Land subsidence in response to declining water table at plains under sparse data is investigated using ALPRIFT, introduced recently by the authors at the stage of its proof-of-concept. ALPRIFT is a framework, which pools consensually together seven general-purpose data layers with a scoring system of prescribed rates accounting for local variations and prescribed weights accounting for their relative importance. It is a subsidence vulnerability indexing (SVI) approach, which estimates relative values and is subject to inherent subjectivities. The paper treats the transformation of SVI into a risk indexing (RI) capability through a scheme, in which ALPRIFT breaks down into ALRIF, characterising passive local effects and into water-driven PT, characterising active system-wide effects. The addition of passive and active processes renders total vulnerability but their products render a measure of risk index. A modelling strategy is formulated for SVI and RI at two levels to treat inherent subjectivities and involves data fusion by using catastrophe theory. The strategy is applied to an aquifer subject to decline in water table at the coast of Lake Urmia, with sparse data. The results provide evidence for the proof-of-concept on SVI and RI using ROC/AUC performance metrics.

作者使用最近在概念证明阶段引入的ALPRIFT来研究稀疏数据下平原上地下水位下降引起的地面沉降。ALPRIFT是一个框架，该框架将七个通用数据层汇总在一起，并具有计分汇率的计分系统，计分汇率时考虑了本地差异，而计价权重时考虑了它们的相对重要性。这是一种沉降脆弱性索引（SVI）方法，它估计相对值并受固有主观性的影响。本文通过一种方案将SVI转换为风险索引（RI）能力，其中ALPRIFT分解为ALRIF（表征被动局部效应）和水驱动PT（表征主动系统范围效应）。被动和主动流程的添加会带来总体漏洞，但其产品会带来一定程度的风险指数。针对SVI和RI在两个级别上制定了建模策略，以处理固有的主观性，并通过使用突变理论涉及数据融合。该策略适用于受稀疏数据影响的乌尔米亚湖沿岸地下水位下降的含水层。结果为使用ROC / AUC性能指标的SVI和RI概念验证提供了证据。