Monitoring The Cold Damages To Double-cropping Rice In The Middle And Lower Reaches Of Yangtze River

Cold damage is one of the major meteorological disasters affectting the growth, development and yield of double-cropping rice in the middle and lower reaches of Yangtze River. Therefore, it is vital to analyze the spatial and temporal distribution of cold damage in the double-cropping rice season, and to establish effective rice cold damage monitoring indicators and reliable schemes to diminish the adverse influences on rice production. To achieve this, daily meteorological data obtained from76meteorological stations during1961and2010, rice observation data from38agro-meteorological stations in1981～2010and field seeding experimental data in2011～2013in Nanjing were used in this paper. Geo-statistical methods, statistical analyzing approaches (e.g. correlation analysis, stepwise regression and principal component regression) and rice crop model ORYZA2000were employed to explore the influence of cold damages on double rice production and to establish comprehensive rice cold damage monitoring schemes for the study area. Meanwhile, the seeding experimental data were used to calibrate the ORYZA2000. After the calibration, reliability of ORYZA2000in simulating the influence of cold damages was examined, based on which the ORYZA2000was improved in this paper. Comsequently, the established monitoring scheme takes the improved ORYZA2000as the core component. The summarized results and conclusions are listed as follows:1. The study area is mainly suffered mild cold damages at a rate of0.75times per year per station for early rice, taking up55%of the whole region. The cold damage in the northeast part of the study area is more serious than that in the southwest part by1.1times per year. Cold damage to early rice has inter-annual and decadal fluctuation characteristics, which shows a highest occurrence in the1960s and a lowest occurrence at the beginning of the21st century. The frequency of cold damage has reduced significantly (-0.23times/decade, P<0.05) during the study period, and the moderate cold damage becomes the main cold damage type for the early rice growth. Furthermore, the periods of cold damages to early rice showes several obvious inter-annual and decadal cycle oscillation characteristics, but only the small scale of3-5years pass the red noise significant test with the significance level a=0.05.2. The study area is mainly suffered from mild cold damages at one year0.63times per station for late rice, taking up47%of the whole region. It shows roughly north-south distribution in space with the north occurring more. The inter-annual and decadal fluctuation characteristics on cold damage to late rice are obvious.1960s shows a highest occurrence and the beginning of the21st century shows a lowest occurrence. The frequency of cold damage has reduced significantly (-0.21times/decade, P<0.05) for late rice, and the mild cold damage reduces the most. Several inter-annual and decadal cycle oscillation characteristics of late rice’s cold damage are obvious, but only the scale of3years and7years are significant.3. Taking13℃as the base temperature, early rice needs300～390℃·d accumulated temperature during the transplanting to the panicle initiation stage. On average, days of the panicle initiation stage extend about6days when13℃based accumulated temperature reduces100℃·d in this stage. In addition, taking15℃as the base temperature, early rice needs195～235℃·d accumulated temperature during the panicle initiation to the heading stage. Days of the heading stage delay9days with accumulated temperature reducing100℃·d in this stage. Mild, moderate and severe growth-delayed cold damage indicators are3-6days,6-9days and>9days respectly. The most important climatic factors for the relative meteorological yields of early rice are precipitation, which are negatively correlated with yield. The established regression models all have significane with a=0.01.4. The optimum temperature for late rice panicle and yield formation is between23～28℃. Taking22℃as the cold accumulated temperature borderline indicator, mild, moderate and severe sterile-type cold damage indicators are2～8℃·d,8～12℃·d and^12℃·d respectively for the phase from panicle initiation to rice heading. Meanwhile, taking20℃as the cold accumulated temperature borderline indicator, mild, moderate and severe sterile-type cold damage indicators are2～16℃d,16～35℃·d and≥35℃·d respectively for the phase from heading to end of grain-filling stage. The most important climatic factors for the relative meteorological yields of the late rice are the cold accumulated temperature and the extreme low temperatures during the growth period from heading to the end of grain-filling stage. They are positively and negatively correlated with the yields, respectively. The established regression models pass the significant test (α=0.01). Indicators verification results agree well with the fact that shows indicators are reliable.5. REDFTT parameter values in ORYZA2000model were substituted by the response curve of photosynthetic rates to temperatures fitting by photosynthetic observation data from the climatic control experiment. Meanwhile, the empty-percent simulation module was added to improve and supplement the capability of ORYZA2000in cold damage simulation. Results from improved crop model show that simulation errors of LAI, ear and green leaf biomass decrease under low temperature stress. Besides, two low temperature scenarios are set to analyze the simulation uncertainty of Cold damage using improved ORYZA2000model. The result illustrates that improved ORYZA2000model can reflect the impact of low temperatures on growth, phenology and yield formation.6. The established monitoring scheme for rice cold damage was effective and reliable not only for site monitoring but also for reginal application. Within the monitoring scheme, the established cold damage indicators and improved ORYZA2000were coupled. The phenological delay of early rice during the transplanting to the panicle initiation stage, and the sterile-type cold damage during the panicle initiation to the heading stage, as well as the cold damages during the grain-filling stage of late rice can be monitored. The results indicated that the delay in the phase from transplanting to panicile initiation of early rice can be well estimated less than or equal to3d with RMSE about1.7d. According to the simulated empty-percent of late rice, the degrees of moderate and severe sterile-type cold damage were estimated with an accuracy around78%. On regional scale, NRMSE of actual empty-percent less than30%of late rice is26.4%. The simulation agrees well with the actual situation, which proved that the established monitoring scheme was effective in cold damage monitoring in the double-cropping rice season in the middle and lower reaches of Yangtze River.