Solar radiation is an important parameter in the fields of computer modeling,engineering technology and energy development.This paper evaluated the ability of three machine learning models,i.e.,Extreme Gradient Boosting(XGBoost),Support Vector Machine(SVM)and Multivariate Adaptive Regression Splines(MARS),to estimate the daily diffuse solar radiation(Rd).The regular meteorological data of 1966-2015 at five stations in China were taken as the input parameters(including mean average temperature(Ta),theoretical sunshine duration(N),actual sunshine duration(n),daily average air relative humidity(RH),and extra-terrestrial solar radiation(Ra)).And their estimation accuracies were subjected to comparative analysis.The three models were first trained using meteorological data from 1966 to 2000.Then,the 2001-2015 data was used to test the trained machine learning model.The results show that the XGBoost had better accuracy than the other two models in coefficient of determination(R2),root mean square error(RMSE),mean bias error(MBE)and normalized root mean square error(NRMSE).The MARS performed better in the training phase than the testing phase,but became less accurate in the testing phase,with the R2 value falling by 2.7-16.9%on average.By contrast,the R2 values of SVM and XGBoost increased by 2.9-12.2%and 1.9-14.3%,respectively.Despite trailing slightly behind the SVM at the Beijing station,the XGBoost showed good performance at the rest of the stations in the two phases.In the training phase,the accuracy growth is small but observable.In addition,the XGBoost had a slightly lower RMSE than the SVM,a signal of its edge in stability.Therefore,the three machine learning models can estimate the daily Rd based on local inputs and the XGBoost stands out for its excellent performance and stability.
Jianhua DongLifeng WuXiaogang LiuCheng FanMenghui LengQiliang Yang
Traditional water and fertilizer inputs are often much higher than the actual demands of tomato,which causes a reduction in water-and fertilizer-use efficiencies.To investigate the advantage of alternate partial root-zone irrigation(AI)on water-and nitrogen(N)-use efficiencies of tomato modified by water and N management,taking conventional irrigation(CI)as the control,the effects of AI on root morphology and activity,fruit yield and water and N use efficiency were studied using pot experiments.There were four combinations of irrigation levels and growing stages of tomato for AI,i.e.AI_(1)(high water(W_(H))from blooming to harvest stage(BHS)),AI_(2)(W_(H)from blooming to fruit setting stage(BFS)and low water(W_(L))at the harvest stage(HS)),AI_(3)(W_(L)at BFS and W_(H)at HS)and AI_(4)(W_(L)at BHS)at three urea rates,i.e.low urea rate(NL),middle urea rate(N_(M))and high urea rate(N_(H))in the form of urea.Irrigation quotas for W_(H)and W_(L)in AI at BFS or HS were 80%and 60%of that in CI,respectively.Compared to CI,AI decreased water consumption by 16.0%-33.1%and increased water use efficiency of yield(WUE_(y))and dry mass(WUE_(d))by 6.7%-11.9%and 10.2%-15.9%,respectively.AI_(1)did not decline yield,total N uptake(TNU)and N use efficiency(NUE)significantly.Compared to NL,N_(M)enhanced tomato yield,TNU,WUE_(y)and WUE_(d)by 28.5%,35.3%,22.6%and 16.3%,respectively.Compared to CINL,AI_(1)N_(M)reduced water consumption by 12.5%,but increased tomato yield,TNU,WUE_(y)and WUE_(d)by 35.5%,58.4%,54.4%and 53.7%,respectively.Therefore,AI_(1)can improve water use efficiency and total N uptake of tomato simultaneously at medium urea rate.
Liu XiaogangLi FushengZhang FucangCai HuanjieYang Qiliang
The objective of this study was to obtain the water-saving and efficient production mode of Arabica coffee. The effects of three drip irrigation modes,conventional drip irrigation( CDI),alternate drip irrigation( ADI) and fixed drip irrigation( FDI) on growth,photosynthetic characteristics,biomass accumulation and irrigation water use efficiency of Arabica coffee were investigated under three nitrogen levels,high nitrogen( NH),middle nitrogen( NM) and low nitrogen( NL). The results show that there was a significant Logistic curve between the plant height,the stem diameter of Arabica coffee and growth days. Compared with CDI,ADI had no significant effects on leaf net photosynthetic rate,stomatal conductance,instantaneous water use efficiency and biomass accumulation above ground of Arabica coffee,while FDI decreased significantly,ADI and FDI increased irrigation water use efficiency by 50. 59% and 32. 85%,respectively. Compared with NH,with the reduction of N application rate,net photosynthetic rate,stomatal conductance,biomass accumulation above ground and irrigation water use efficiency decreased by 6. 81%-12. 30%,13. 70%-22. 69%,9. 61%-16. 67% and 9. 78%-15. 64%,respectively. Compared with CDINH,ADINHdecreased net photosynthesis rate and the stomatal conductance not significantly,other treatments decreased by 9. 16%-19. 22%,14. 49%-32. 91%,and decreased biomass accumulation above ground by 8. 26%-27. 34% except ADINH,and increased irrigation water use efficiency by 16. 46%-60. 95% except CDINMand CDINL. Therefore,alternate drip irrigation under high N level( ADINH) is the best water and nitrogen coupling mode of young Arabica coffee tree for water efficiency.