soc. Agric. Eng. 25, 992–996. Ibraimo, N.A., Taylor, N.J., Steyn, J.M., Gush, M.B., Annandale, J.G., 2016. Estimating water use of mature pecan orchards: a six stage crop growth curve approach. Agric. Water Manage. 177, 359–368. https://doi.org/10.1016/j.agwat.2016.08.024. Inman-Bamber, N.G., McGlinchey, M.G., 2003. Crop coefficients and water-use estimates for sugarcane based on long-term bowen ratio energy balance measurements. For. Crop Res. 83, 125–138. https://doi.org/10.1016/S0378-4290(03)00069-8. Kashyap, P.S., Panda, R.K., 2001. Evaluation of evapotranspiration estimation methods and development of crop-coefficients for potato crop in a sub-humid region. Agric. Water Manage. 50, 9–25. https://doi.org/10.1016/S0378-3774(01)00102-0. Kottek, M., Grieser, J., Beck, C., Rudolf, B., Rubel, F., 2006. World map of the KöppenGeiger climate classification updated. Meteorol. Zeitschrift 15, 259–263. https://doi. org/10.1127/0941-2948/2006/0130. Liu, Y., Luo, Y., 2010. A consolidated evaluation of the FAO-56 dual crop coefficient approach using the lysimeter data in the North China Plain. Agric. Water Manage. 97, 31–40. https://doi.org/10.1016/j.agwat.2009.07.003. Liu, C., Zhang, X., Zhang, Y., 2002. Determination of daily evaporation and evapotranspiration of winter wheat and maize by large-scale weighing lysimeter and microlysimeter. Agric. For. Meteorol. 111, 109–120. https://doi.org/10.1016/S01681923(02)00015-1.

76

Agricultural Water Management 212 (2019) 68–77

S.R. Peddinti, B.P. Kambhammettu

Wilson, K., 2002. Energy balance closure at FLUXNET sites. Agric. For. Meteorol. 113, 223–243. https://doi.org/10.1016/S0168-1923(02)00109-0. Zhao, N., Liu, Y., Cai, J., Paredes, P., Rosa, R.D., Pereira, L.S., 2013. Dual crop coefficient modelling applied to the winter wheat-summer maize crop sequence in North China Plain: basal crop coefficients and soil evaporation component. Agric. Water Manage. 117, 93–105. https://doi.org/10.1016/j.agwat.2012.11.008. Zhao, P., Li, S., Li, F., Du, T., Tong, L., Kang, S., 2015. Comparison of dual crop coefficient method and Shuttleworth-Wallace model in evapotranspiration partitioning in a vineyard of northwest China. Agric. Water Manage. 160, 41–56. https://doi.org/10. 1016/j.agwat.2015.06.026. Zuo, H., Chen, B., Wang, S., Guo, Y., Zuo, B., Wu, L., Gao, X., 2016. Observational study on complementary relationship between pan evaporation and actual evapotranspiration and its variation with pan type. Agric. For. Meteorol. 222, 1–9. https:// doi.org/10.1016/j.agrformet.2016.03.002.

variables in Iran. Atmos. Res. 100, 132–140. https://doi.org/10.1016/j.atmosres. 2011.01.005. Tabari, H., Grismer, M.E., Trajkovic, S., 2013. Comparative analysis of 31 reference evapotranspiration methods under humid conditions. Irrig. Sci. 31, 107–117. https:// doi.org/10.1007/s00271-011-0295-z. Valipour, M., Gholami Sefidkouhi, M.A., Raeini−Sarjaz, M., 2017. Selecting the best model to estimate potential evapotranspiration with respect to climate change and magnitudes of extreme events. Agric. Water Manage. 180, 50–60. https://doi.org/10. 1016/j.agwat.2016.08.025. Van Bavel, C.H.M., Newman, J.E., Hilgeman, R.H., 1967. Climate and estimated water use by an orange orchard. Agric. For. Meteorol. 4, 27–37. Villalobos, F.J., Testi, L., Moreno-Perez, M.F., 2009. Evaporation and canopy conductance of citrus orchards. Agric. Water Manage. 96, 565–573. https://doi.org/10.1016/j. agwat.2008.09.016.

77