A Comprehensive Evaluation of the National Water Model (NWM) – Height Above Nearest Drainage (HAND) Flood Mapping Methodology

Abstract

Abstract. Flood maps are needed for emergency response, research, and planning. The Height Above Nearest Drainage (HAND) technique is a low-complexity, terrain-based approach for inundation mapping from elevation, discharge-height relationship, and streamflow data. The recent operational capacities of the NOAA National Water Model (NWM) and pre-processed HAND products from the University of Texas offer an operational framework for real-time and forecast flood mapping across the United States. This paper offers the first detailed evaluation of the coupled NWM-HAND approach by comparing NWM-HAND simulations to a database of remotely sensed flood maps. These comparisons are made both quantitatively and through a detailed evaluation of selected results. The results show that NWM-HAND performs well in the majority of cases with 70 % of the calculated flood maps yielding an agreement of 80 % or better with the remote sensing observations. Persistent inaccuracies occur in lower-order reaches, regions of low relief, and when the NWM misses substantially either in magnitude or timing.

Reply on RC1

“…That is, the model is more likely to generate false alarms than it is to miss areas of true inundation in the benchmark dataset. These performance metrics are comparable with those obtained in previous studies using models with more explicit computational complexity (Sampson et al, 2015;Sosa et al, 2020;Wing et al, 2017) or with similar sub-grid topographic representations (Johnson et al, 2019;Yamazaki et al, 2011).…”

“…After further extreme flooding in 2015-2016, the National Flood Resilience Review called for "a more integrated flood risk modelling approach to allow simulations to be run which link meteorology, hydrology and flooding across England" (Cabinet Office, 2016, p.26). To address these needs, simplified approaches have been applied to reduce computation time further, although their accuracy is yet to be evaluated fully (Afshari et al, 2018;Hall et al, 2003;Johnson et al, 2019;Nobre et al, 2016).…”

Assessment of Relative Accuracy of TxRR and NWM in Simulating Streamflow for a Cluster of Watersheds along the south Texas Coast

“…Earlier studies assessing NWM analysis also reported similar negative biases. Johnson et al, (2019), for example, found that the inundation maps generated by combining NWM 1.2 reanalysis with the HAND method (Nobre et al, 2016) underpresernted the extent of inundation, and cited the underprediction of streamflow by NWM as one of the primary causes. It should be noted, however, that this underprediction can be at least partially explained by the negative biasse in NLDAS-2 precipitation data during these events -there is a negative bias in the NLDAS precipitation data evident in major storm events that is indicative of data quality issues.…”

“…As NWM is a relatively new model, to date only limited works have touched on its performance, and few of these focused on the skills of NWM vis-à-vis simpler, conceptual models (Nobre et al, 2016, Hansen et al, 2019Johnson et al, 2019;Viterbo et al, 2020a;2020b). Loosely related, there is a body of literature on relative performance of physicallybased, distributed and lumped, conceptual hydrologic models.…”