Merritts D, Hesterberg T
Science. 1994 Aug 19;265(5175):1081-4. doi: 10.1126/science.265.5175.1081.
Stream networks are sensitive to low rates of surface uplift and can be used to decipher the history of large earthquakes even where faults do not rupture the surface, as in intraplate seismic zones. Statistical analysis of alluvial network data from topographic maps in the New Madrid seismic zone, in the central United States, shows that stream-segment gradients deviate the most from an estimated natural stream profile where surface uplift is greatest. Evidence of cumulative deformation distilled from stream network patterns represents at least several meters of differential surface uplift during Holocene time, which suggests that more than one cycle of surface deformation occurred.
水系网络对地表缓慢隆升很敏感，即使在板块内部地震带等断层未切穿地表的地方，也可用于解读大地震的历史。对美国中部新马德里地震带地形图上的冲积网络数据进行统计分析表明，在地表隆升最大的地方，河段坡度与估计的自然河流剖面偏差最大。从水系网络模式中提取的累积变形证据表明，全新世期间至少有几米的地表差异隆升，这表明发生了不止一个地表变形周期。
