基于山东省 29 个气象站 1951—2017 年的实测气象数据，主要采用降水量距平百分率作为气象干旱等级评 价指标，通过云模型分析山东省干旱时空分布的均匀性和稳定性特征，并采用 ArcGIS 的反距离权重法分析不同 时间尺度下的干旱频次空间分布特征。结果表明：山东省降水量距平百分率呈逐年波动式下降趋势，但下降趋势 不明显，干旱发生具有明显的季节性，呈现出季节性连旱现象；从时间尺度来看，年尺度上干旱分布的模糊性和随 机性较小，干旱的确定性更容易度量，干旱分布不均匀性的稳定程度较低；对季节间云数字特征的对比显示，多年 来冬、秋季干旱分布的模糊性和随机性较大，冬季干旱分布的确定性最难以度量；在稳定性方面秋季最稳定，春季 次之，夏季最不稳定；在空间尺度上，鲁西北地区发生干旱、严重干旱和特大干旱的频次最高；发生轻度干旱的频 次最高区位于胶东半岛，鲁南地区次之；发生中度干旱的频次最高区位于鲁中地区，鲁南地区次之；由超熵值大小 判定干旱分布不均匀性的稳定程度，表现为鲁西北地区最稳定，鲁中地区次之，胶东半岛最不稳定；从干旱分布的 均匀性分析得出鲁西南地区干旱分布最不均匀，鲁西北地区干旱分布最均衡；山东省干旱频次在空间上的分布表 现为春旱和冬旱呈从西北向东南递减的趋势，秋旱比较严重。云模型针对半湿润地区的干旱分布具有较好的应 用前景。

With?the?acceleration?of?global?warming?and?frequent?occurrence?of?agricultural?natural?disasters,?drought has?become?an?important?restrictive?factor?for?agricultural?production?in?Shandong?Province?and?has?caused?huge losses?to?agriculture.?As?an?important?region?of?the?national?ecological?protection?and?high-quality?development strategy?in?the?Yellow?River?basin,?the?high-quality?development?of?ecological?agriculture?is?particularly?important in ?Shandong ?Province, ?and ?drought ?is ?one ?of ?the ?limiting ?factors ?affecting ?the ?high-quality ?development ?of ecological?agriculture.?So?the?study?of?drought?is?becoming?more?and?more?important.?In?recent?years,?there?have been?more?and?more?research?on?extreme?drought?monitoring,?risk?assessment?models?and?systems,?and?a?variety?of meteorological?hydrological?drought?index?models?have?been?widely?used.?The?precipitation?anomaly?percentage model ?is ?superior ?to ?the ?other ?drought ?index ?models ?because ?it ?can ?directly ?reflect ?the ?drought ?caused ?by precipitation.?Although?some?research?has?been?conducted?in?recent?years?on?temporal-spatial?drought?distribution, there?are?few?researches?on?the?fuzziness,?randomness?and?unevenness?of?drought?distribution. ??????Based?on?the?measured?data?of?meteorological?data?from?1951?to?2017?in?29?meteorological?stations?in?Shandong Province,?the?precipitation?anomaly?percentage?is?used?as?the?evaluation?index?of?meteorological?drought?grade,?the cloud ?model ?is ?used ?to ?study ?the ?uniformity ?and ?stability ?characteristics ?of ?the ?drought ?spatial ?and ?temporal distribution ?in ?Shandong ?Province ?and ?the ?inverse ?distance ?weighting ?method ?is ?used ?to ?analyze ?the ?spatial distribution?characteristics?of?drought?frequency?at?different?time?scales. ??????Results?showed?that:?the?precipitation?anomaly?percentage?in?Shandong?Province?shows?a?downward?trend?of fluctuation ?year ?by ?year, ?but ?the ?decreasing ?trend ?was ?not ?obvious; ?the ?occurrence ?of ?drought ?has ?obvious seasonality, ?showing ?the ?phenomenon ?of ?seasonal ?continuous ?drought. ?On ?the ?time ?scale, ?the ?ambiguity ?and randomness?of?the?drought?distribution?on?the?annual?scale?are?smaller?than?the?seasonal?scale,?the?determinism?of the?drought?is?easier?to?measure?than?the?seasonal?scale,?and?the?stability?of?the?unevenness?of?distribution?is?low;?the comparison?of?the?numerical?characteristics?of?the?inter-seasonal?cloud?shows?that?the?distribution?of?winter?and autumn ?droughts ?has ?been ?relatively ?vague ?and ?random ?over ?the ?years, ?and ?the ?certainty ?of ?winter ?drought distribution?is?the?most?difficult?to?measure.?In?terms?of?stability,?autumn?is?the?most?stable,?followed?by?spring,?and the?most?unstable?is?summer. ??????On ?the ?spatial ?scale, ?the ?frequency ?of ?drought, ?severe ?drought ?and ?extreme ?drought ?in ?the ?northwestern Shandong?area?is?the?highest;?the?frequency?of?mild?drought?in?the?Jiaodong?Peninsula?area?is?the?highest,?followed by?the?southern?Shandong?area;?the?frequency?of?moderate?drought?is?the?highest?in?middle?Shandong?Province, followed?by?south?Shandong?Province.?The?stability?of?drought?distribution?was?judged?by?the?value?of?super- entropy,?which?showed?that?the?northwest?of?Shandong?was?the?most?stable,?the?middle?of?Shandong?was?the?second, and?the?Jiaodong?Peninsula?was?the?most?unstable.?Based?on?the?analysis?of?the?uniformity?of?drought?distribution,?it is ?concluded ?that ?the ?drought ?distribution ?in ?southwest ?Shandong ?is ?the ?most ?uneven, ?and ?that ?the ?drought distribution?in?northwest?Shandong?is?the?most?balanced.?The?spatial?distribution?of?drought?frequency?in?Shandong Province?showed?that?spring?drought?and?winter?drought?decreased?from?northwest?to?southeast,?and?autumn?drought was?more?serious.The?cloud?model?has?a?good?application?prospect?for?drought?distribution?in?subhumid?area.