Rice Expansion And Its Effect On Near-surface Air Temperature And Evapotranspiration In Heilongjiang Province

In recent years,there has been an expansion of rice planting area in Heilongjiang Province,the crop planting structure of the Sanjiang Plain and Songnen Plain has been continuously adjusted,and crop water consumption and water resources management in the reclamation area have changed year by year.Changes in crop planting structure will affect the earth's environment(atmosphere,soil,hydrological cycle,etc.),among which near-surface temperature and evapotranspiration are the most basic observational parameters in the earth's environment.Therefore,objectively and accurately deriving the temporal and spatial variation of rice planting area,near-surface temperature and evapotranspiration in Heilongjiang Province from these parameters is of great significance to provincial-scale water resources management,ensuring food production safety,and promoting the harmonious development of the relationship between man and nature.Remote sensing technology has the advantage of continuity in time and space,and has been widely used in retrieving provincial-scale rice planting area,near-surface temperature and evapotranspiration.This paper takes Heilongjiang Province as the research object,and based on the surface reflectance data and surface temperature data of the Moderate Resolution Imaging Spectroradiometer(MODIS)from 2003 to 2018,the land type identification,near surface temperature inversion and daily evapotranspiration inversion.in Heilongjiang Province are carried out respectively.The decision tree classification model was established by means of integration threshold method,hierarchical classification,texture characteristics,supervision classification and index time series,and the forest,water body,towns,meadows,beaches,paddy fields and dry land were sequentially extracted.The spatial and temporal distribution of land use in Heilongjiang Province from 2003 to 2018 was qualitatively and quantitatively analyzed.The spatial distribution of the highest temperature and the lowest temperature in Heilongjiang Province was reconstructed using the Temperature-Vegetation Index(TVX)method,and qualitative and quantitative analysis of the effect of rice field expansion on the near-surface temperature of the whole year,the period of birth,summer and non-fertility period in 2003-2018;The surface temperature and vegetation index feature spaces are used to construct linear and polynomial dry and wet edge equations,and the Conditional Vegetation Temperature Index(VTCI)is got through calculation.The daily evapotranspiration was calculated based on the net surface radiant flux Rn and VTCI,and the influence of paddy field expansion on the daily evapotranspiration and total monthly evapotranspiration during the growth period was qualitatively and quantitatively analyzed.The main conclusions are as follows:1)Based on the phenological characteristics of different vegetations in Heilongjiang Province,the decision tree classification criteria for different land types are determined,which can accurately identify land use types at the provincial scale.From 2003 to 2018,the Kappa coefficient of rice field area identification in Heilongjiang Province was 0.899-0.961,and the total accuracy was 85.5%-92.3%,which can effectively meet the needs of remote sensing monitoring of rice field area.The area of rice fields in Heilongjiang Province increased linearly from 2003 to 2018,with an average annual expansion of 158 100 hm~2.The area of rice fields in 2018 was three times the area of rice fields in 2003.The center of the planting area extends northward by about 160 km.The transformation of meadows and beaches is mainly distributed in the Sanjiang Plain and Songnen Plain.2)According to the near-surface high temperature images in Heilongjiang Province during the transplanting and milking periods,the temperature in the Sanjiang Plain generally ranges from 10 to 20°C,and the temperature in the Songnen Plain ranges from 20 to 30°C.The high temperature zone during the transplanting period is in the northeast of the Songnen Plain,and the high temperature zone during the milk maturity period is in the northwest of the Songnen Plain.According to the images of the lowest surface temperature near the surface during the transplanting and milking periods in Heilong Province,the temperature in the Sanjiang Plain generally ranges from 0 to 11°C,and the temperature in the Songnen Plain ranges from 9 to 20°C.The low temperature zone during the transplanting period is in the east of the Sanjiang Plain,and the low temperature zone during the milking period is in the west and east of the Sanjiang Plain.In the time window of summer and rice growth period,the time series of near-surface temperature is significant within 90%confidence interval,and the cooling effect of rice planting area is more obvious.The cooling effect of the highest temperature in rice fields in different months is May>June>July>August>September.The cooling effect is related to the rice growth period and temperature changes.Only in June,the rice field area has the effect of cooling down throughout the day,while in May,August and September,the rice field area has an obvious minimum temperature insulation effect;3)According to the distribution of the highest temperature near the surface in the paddy field expansion area from 2003 to 2018,it can be seen that from May to August,the cooling effect of the paddy field expansion area in the Sanjiang Plain was the most obvious.In most areas,the temperature dropped from 32 to 40°C to 16 to 32°C.The central part also has a cooling effect with a smaller temperature difference.In October,the expansion area of the rice field has a thermal insulation effect,and most of the highest temperature areas near the surface increase from 0 to 16°C to 8 to 24°C.According to the distribution of the lowest temperature near the surface of the rice field expansion area from 2003 to 2018,it can be seen that there is a thermal insulation effect in the rice field expansion area in May,August and September,and the thermal insulation effect is most obvious in August.In August,most of the pixels in the lowest temperature near the surface image increased from 0 to 12°C to 12 to 24°C.From June to July,there is a cooling effect in the expansion area of the rice field,and the temperature difference of the same pixel ranges from 0 to 10°C.The influence of paddy field expansion on near-surface temperature is non-linear,and the change of near-surface temperature is related to the ratio of paddy field area.4)Using linear fitting and polynomial fitting respectively,the dry and wet edge equations are calculated and the feature space model is constructed,and the conditional vegetation temperature index VTCI is calculated from the feature space.Based on short-wave radiation,incident long-wave radiation and emitted long-wave radiation,calculate the net surface radiant flux Rn.The daily evapotranspiration from 2004 to 2006 is calculated from the above two parameters,and the daily average evapotranspiration data from Sanjiang Station is used to verify that the characteristic space calculated by polynomial fitting is better than the characteristic space calculated by linear fitting.Based on the feature space fitted by the polynomial,the conditional vegetation temperature index VTCI from 2003 to 2018 was calculated.The daily evapotranspiration of crops varies greatly.Dryland areas have greater evapotranspiration in September,and paddy fields have greater evapotranspiration in May and September.Calculate the net surface radiation flux Rn from 2003 to 2018.During the growth period,the net radiation flux in Heilongjiang Province was the largest in November;5)Calculate the daily evapotranspiration from 2003 to 2018.Water bodies and wetland evapotranspiration are relatively large,and the daily evapotranspiration during May to September period from 2003 to 2018 ranges from 10 to 18 mm.Urban evapotranspiration is relatively small,and the range of daily evapotranspiration during May to September period from 2003 to 2018 is 0-8 mm.The evapotranspiration of the farmland area varies with time interval.Explore the impact of rice field expansion on daily and monthly evapotranspiration.From the end of April to the beginning of September,the daily evapotranspiration in the expansion area of rice fields increased,and the daily evapotranspiration in the expansion area of rice fields decreased from mid-September to the beginning of November.The cumulative evapotranspiration from May to October increased by 1570587.1 mm.