| Although China is a resource-deficit country,the phenomenon of wasting irrigation water in vegetable cultivation is extremely serious.Accurate estimation of reference crop evapotranspiration(ETO)can provide theoretical support for developing irrigation strategies and water-saving agriculture for vegetables in the facility.In this study,in view of the current serious problem of vegetable wastage in greenhouse cultivation and the lack of systematic evaluation of greenhouse crop ETO model,the autumn quail eggplant was used as a test crop to study the evapotranspiration of greenhouse autumn quail eggplant.Eight models were used and modified,including 3 Comprehensive model models FA056 Penman-Monteith(FA056),1963 Penman(Pen 63)and FA01979 Penman(FA079),three radiation models Priestley-Taylor(PT),DeBruin-Keijman(DK),and Makkink(Mak)and 2 The temperature model Mcloud(Mcl)and Hargreaves(Har)estimated the daily reference evapotranspiration of the autumn quail eggplant in the greenhouse,and verified the results at each growth period with the measured value of the weighed lysimeter.To obtain the best estimation method for daily reference evapotranspiration for each growth stage of greenhouse eggplant Get the following main results1 The cumulative evapotranspiration during the whole growth period of greenhouse autumn quail eggplant based on the measured value of the lysimeter was 254.91 mm,in which the cumulative transpiration and soil evaporation were 135.00 mm and 119.54 mm,which accounted for 53%and 47 of the total volume respectively.%.The environmental factors except the mean relative humidity were significantly negatively correlated with plant transpiration,soil evaporation and evapotranspiration(P<0.01),and other environmental factors were significantly positively correlated with surface evaporation(P<0.01).Photosynthetically active radiation,temperature,and saturation difference were the main influencing factors and showed good correlation with surface evaporation2 Based on the dynamic changes of the whole growth period,the estimation effects of each model are ranked in order:comprehensive method>radiation method>temperature method.The three comprehensive methods overestimated the mean value of 0.32-0.72 mm/d,of which FA056 performed best,followed by Pen63,and FA079 had the worst performance.The higher the temperature,the more overestimated the more obvious the maximum daily deviation was 8.66 mm/d.The overall value of the French model is underestimated.The underestimated range is 0.40-0.76 mm/d.The higher the temperature,the more underestimated the phenomenon is.The two temperature methods are obviously overestimated.The overestimation range is 2.11-6.43 mm/d,and the temperature is lower.The more serious the problem of overestimation,the greater the gap between Har and the measured values during the whole growth period The calculated value of the Mcl model has a large gap between actual evapotranspiration and the actual evapotranspiration at the early stage of childbirth.It is consistent with the measured value change trend after the maturity period begins.3According to the differences in cumulative evapotranspiration at each growth stage,the integrated model had the smallest gap at the seedling stage at higher temperatures.The cumulative water requirement of greenhouse eggplant at seedling stage was 86.58 mm,and the three comprehensive model estimates ranged from 94.51.-105.28mm,the difference range is only 7.73-18.70mm;the three radiation methods have better estimation effect in the lower temperature mature period,the actual measurement value is 95.19 mm,the estimated value range is 77.50-110.25mm,the difference range is 10.70-17.69 mm;in the two temperature methods,the minimum difference between the Har estimation and the measured value at the seedling stage is estimated to be 227.13 mm with a difference of up to 140.55 mm;the Mcl model has the smallest gap between the mature and measured values.It is 89.21mm,which is 5.98mm different from the measured value.The comprehensive model has the best performance in the seedling stage,and can be used to calculate the water requirement at the seedling stage.It can be corrected for the lower flowering fruit setting and maturity in the future;the radiation method can be used to estimate the effect of maturation at low temperature and radiation.Better,the pre-fertility estimation results are not good,and the difference between the calculated temperature method and the actual evapotranspiration is greatest,and it is not suitable for direct application in greenhouses4 Based on the correlation R2 and the statistical eigenvalues RMSE and MBE,the rankings of the estimation effects of all models are:FA056>Pen63>FA079>D-K=P-T>Mak>Mcl>Har.The estimated value of the synthetic method is closest to the measured value,with the highest correlation and the smallest RMSE.The R2 is as high as 0.7306 and the RMSE is only 0.32 mm/d.The three synthetic method ranks are FA056>Pen63>FA079;three radiation estimates It is close to the measured values,but the correlation is low.The range of R2 is 0.0151-0.3837 and the RMSE is 0.40-1.76 mm/d.The gap between PT and DK estimates is very small and can be considered as equal.The order of consistency is:D-K=P-T>Mak.The correlation between the calculated and measured values of the temperature method is the worst,and the overestimation phenomenon is obvious.The R2 of Mcl and measured values is as high as 0.4139,but the RMSE is also high as 2.11 mm/d.The performance of the Har formula is poor.The R2 is only 0.0121 and the RMSE is 6.43 mm/d.The modified comprehensive method performs best in all models,followed by radiation,and the temperature method has the worst estimation.Among the three comprehensive methods,FA056 is often used as a calibration method,with obvious advantages in terms of correlation and precision,which can be used to estimate the amount of water in the greenhouse. |
PDF Full Text Request