Wetlands play a significant role in flood mitigation. Remote sensing technologies as an efficient and accurate approach have been widely applied to delineate wetlands. Supervised classification is conventionally applied for remote sensing technologies to improve the wetland delineation accuracy. However, performing supervised classification requires preparing the training data, which is also considered time-consuming and prone to human mistakes. This paper presents a deterministic topographic wetland index to delineate wetland inundation areas without performing supervised classification. The classic methods such as Normalized Difference Vegetation Index, Normalized Difference Water Index, and Topographic Wetness Index were chosen to compare with the proposed deterministic topographic method on wetland delineation accuracy. The ground truth sample points validated by Google satellite imageries from four different years were used for the assessment of the delineation overall accuracy. The results show that the proposed deterministic topographic wetland index has the highest overall accuracy (98.90%) and Kappa coefficient (0.641) among the selected approaches in this study. The findings of this paper will provide an alternative approach for delineating wetlands rapidly by using solely the LiDAR-derived Digital Elevation Model.

Flooding is one of the most frequent natural hazards and causes more economic loss than all the other natural hazards. Fast and accurate flood prediction has significance in preserving lives, minimizing economic damage, and reducing public health risks. However, current methods cannot achieve speed and accuracy simultaneously. Numerical methods can provide high-fidelity results, but they are time-consuming, particularly when pursuing high accuracy. Conversely, neural networks can provide results in a matter of seconds, but they have shown low accuracy in flood map generation by all existing methods. This work combines the strengths of numerical methods and neural networks and builds a framework that can quickly and accurately model the high-fidelity flood inundation map with detailed water depth information. In this paper, we employ the U-Net and generative adversarial network (GAN) models to recover the lost physics and information from ultra-fast, low-resolution numerical simulations, ultimately presenting high-resolution, high-fidelity flood maps as the end results. In this study, both the U-Net and GAN models have proven their ability to reduce the computation time for generating high-fidelity results, reducing it from 7–8 h down to 1 min. Furthermore, the accuracy of both models is notably high.