Measurement And Partitioning Evapotranspiration:Application Of Sap Flow Sensors And Heat-pulse Sensors

Field evapotranspiration（ET）is an important part of soil surface energy balance and water balance in farmland ecosystem is mainly composed of plant transpiration and soil water evaporation.Accurate measurements of plant transpiration and soil water evaporation are of great significance for the guidance of field irrigation and improving water use efficiency.But now heat-balance sap-flow（HBSF）sensors which are widely used will overestimate the stem flow rate,and multi-needle heat-pulse sensors based on sensible heat balance theory can not obtain the E rate above 1.5 mm soil layer and its accuracy and precision are unclear.The main purpose of this study is calibrating HBSF sensors based on mass balance method,improving the calculation method of surface soil water evaporation,measuring E and T dynamic during whole growth period in maize field.Then,we apply multi-needle heat-pulse sensors and HBSF sensors to no-tillage,to further explore the effect of no-tillage on E and T in spring maize field.The main study findings are as follows:First,HBSF sensors measured sap flow velocity(V,g plant^-1 h^-1)overestimated transpiration rate(T,g plant^-1 h^-1),and the errors displayed diurnal dynamics:small in the evening and early morning,became larger with increasing transpiration rate,and reached a maximum(about 67 g plant^-1 h^-1 in the lysimeter experiment,154 g plant^-1 h^-1 in the pot study,and 30 g plant^-1 h^-1 in the open field study)when solar irradiance was the largest.On the basis of transpiration rate measured by weighing method,an empirical linear calibration equation relating T and V measured with HBSF sensors was established:T =0.65V+0.39.Using this equation,the largest sap flow error was reduced by 60%,50%,and 50%in the lysimeter experiment,pot experiment,and field study,respectively.The linear calibration equation improves maize sap flow accuracy with the HBSF sensors.Second,evaluate the accuracy and precision of the multi-needle heat-pulse technique using a weighing lysimeter（9m²）,and the results showed that heat-pulse technique can provide continuous daily E measurements with good accuracy and precision.Four multi-needle heat-pulse probes were installed in weighing lysimeter at four locations to measure soil thermal conductivity,heat capacity and temperature gradient of the 0-to 5-cm layer.Daily E rates were estimated with sensible heat balance theory and modified sensible heat balance theory.Potential daily evaporation（E0）rate was also computed using the Penman-Monteith equation.The E/E0 ratio was greater than 0,9 for the first four days,but fell sharply thereafter,and approached a stable value of 0.4.There was a linear relationship between heat-pulse results and lysimeter data,with a slope of 0.95 and an offset of 0.09 mm d^-1.The absolute error of the heat-pulse E rate ranged from-0.02 mm d^-1 to 0.29 mm d^-1(or^-1.17%to 8.49%of the true value),and the root-mean-square error was 0.15 mm d^-1.The standard deviation of the heat-pulse E data ranged from 0.07-0.51 mm d^-1 and the coefficient of variation varied from 3.30%to 14.45%.Third,combination of HBSF sensors and multi-needle heat-pulse sensors can obtain accurate ET.We used multi-needle heat-pulse probes and weighing lysimeter to measure the dynamic variation of E and ET rate in maize（Zea mays L.）field for the whole growth period.We calculated T rate following the FAO-56 approach in early-season which was unmeasureable with HBSF sensors and measured T rate using HBSF sensors in the late-season.Finally,evaluate the accuracy of the combination of the proposed method for measuring ET by comparison with weighing lysimeter.Results showed that the E,T and ET values ranged from 0.13 mm d^-1 to 5.10 mm d^-1,0 mm d^-1 to 8.67 mm d^-1 and 0.39 mm d^-1 to 10.11 mm d^-1 in the whole growth period,with the average value of 1.42 mm d^-1,2.87 mm d^-1 and 4.32 mm d^-1,respectively.The cumulative E value in the corn growth period reached 131.78 mm,accounting more than 30%of ET.Comparing the ET obtained by combination of proposed methods to ET monitored by weighing lysimeter,the results were in good correlation with r2 = 0.80.Finally,applying multi-needle heat-pulse sensors and HBSF sensors to two tillage methods of no-tillage and ploughing to compare E and T dynamic in spring maize field.The results showed that E and T rates ranged from 0.29 to 2.59 mm d^-1 and 0.35 to 6.86 mm d^-1 during the measurement period in no-tillage,E and T rates ranged from 0.10 to 4.15 mm d^-1 and 0.28 to 7.24 mm d^-1 in ploughing,respectively.The magnitude of E and T rates were mainly affected by solar radiation,soil water content and leaf area index together.Comparing with ploughing,no-tillage method could reduce 15 mm soil water evaporation and increase 22 mm plant transpiration during the measurement period,the non-productive soil water content reduced.