分层神经网络（LNN）可以用来分析地下水中的氡浓度,而且试图给出氡浓度和环境参数之间的函数关系。由于环境（例如：降雨量）对地下水中氡浓度的影响可能是非线性的,与目前时间脉冲响应线性计算方法相比，该方法能够较准确的估计出环境参数造成的氡浓度的变化。从厦门市东孚水氡观测井获取的数据分析表明，该方法能够较准确的分析出由地震导致的氡浓度变化,同时能够排除环境因素引起的氡浓度变化。

LNN had been used to employ the current radon concentration in groundwater, and attempt to analysis equation function of radon concentration with environmental parameters. The environment (For example: rainfall) on the radon concentration in groundwater may be nonlinear, which can estimate accurately the radon concentration by environmental parameters change compared with the Linear Computational Technique (CLT). Radon observation wells of Xiamen Dong Fu show that LNN can accurate to find out the change of radon concentration by the earthquake from environmental factors (For example: rainfall) , In addition, LNN can tell the change of radon concentration by the environmental factors from other factors.