The complexity of the atmosphere renders the modelling of the atmospheric delay in multi temporal InSAR difficult. This limits the potential of achieving millimetre accuracy of InSAR-derived deformation measurements. In this paper we review advances in tropospheric delay modelling in InSAR, tropospheric correction methods and integration of the correction methods within existing multi temporal algorithms. Furthermore, we investigate ingestion of the correction techniques by different InSAR applications, accuracy performance metrics and uncertainties of InSAR derived measurements attributed to tropospheric delay. Spatiotemporal modelling of atmospheric delay has evolved and can now be regarded as a spatially correlated turbulent delay with varying degree of anisotropy random in time and topographically correlated seasonal stratified delay. Tropospheric corrections methods performance is restricted to a case by case basis and ingestion of these methods by different applications remains limited due to lack of their integration into existing algorithms. Accuracy and uncertainty assessments remain challenging with most studies adopting simple statistical metrics. While advances have been made in tropospheric modelling, challenges remain for the calibration of atmospheric delay due to lack of data or limited resolution and fusion of multiple techniques for optimal performance.

大气的复杂性使得在多时间InSAR中对大气延迟的建模变得困难。这限制了获得InSAR衍生的变形测量​​结果的毫米精度的可能性。在本文中，我们回顾了InSAR中对流层延迟建模，对流层校正方法以及将校正方法集成到现有的多时间算法中的进展。此外，我们研究了不同InSAR应用对校正技术的摄取，准确性性能指标以及对流层延迟引起的InSAR衍生测量的不确定性。大气延迟的时空模型已经发展，现在可以看作是空间相关的湍流延迟，其各向异性程度随时间变化是随机的，并且与地形相关的季节性分层延迟也是如此。对流层校正方法的性能受限于具体情况，并且由于缺乏将它们集成到现有算法中的能力，因此不同应用对这些方法的摄取仍然受到限制。在大多数采用简单统计指标的研究中，准确性和不确定性评估仍然具有挑战性。尽管在对流层建模方面取得了进展，但由于缺乏数据或分辨率有限以及融合多种技术以获得最佳性能，大气延迟校准仍面临挑战。