Based on the 20%,40%,60%and 80%quantiles of the daily cumulative energy of downward short-wave radiation,the four-component radiation observation data observed by the Haibei Animal Husbandry Experi-mental Station of China Meteorological Administration from September 2014 to December 2020 were divided in-to five groups.The influence of difference in downward shortwave radiation on the radiation balance of meadow grassland has been analyzed,which provided a basis for clarifying the changes and causes of radiation balance under different sunlight conditions.The results show that the downward shortwave radiation energy in the whole year is 8192.9 MJ·m-2·a-1,which can be theoretically received under the condition of abundant sunlight.Under normal sunlight,it is only 80%of that in the condition of abundant sunlight,while this proportion is only 40%under the condition of less sunlight.The downward shortwave radiation received by the surface under different sunlight conditions showes a very significant logarithmic increase trend in the same month,and the difference be-tween the groups exceeds the difference between different season.There are also large differences between the groups of upward shortwave radiation and atmospheric longwave radiation.However,the upward shortwave radi-ation increases with the increase of solar radiation,due to the change of air humidity and cloudiness,while the atmospheric longwave radiation decreases with the increase of solar radiation.The variation of surface longwave radiation under different sunlight conditions is small,and does not exceed 3%.Although the difference in surface longwave radiation is not significant between the three sunlight conditions,due to the small amount of down-ward shortwave radiation received under the condition of less sunlight,the surface longwave radiation is 3.4 times more than the downward shortwave radiation,while it is only 1.4 and 1.7 times under the other two light-ing conditions.As the increase of sunlight,the net shortwave radiation received by the surface increases,and the net longwave radiation also increases.Therefore,the energy absorbed by the radiation under different sunlight conditions accounts for 36%to 39%of the downward shortwave radiation energy received in each case.In addi-tion,the difference of sunlight conditions also affect the linear regression model between downward shortwave radiation and net total radiation.The coefficient of determination and slope of the relationship model are the smallest under the condition of less sunlight.The coefficient of determination of the relationship model are above 0.85 for both normal and abundant sunlight.However,the slope is 1 when the sunlight is normal,while the slope is 0.87 when the sunlight is abundant.

利用1999—2018年逐月ERA-interim再分析数据,选取青海省东部农业区的15个气象站点平均气温和土壤温度,采用误差和相关性分析,对比两种资料时间和站点空间变化规律,并对ERA-interim再分析资料的准确性进行验证评估.结果表明:ERA-interim再分析数据平均气温、5和15 cm土壤温度均低于观测值,分别偏低3.5、4.3和4.5℃;观测值与订正前后的再分析资料月际和季节变化基本一致,空间变化均表现出从东南部向西北部减小的趋势,而且最大值和最小值出现的站点完全一致.近20年来,观测值与再分析资料平均气温均呈现上升趋势,5和15 cm土壤温度观测值表现为升高趋势,而再分析数据呈减小趋势.与订正前相比,订正后的再分析资料与观测值的平均偏差和均方根误差明显减小,同时平均偏差和均方根误差年际变化均呈先增大后减小的趋势.年内平均偏差和均方根误差均呈现"M"型,最大值均出现在4月,最小值出现在12或1月.东部农业区地形复杂、海拔高度差异是再分析资料比观测值偏低的主要原因,利用回归方程实现平均气温和土壤温度再分析数据的订正,有效降低了ERA-interim再分析数据的偏差,提高再分析数据评估的准确性,提升再分析资料对青海省东部农业区农时农事服务及评估.

干旱是影响祁连山地区畜牧业发展最主要的自然灾害之一.本文利用1961～2017年月降水观测资料,以及《北方牧区草原干旱等级指标》中降水距平百分率指标的划分依据,分析了祁连山地区降水的变化和干旱发生的时空特征,结果表明:(1) 1961 ～ 2017年降水量呈震荡递增趋势,其增长速率为13.6mm/10a;(2)降水量的季节变化特征明显,各季节降水量也表现出逐年递增的趋势,夏季降水量的年际增长率最大,春季、秋季年降水量的年际增长率较一致,冬季降水量的年际增长率最小;(3)祁连山地区全年当中特旱的发生概率为0.6％;重旱发生概率为4.7％;中旱发生概率为27.5％;轻旱发生概率为44.4％;无旱发生概率为22.8％.

利用1961～2010年西宁市三县灾情普查资料和降水资料,通过数理统计法分析了西宁市基于降水距平百分率的气象干旱指标对农业气象干旱的适用性,同时采用相关性分析及逐步回归等方法,用近年来自动站土壤墒情资料,建立了简单易行的农业气象干旱预测预警模型.研究结果表明,西宁市生长季各月降水主要为正常年型,其次为轻旱和稍丰年型;季节正常为主要降水年型,其次为轻旱、稍丰,其余降水年型出现频率小.干旱发生频率从大到小均为春、秋和夏;年降水量基本不出现重旱或超丰以上的降水年型,正常年型占58％～66％,气象干旱发生频率为14％ ～ 22％.轻旱出现频率为15％左右,而中旱和重旱出现频率不足5％.用降水、气温与土壤水分间的关系建立墒情模型,参数获取简单,实用性强,开展农业气象干旱发生、发展趋势预测,对农业生产管理具有一定的指导作用.