Abstract
Studies on dual crop coefficient method in a greenhouse require accurate values of reference evapotranspiration (ETo). This study was conducted in a solar greenhouse at the experimental station of the Chinese Academy of Agricultural Sciences during 2015 and 2016. An automatic weather station was installed in the center of the same greenhouse to record weather parameters at 30-minute intervals. Five ETo models (Penman-Monteith, Penman, radiation, pan evaporation, and Priestley-Taylor) were employed, and their performance was evaluated using the dual crop coefficient method. The basal crop coefficient Kcb and soil evaporation coefficient Ke were adjusted according to the surrounding climate inside the greenhouse. Crop evapotranspiration (ETc) was continuously measured using sap flow system combined with microlysimeter in 2015 and weighing lysimeters in 2016. Daily ETo was simulated from the five models and compared with the measurements. Results show that the adjusted Kcb values were 0.15, 0.94, and 0.65 in 2015 and 0.15, 1.02, and 0.70 in 2016 at initial, midseason, and late-season, respectively. The Ke varies between 0.10 and 0.45 during the whole growth period. The ETc was ≈345 mm for drip-irrigated tomato in solar greenhouse at the whole growth stage. The radiation and pan evaporation models tend to overestimate ETo values. Results of the Penman-Monteith, Penman, and Priestley-Taylor models show comparatively good performance in estimating ETo. Considering the robustness and simplicity, the Priestley-Taylor was recommended as the first choice to estimate ETo of tomato grown in a solar greenhouse. This work can help farmers to optimize the irrigation scheduling based on an ETo model for solar greenhouse vegetables in northern China.
A solar greenhouse is an energy-saving greenhouse that consists of gables on both sides, a maintenance rear wall, a supporting skeleton, and covering materials. It is different from glass and plastic greenhouse in two ways. First, there is no heating system in a solar greenhouse. The heat in the daytime comes from solar radiation and in the evening comes from the rear wall and surface soil. Second, the top and south vents are used to regulate temperature and humidity according to the characteristics of crop growth. These simple-structured greenhouses have emerged as a favorite technique in northern China because it bridges the season gap in vegetable and fruit production (Wei and Sun, 2014). Tomato (Lycopersicon esculentum) has high nutritional value, rich in vitamins A, B1, B2, and C in carotene and calcium (Lupini et al., 2017). In recent years, greenhouse tomato cultivation has been becoming an important industry for the local economy. A large amount of water is needed to supply tomatoes cultivated in a greenhouse, especially in the flowering and fruit setting stages (Gong et al., 2017a). Deep groundwater has been extracted for irrigation to maintain tomato production. Long-term overexploitation of groundwater resources will lead to a series of environmental degradations (e.g., soil salinization and desertification) (Xia et al., 2016). Therefore, improving water use efficiency is an effective way to maintain long-term greenhouse development.
ETc is an important parameter for determining a scientific and rational irrigation schedule to improve water use efficiency (Qiu et al., 2017, 2019; Yan et al., 2017; Yuan et al., 2001). Establishing an ETc model is a simple method to get water requirements of crops, thus many greenhouse ETc models have been developed. For example, the FAO56 Penman-Monteith (PM) model is the most commonly used because of its simple structure and relatively high precision (Allen et al., 1998). However, its application was still limited due to less information on two forms of resistance (e.g., aerodynamic resistance and canopy resistance). The structure of PM model must be redefined if applied in a greenhouse. Aerodynamic resistance has usually been considered a function of wind velocity, but it can become infinite when wind velocity is close to zero. Zhang and Lemeur (1992) reported that the aerodynamic resistance can be calculated by using the heat transfer coefficient inside greenhouse, but this coefficient needs canopy temperature, which is difficult to obtain. Because the PM model was insensitive to the aerodynamic resistance, several studies treat it as a constant in different greenhouse types (Fernández et al., 2010, 2011; Gong et al., 2017b; Yan et al., 2018). For surface resistance, it is usually estimated using the ratio of stomatal resistance to the effective leaf area index (Allen et al., 1998). Gong et al. (2017b) developed a bulk surface resistance model by combining the restrictions of soil and canopy to water vapor transfer for estimating the tomato ETc in solar greenhouse. Yan et al. (2018) indicated that the establishment of the canopy resistance model was related to days after transplant and net radiation. However, there are arguments about whether these resistance models can be appropriated for the whole growth stages in a solar greenhouse.
The dual crop coefficient method has also been used widely to determine crop water requirements, especially in solar greenhouse (Gong et al., 2017c; Qiu et al., 2015). This method consists of three parts: the basal crop coefficient (Kcb), soil evaporation coefficient (Ke), and reference ETo. In most cases, the Kcb and Ke need to be adjusted according to surrounding environmental and crop factors (Allen et al., 1998, 2005). For example, Ding et al. (2013a) developed an improved dual crop coefficient method for better predicting crop transpiration through Kcb and soil evaporation through Ke. Qiu et al. (2013) investigated effect of planting density on the Kcb by introducing a density coefficient. The adjusted Kcb and Ke were also used to estimate the greenhouse tomato ETc, and obtained the Kcb values at the initial, midseason, and late-season stages (Gong et al., 2017c). In addition, Rosa et al. (2012a, 2012b) developed a SIMDualKc software based on dual crop coefficient method and applied it to make irrigation scheduling for rainfed and basin irrigated maize and furrow irrigated cotton. The SIMDualKc model was also calibrated and validated for estimating hot pepper ETc in a solar greenhouse (Qiu et al., 2015). We have more research on the crop coefficient, however, fewer studies focus on the ETo, resulting in limitation of the dual crop coefficient method in a solar greenhouse. It is thus necessary to evaluate several models for estimating greenhouse ETo. Doorenbos and Pruitt (1977) proposed four methods to calculate ETo in the FAO Irrigation and Drainage Paper No. 24 (FAO24), including the FAO24 Penman, radiation, pan evaporation, and Hargreaves models. Fernández et al. (2010) indicated that the Hargreaves and the radiation methods were recommended to calculate ETo in the plastic greenhouse due to simplicity and precision. However, these models are validated only in the Mediterranean climatic regions. Considering the differences of climatic region and greenhouse characteristics, the ETc simulation of solar greenhouse with natural ventilation cannot be well presented by current research results. Therefore, whether these models can be applied to semihumid continental climate in northern China remains to be validated. In addition, as a simplification of the Penman equation, the Priestley-Taylor model (Priestley and Taylor, 1972), was often used to calculate greenhouse ETo. In this model, the ETc process was governed by net radiation, air temperature, and pressure for a surface that extends over a large surface area. Accuracy of the Priestley-Taylor model mainly depends on parameter α, which is a function of environmental factors (such as leaf area index and solar radiation) (Sumner and Jacobs, 2005). Ding et al. (2013b) indicated that the α was a function of leaf area index, vapor pressure deficit, and soil moisture content. Many studies have shown that taking α as 1.26 can be applied to many vegetated areas (Valdés-Gómez et al., 2009; Szilagyi, 2014; Tongwane et al., 2017). However, how to determine the value of α in a solar greenhouse has not been defined yet.
As stated earlier, the purpose of this work was to determine the best model for estimating ETo in a solar greenhouse by comparing the PM, Penman, radiation, pan evaporation, and Priestley-Taylor models. The authors hope to apply it for greenhouses to develop an irrigation scheduling.
Materials and Methods
Experimental site and design
Site description.
The experiment was conducted at the experimental station of the Xinxiang Comprehensive Experimental Base of the Chinese Academy of Agricultural Sciences, located in the northern of Henan Province, China (lat. 35°9′ N, long. 113°5′ E, altitude 78.7 m). The experimental site was located in a warm temperate continental monsoon climate zone with a mean annual temperature of 14.5 °C. The materials of the solar greenhouse included a steel frame and covering with plastic film (0.2-mm-thick nondrop polyethylene sheet). The length and width of the solar greenhouse was 60 and 8.5 m, respectively, it was oriented east–west, without heating equipment and passively ventilated by opening the plastic film on the top and the south. The experimental soil of 0.0 to 1.0 m was silt loam, with a mean bulk density of 1.49 g·cm−3, wilting point water content, and field soil water capacity of 0.09 and 0.32 cm3·cm−3, respectively. The main characteristics are presented in Table 1.
Textural and basic soil hydraulic properties of the greenhouse experimental site.
The greenhouse was uniformly planted with tomato crops (Mill. cv. Jingding). The plot was 8.8 m2 (8.0 m long × 1.1 m wide) with 50 plants planted in two rows. There were 18 plots in total, and the planting density was 5.7 plants/m2. Tomato seedlings were transplanted on 10 Mar. 2015 and 9 Mar. 2016 and irrigated with a drip irrigation system (the distance between emitters was 30 cm, and flow rate was 1.1 L·h−1). The irrigation amount and frequency were determined on the basis of the accumulated water evaporation of the D20 pan placed above the canopy. Irrigation events were conducted when the cumulative pan evaporation reached 20 ± 2 mm. Twenty millimeters of water were applied by drip irrigation after transplanting to ensure the survival of seedlings. Fifteen times of irrigation were performed, and the amount of water was 282.4 and 280.7 mm in 2015 and 2016, respectively. The length of initial, development, midseason, and late-season stages are shown in Table 2. Similar agronomic management, such as fertilization, pollination, and pest control, was conducted throughout the study period.
Tomato planted at the study site and its length of growing stage.
Sample collection and analyses
Meteorological factors and pan evaporation.
An automatic weather station was installed in the center of the same greenhouse. Solar radiation (Rs), net radiation (Rn), air temperature (Ta), relative humidity (RH), and air speed (u) at 2.0 m aboveground level were constantly monitored. These sensors were radiometer (LI200X; Campbell Scientific Inc., Logan, UT), net radiometer (NR LITE2; Kipp and Zonen BV, Delft, The Netherlands), temperature and humidity recorder (CS215), and air velocity meter (Gill WindSonic, Lymington, UK), respectively. In addition, the soil heat flux (G) was measured using heat flux plates (HFP01; Hukseflux, Delft, The Netherlands) at 5 cm below the ground surface. All data were collected with a CR-1000 data logger (Campbell Scientific Inc.) every 10 s, and 30-min averages were calculated.
A D20 pan is used to measure the pan evaporation at 30 cm above the canopy surface. The rain gauge with an accuracy of 0.1 mm was used to measure pan evaporation at 8:00 am every day, and 20 mm of water was added after each measurement to ensure water quality.
Soil evaporation.
Soil evaporation within two rows was continuously measured from March to July in 2015 and 2016 using microlysimeters, which was made from galvanized iron. It consisted of an inner and outer cylinder, 10 and 12 cm diameter, respectively, and 15 cm height. To reduce measurement errors and ensure easy operation without destroying the soil structure, the following operations were performed: 1) the outer cylinder was embedded in the soil beforehand, keeping the top edge level with the soil surface; 2) the inner cylinder was pushed inside the soil and the base sealed with plastic foil after it was removed; 3) inner cylinder mass was weighed using an electronic balance with a precision of 0.1 g every day at 8:00 am; 4) inner soil was replaced at 2-d intervals and after irrigation; and 5) six repetitions were conducted.
Plant transpiration.
Sap flow system (Flow 32-1k system; Dynamax, Houston, TX) was used to measure plant transpiration continuously from 15 May to 15 July 2015 and 14 Apr. to 10 July 2016. To ensure the rationality of measurement and reduce errors, the following processing were conducted: 1) eight healthy plants consistent in leaf area index were randomly selected; 2) sap flow meters were fixed at 20 cm above ground to avoid the effect of soil radiation on probes; 3) the specifications were required to meet the requirements of the tomato stem diameter to ensure close contact between the probes and the stem; 4) in the first stage, plant transpiration was obtained by minus soil evaporation from crop ETc, which was measured by water balance approach in 2015 and weighing lysimeter in 2016. The plant transpiration data were collected every 15 min by a CR-1000 data logger (Campbell Scientific Inc.).
Evapotranspiration.
The ETc was measured from March to July 2016 using two weighing lysimeters (1.0 m long, 1.0 m wide, and 1.2 m deep), which were located in the same greenhouse. To obtain more accurate measurements, the following procedures were followed: 1) soil texture in the weighing lysimeter was the same as in the field; 2) six seedlings were randomly selected, and the planting pattern was the same as outside; and 3) plants were supported with bamboo poles to avoid interference from outside crops.
Soil moisture content and growth indexes.
Five ECH2O sensors (5TE; Decagon Devices, Inc., Pullman, WA) were used to continuously monitor volumetric soil moisture content, which were installed in the middle of two drippers on the same drip tape. The measured depths were 10, 20, 30, 40, and 60 cm respectively. All data were collected every 30 min using an EM50 data logger (Decagon Devices, Inc.). Three repetitions were conducted to improve measurement accuracy in each plot.
Growth indexes included tomato height, leaf area index, and root length. Thirty plants were sampled at the center to measure height and leaf area index at 7-d intervals. Detailed measurement methods can be found in Gong et al. (2017b). Root length was measured using a root auger at the initial, development, midseason, and late-season stages. Roots were removed every 10 cm until no roots was found. Two repetitions were conducted because of the heavy workload.
Models and parameters
Five reference ETo models were evaluated and compared in this work, including the FAO56 PM, FAO24 Penman, FAO24 radiation, FAO24 pan evaporation, and Priestley-Taylor models.
PM model.
FAO 24 Penman model.
FAO 24 radiation model.
FAO 24 pan evaporation model.
Priestley-Taylor model.
Dual crop coefficients method.
When calculating Ke, the soil field capacity and soil wilting point between 0 and 15 cm were measured to be 0.31 m3·m−3 and 0.10 m3·m−3. Other parameters, such as root-zone soil field capacity, root-zone soil wilting point, maximum root depth, average fraction of soil surface wetted by irrigation, and leaf senescence factor are 0.32 m3·m−3, 0.09 m3·m−3, 1.0 m, 0.35, and 0.2, respectively. Standard (initial) and calibrated Kcb at initial, midseason, and late-season stages, p depletion fractions, and soil evaporation parameters for the greenhouse tomato are presented in Table 3.
Standard (initial) and calibrated basal crop coefficients, p depletion fractions, and soil evaporation parameters for the greenhouse tomato.
Performance indicators
Results and Discussion
Variations of meteorological factors in greenhouse.
To reveal the variations of meteorological factors inside the solar greenhouse, solar radiation (Rs), water vapor pressure deficit (VPD), air temperature (Ta), and pan evaporation (Epan) were investigated primarily in 2015 and 2016.
Figure 1A shows that the daily Rs changed from 0.74 to 19.97 MJ·(m−2·d−1) with an average of 11.73 and 11.02 MJ·(m−2·d−1) in 2015 and 2016, respectively. The maximum and minimum Rs occurred in July and March, respectively. Daily VPD varied from 0.02 to 2.46 kPa with a mean of 0.74 kPa for the 2 study years (Fig. 1B). In contrast to Rs, VPD did not present a clear variation trend. However, Ta shows a similar trend with Rs, which often reached the peak value in July. Daily Ta changed from 10.87 to 31.75 °C with a mean of 22.90 °C for the 2 study years (Fig. 1C). The variation trend was the closest between Rs and Epan. It is noteworthy that the maximum daily Epan was 7.8 mm and the minimum was only 0.12 mm for the entire period (Fig. 1D). Many studies found that Epan was close to ETc. For example, McVicar et al. (2007) stated that monthly ETc can be represented by Epan. Zuo et al. (2016) indicated that there was a complementary relationship between Epan and ETc. Bian et al. (2016) reported that the correlation coefficients between the monthly ETc and Epan was 0.894, and an empirical evapotranspiration model based on the relationships between monthly ETc and the leaf area index, precipitation, and Epan was established.
Variations of (A) solar radiation (Rs), (B) water vapor pressure deficit (VPD), (C) air temperature (Ta), and (D) pan evaporation (Epan) during 2015 and 2016 in the study’s solar greenhouse.
Citation: HortScience horts 55, 2; 10.21273/HORTSCI14514-19
Crop coefficient curves.
Following the proposed segment approach in the FAO56 manual, the tomato growing phase was divided into four stages: initial, development, midseason, and late-season. The initial and midseason stages were characterized by horizontal line segments, whereas the development and late-season were characterized by rising and falling line segments, respectively. Figure 2 shows the variations of the basal crop coefficient, Kcb, and soil evaporation coefficient, Ke, which was adjusted according to the greenhouse environment in 2015 and 2016. Results show that the change trend was similar between adjusted Kcb and the standard Kcb, but there were small differences at the midseason and late-season stages. Detailed descriptions are as follows: at the initial stage, Kcb ini was not adjusted and reached a value of 0.15 based on the FAO56 manual. At midseason and late-season stages, Kcb mid and Kcb end had to be adjusted due to environmental differences inside the greenhouse (RHmin was not equal to 45%, u2 was not equal to 2 m·s−1). Meanwhile, Kcb mid and Kcb end were greater than 0.45. Thus, the values of adjusted Kcb mid and Kcb end were 0.94 and 0.65 in 2015 and 1.02 and 0.70 in 2016, respectively. Kcb mid was lower than the standard value (1.10), whereas Kcb end was between the standard value (0.6 to 0.8). Many studies have shown that the adjusted Kcb mid was often lower than the standard Kcb mid recommended by FAO56 manual. Similar studies can also be found in Qiu et al. (2015) and Gong et al. (2017c). The possible reasons for Kcb mid lower than the standard Kcb mid are the following: 1) higher humidity and lower air transfer in the solar greenhouse can result in lower Kcb values (Allen et al., 1998); and 2) the walkway was set at ≈30 cm, resulting in reduction of total plant transpiration, which lowered the Kcb (Qiu et al., 2013).
Variations of the basal crop coefficients (Kcb), soil evaporation coefficient (Ke) and standard Kcb for greenhouse tomato in 2015 (A) and 2016 (B).
Citation: HortScience horts 55, 2; 10.21273/HORTSCI14514-19
Ke was affected by topsoil moisture content and canopy coverage rate primarily (Rosa et al., 2012a). In the initial stage, because of the small LAI and large exposed surface area, Ke was between 0.11 and 0.41 in 2015 and between 0.07 and 0.43 in 2016. In the development stage, Ke changed from 0.02 to 0.35. With the increasing of canopy coverage ratio, Ke decreased gradually but increased after irrigation. At the midseason and late-season stages, Ke was between 0.07 and 0.19 in both study years. Normally, the higher topsoil moisture content, the larger value of the Ke (Gong et al., 2017c). Rosa et al. (2012b) also indicated that Ke was high when the topsoil was wet after irrigation and the canopy was small.
Evaluation of five reference evapotranspiration models.
Variations of tomato ETc during the entire growth period are presented in Fig. 3. Results show that daily ETc was smaller at the initial stage and increased gradually at the late-season stage. Daily ETc changed from 0.12 to 6.66 mm·d−1 with a mean of 2.82 mm·d−1 in 2015 and from 0.32 to 6.65 mm·d−1 with a mean of 3.01 mm·d−1 in 2016. The average daily ETc was 1.65 and 1.15 mm·d−1, 2.53 and 2.70 mm·d−1, 3.98 and 3.52 mm·d−1, 3.42 and 3.86 mm·d−1 at initial, development, midseason, and late-season stages in 2015 and 2016, respectively. With the increase of Rs and Ta, the water requirement of crops increases gradually. Niu et al. (2011) indicated that daily ETc of greenhouse cucumber was largest at full fruit stage under drip irrigation, and the amount of ETc at this stage accounts for 69.6% of total ETc. Wang et al. (2009) also indicated that the maximum daily ETc occurs at the peak fruit stage, when total ETc changes from 60.9 to 100.42 mm. Similar results were validated in greenhouse melon (Sensoy et al., 2007) and pepper (Qiu et al., 2013). The reasons for this phenomenon can be described as follows: high radiation with air temperature, and low humidity will lead to the increase of leaf transpiration rate and crop water requirement. In this case, crop roots need to constantly absorb water from the soil to meet the water demand of fruit expansion (Gong et al., 2017a; Qiu et al., 2013).
Dynamics of tomato evapotranspiration in solar greenhouse during the whole growth period in 2015 and 2016.
Citation: HortScience horts 55, 2; 10.21273/HORTSCI14514-19
The parameters b, c, Kp, and α from the radiation, Penman, pan evaporation, and Priestley-Taylor models were determined through simultaneous measurements of ETc and (Lycopersicon esculentum) coefficient using the inversing method. Here, the ratio of ETc to (Kcb+Ke) was the measured ETo. Comparison of daily crop ETo between the estimated and measured values in the solar greenhouse are presented in Fig. 4. Results show that five models can better estimate daily ETo, most estimated data are closely distributed around the 1:1 line. The performance of five models are as follows:
Comparison between daily values of estimated vs. measured evapotranspiration (ETo) in solar greenhouse. ETo values were estimated by FAO56 Penman-Monteith (A), FAO24 Penman (B), FAO24 radiation (C), FAO24 pan evaporation (D), and Priestley-Taylor (E). Each point corresponds with a daily ETo value from 2015 to 2016.
Citation: HortScience horts 55, 2; 10.21273/HORTSCI14514-19
The PM model was recommended as the standard method for calculating field crop ETo (Allen et al., 1998). However, it underestimated greenhouse tomato ETo by 10.1% over 2 study years, with the slopes of regression equation of 0.89, MAE of 0.45 mm·d−1, RMSE of 0.58 mm·d−1, and dl of 0.95 (Table 4). Many meteorological factors of the PM model should be measured, resulting in a decrease in precision. In addition, the parameter of aerodynamic resistance (ra) in the PM model is a function of wind velocity, and it will reach infinity when wind velocity is close to zero. Some studies indicated that ra oscillates between 100 and 500 s·m−1 in greenhouse (Bailey et al., 1993; Fernández et al., 2010, 2011; Gong et al., 2017b; Katsoulas et al., 2001). Villarreal-Guerrero et al. (2012) indicated that adjustment of the parameters ra and stomatal resistance limits the application of the PM model in a greenhouse under variable high-pressure fogging.
Summary of statistics from the comparison between predicted (P, mm·d−1) and measured (Q, mm·d−1) values of mean daily greenhouse ETo.
The Penman and Priestley-Taylor models were similar in estimating ETo, and slightly overestimated ETo by 8.3% and 6.7%, at two growth seasons with slopes of 1.08 and 1.06, MAE of 0.45 and 0.46 mm·d−1, RMSE of 0.59 and 0.60 mm·d−1 respectively. The correction parameter values of c for the Penman and α for the Priestley-Taylor methods were stable in the solar greenhouse with mean values of 1.10 and 0.55, respectively. Muniandy et al. (2016) indicated that the Penman model (c = 0.35) was found to be the best model for estimating daily ETo of bittergourd and chili in the field. Fernández et al. (2010) indicated that taking c as 0.96 was more suitable in the whitened greenhouse. For the Priestley-Taylor model, net radiation (Rn) and α are two necessary parameters, where α also needs to be adjusted according to surrounding environment. Ding et al. (2013b) indicated that α was a function of leaf area index, vapor pressure deficit, and soil moisture content, and the value was different at each growth stage. For example, Valdés-Gómez et al. (2009) reported that the Priestley-Taylor model predicted greenhouse tomato ETc with an error of 6.1% when the α value was 1.26. This parameter was often applied in the field crops (Szilagyi, 2014; Tongwane et al., 2017). Although the Priestley-Taylor model slightly overestimated the measured ETo, considering its robustness and simplicity, this model was used as the first choice to estimate crops ETo in a solar greenhouse.
The performance of the radiation and pan evaporation models was slightly worse in estimating greenhouse ETo; the correction parameter values of b for the radiation model and Kp for the pan evaporation model was 1.14 and 1.00, respectively. The radiation and pan evaporation models overestimated ETo by 14.1% and 14.4% at two growth seasons, with regression coefficients of 1.14 and 1.16. The reason for the large deviation can be attributed to the effect of solar radiation on leaf stomata. Except for solar radiation, the VPD was also an important factor affecting ETc that cannot be ignored in a greenhouse condition (Gong et al., 2017a). Bonachela et al. (2006) also indicated that solar radiation was the main environmental factor affecting greenhouse ETc, followed by VPD. The restrictive conditions of water entering the atmosphere from plants are related to VPD, and the effect of meteorological factors on latent heat flux is largely realized by changing the VPD (Zhang et al., 2016).
Conclusions
Total evapotranspiration for greenhouse tomato with drip irrigation was ≈345 mm during the entire growth period. The amount of ETc was the largest at the midseason stage, which accounted for 69.6% of total ETc. The Kcb ini was 0.15, and the adjusted Kcb mid were lower than the standard (0.94 for 2015 and 1.02 for 2016). Adjusted Kcb end was 0.65 and 0.70 in 2015 and 2016, respectively. Ke was mainly affected by surface coverage and soil moisture and varied between 0.10 and 0.45 throughout the growth period. Daily greenhouse tomato ETo was estimated by the PM (ra = 308 s·m−1), Penman (c = 1.1), radiation (b = 1.14), pan evaporation (Kp = 1.0), and Priestley-Taylor (α = 0.55) methods. The PM, Penman, and Priestley-Taylor were found to be a better methods than radiation and pan evaporation methods, although ETo was slightly underestimated by PM and slightly overestimated by the radiation and pan evaporation methods. Because it required fewer parameters and less precision, the Priestley-Taylor method is recommended as the first choice to estimate crop ETo for solar greenhouse. This work can not only be applied to other solar greenhouses with similar weather conditions, but can also help farmers develop irrigation scheduling for greenhouse vegetables.
Literature Cited
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. FAO, Rome, Italy
Allen, R.G., Pereira, L.S., Smith, M., Raes, D. & Wright, J.L. 2005 FAO-56 dual crop coefficient method for estimating evaporation from soil and application extensions J. Irrig. Drain. Eng. 131 2 13
Bailey, B.J., Montero, J.I., Biel, C., Wilkinson, D.J., Anton, A. & Jolliet, O. 1993 Transpiration of Ficus benjamina: Comparison of measurements with predictions of the Penman-Monteith model and a simplified version Agr. For. Meteorol. 65 229 243
Bian, Z.H., Gu, Y., Zhao, J., Panpan, Y., Li, Y.T., Zeng, C.F. & Wang, L.C. 2016 Simulation of evapotranspiration based on leaf area index, precipitation and pan evaporation: A case study of Poyang Lake watershed, China Ecohydrol. Hydrobiol. 13 44 53
Bonachela, S., González, A.M. & Fernández, M.D. 2006 Irrigation scheduling of plastic greenhouse vegetable crops based on historical weather data Irrig. Sci. 25 53 62
Ding, R.S., Kang, S.Z., Zhang, Y.Q., Hao, X.M., Tong, L. & Du, T.S. 2013a Partitioning evapotranspiration into soil evaporation and transpiration using a modified dual crop coefficient model in irrigated maize field with ground-mulching Agr. Water Mgt. 127 85 96
Ding, R.S., Kang, S.Z., Li, F.S., Zhang, Y.Q. & Tong, L. 2013b Evapotranspiration measurement and estimation using modified Priestley-Taylor model in an irrigated maize field with mulching Agr. For. Meteorol. 168 140 148
Doorenbos, J & Pruitt, W.O. 1977 Crop water requirements. FAO Irrigation and Drainage Paper No. 24. FAO, Rome, Italy
Fernández, M.D., Bonachela, S., Orgaz, F., Thompson, R., Lo’pez, J.C., Granados, M.R., Gallardo, M. & Fereres, E. 2010 Measurement and estimation of plastic greenhouse reference evapotranspiration in a Mediterranean climate Irrig. Sci. 28 497 509
Fernández, M.D., Bonachela, S., Orgaz, F., Thompson, R., Lo’pez, J.C., Granados, M.R., Gallardo, M. & Fereres, E. 2011 Erratum to: Measurement and estimation of plastic greenhouse reference evapotranspiration in a Mediterranean climate Irrig. Sci. 28 91 92
Gong, X.W., Liu, H., Sun, J.S., Ma, X.J., Wang, W.N. & Cui, Y.S. 2017a Variation of evapotranspiration in different spatial scales for solar greenhouse tomato and its controlling meteorological factors Transactions of the CSAE 33 8 244 250 (in Chinese)
Gong, X.W., Liu, H., Sun, J.S., Gao, Y., Zhang, X.X., Shiva, K.J., Zhang, H., Ma, X.J. & Wang, W.N. 2017b A proposed surface resistance model for the Penman-Monteith formula to estimate evapotranspiration in a solar greenhouse J. Arid Land 4 530 546
Gong, X.W., Liu, H., Sun, J.S., Ma, X.J., Wang, W.N. & Cui, Y.S. 2017c Modeling evapotranspiration of greenhouse tomato under different water conditions based on the dual crop coefficient method Chi. J. Appl. Ecolo. 28 4 244 250 (in Chinese)
Katsoulas, N., Baille, A. & Kittas, C. 2001 Effect of misting on transpiration and conductances of a greenhouse rose canopy Agr. For. Meteorol. 106 233 247
Lupini, A., Princi, M.P., Araniti, F., Miller, A.J., Sunseri, F. & Abenavoli, M.R. 2017 Physiological and molecular responses in tomato under different forms of N nutrition J. Plant Physiol. 216 17 25
McVicar, T.R., Niel, T.G.V., Li, L.T., Hutchinson, M.F., Mu, X.M. & Liu, Z.H. 2007 Spatially distributing monthly reference evapotranspiration and pan evaporation considering topographic influences J. Hydrol. 338 196 220
Monteith, J.L. 1965 Evaporation and environment Symp. Soc. Exp. Biol. 19 205 234
Muniandy, J.M., Yusop, Z. & Askari, M. 2016 Evaluation of reference evapotranspiration models and determination of crop coefficient for Momordica charantia and Capsicum annuum Agr. Water Mgt. 169 77 89
Niu, Y., Liu, H.L., Wu, W.Y. & Yang, S.L. 2011 Cucumber transpiration by large–scale weighting lysimeter in solar greenhouse Transactions of the CSAE 27 1 244 250 (in Chinese)
Priestley, C.H.B. & Taylor, R.J. 1972 On the assessment of surface heat and evaporation using large-scale parameters Mon. Weather Rev. 100 81 92
Qiu, R.J., Du, T.S., Kang, S.Z., Chen, R.Q. & Wu, L.S. 2015 Assessing the SIMDualKc model for estimating evapotranspiration of hot pepper grown in a solar greenhouse in Northwest China Agr. Syst. 138 1 9
Qiu, R.J., Song, J.J., Du, T.S., Kang, S.Z., Tong, L., Chen, R.Q. & Wu, L.S. 2013 Response of evapotranspiration and yield to planting density of solar greenhouse grown tomato in northwest China Agr. Water Mgt. 130 44 51
Qiu, R.J., Liu, C.W., Wang, Z.C., Yang, Z.Q. & Jing, Y.S. 2017 Effects of irrigation water salinity on evapotranspiration modified by leaching fractions in hot pepper plants Sci. Rep. 7 7231 7240
Qiu, R.J., Liu, C.W., Cui, N.B., Wu, Y.J., Wang, Z.C. & Li, G. 2019 Evapotranspiration estimation using a modified Priestley-Taylor model in a rice-wheat rotation system Agr. Water Mgt. 224 105755
Rosa, R.D., Paredes, P., Rodrigues, G.C., Alves, I., Fernando, R.M., Pereira, L.S. & Allen, R.G. 2012a Implementing the dual crop coefficient approach in interactive software. 1. Background and computational strategy Agr. Water Mgt. 103 8 24
Rosa, R.D., Paredes, P., Rodrigues, G.C., Fernando, R.M., Alves, I., Pereira, L.S. & Allen, R.G. 2012b Implementing the dual crop coefficient approach in interactive software: 2. Model testing Agr. Water Mgt. 103 62 77
Sensoy, S., Ertek, A., Gedik, I. & Kucukyumuk, C. 2007 Irrigation frequency and amount affect yield and quality of field-grown melon (Cucumis melo L.) Agr. Water Mgt. 88 269 274
Sumner, D.M. & Jacobs, J.M. 2005 Utility of Penman-Monteith, Priestley-Taylor, reference evapotranspiration, and pan evaporation methods to estimate pasture evapotranspiration J. Hydrol. 308 81 104
Szilagyi, J. 2014 Temperature corrections in the Priestley-Taylor equation of evaporation J. Hydrol. 519 455 464
Tongwane, M.I., Savage, M.J., Tsubo, M. & Moeletsi, M.E. 2017 Seasonal variation of reference evapotranspiration and Priestley-Taylor coefficient in the eastern Free State, South Africa Agr. Water Mgt. 187 122 130
Valdés-Gómez, H., Ortega-Farías, S. & Argote, M. 2009 Evaluation of the water requirements for a greenhouse tomato crop using the Priestley-Taylor Method Chil. J. Agr. Res. 69 3 11
Wang, Z.Y., Liu, Z.X., Zhang, Z.K. & Liu, X.B. 2009 Subsurface drip irrigation scheduling for cucumber (Cucumis sativus L.) grown in solar greenhouse based on 20 cm standard pan evaporation in Northeast China Scientia Hort. 123 51 57
Wei, R.J & Sun, Z.F. 2014 Development and perspective of research on microclimate of sunlight greenhouse in china J Northwest Univ. 42 139 150 (in Chinese)
Xia, J.B., Zhang, S.Y., Zhao, X.M., Liu, J.H. & Chen, Y.P. 2016 Effects of different groundwater depths on the distribution characteristics of soil-Tamarix water contents and salinity under saline mineralization conditions Catena 142 166 176
Yan, H.F., Zhang, C., Gerrits, M.C., Acquah, S.J., Zhang, H.N., Wu, H.M., Zhao, B.S., Huang, S. & Fu, H.W. 2018 Parametrization of aerodynamic and canopy resistances for modeling evapotranspiration of greenhouse cucumber Agr. Meteorol. 262 370 378
Yan, H.F., Zhang, C., Peng, G.J., Darko, R.O. & Cai, B. 2017 Modeling canopy resistance for estimating latent heat flux at a tea field in south China Exp. Agr. 54 563 576
Yuan, B.Z., Kang, Y.H. & Nishiyama, S. 2001 Drip irrigation scheduling for tomatoes in unheated greenhouses Irrig. Sci. 20 149 154
Zhang, L. & Lemeur, R. 1992 Effect of aerodynamic resistance on energy balance and Penman-Monteith estimates of evapotranspiration in greenhouse conditions Agr. Meteorol. 58 209 228
Zhang, B.Z., Xu, D., Liu, Y., Li, F.S., Cai, J.B. & Du, L.J. 2016 Multi-scale evapotranspiration of summer maize and the controlling meteorological factors in north China Agr. Meteorol. 216 1 12
Zuo, H.C., Chen, B.L., Wang, S.X., Guo, Y., Zuo, B., Wu, L.Y. & Gao, X.Q. 2016 Observational study on complementary relationship between pan evaporation and actual evapotranspiration and its variation with pan type Agr. Meteorol. 222 1 9