On the verification of ensemble precipitation forecasts over the Godavari River basin

Manikanta, Velpuri; Teja, K. Nikhil; Das, Jew; Umamahesh, N. V.

doi:10.1016/j.jhydrol.2022.128794

Cited by 9 publications

(3 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Data-driven statistical methods used for flood prediction include autoregressive integrated moving average (ARIMA) [28], multiple linear regression (MLR) [29] Flood frequency analysis (FFA) [30] and regional flood frequency analysis (RFFA) [31] have been proposed. Later, empirical orthogonal function (EOF) [32], Bayesian forecasting models (BFM) [33], quantile regression techniques (QRT) [34], and climatology average method (CLIM) [35] have also been focused for flood prediction. Advanced flood forecasting systems for long-term and short-term prediction of floods are also significant for the generation of early flood warnings.…”

Section: Related Workmentioning

confidence: 99%

FFM: Flood Forecasting Model Using Federated Learning

et al. 2023

View full text Add to dashboard Cite

Floods are one of the most common natural disasters that occur frequently causing massive damage to property, agriculture, economy and life. Flood prediction offers a huge challenge for researchers struggling to predict floods since long time. In this article, flood forecasting model using federated learning technique has been proposed. Federated Learning is the most advanced technique of machine learning (ML) that guarantees data privacy, ensures data availability, promises data security, and handles network latency trials inherent in prediction of floods by prohibiting data to be transferred over the network for model training. Federated Learning technique urges for onsite training of local data models, and focuses on transmission of these local models on the network instead of sending huge data set towards central server for local model aggregation and training of global data model at the central server. In this article, the proposed model integrates locally trained models of eighteen clients, investigates at which station flooding is about to happen and generates flood alert towards a specific client with five days lead time. A local feed forward neural network (FFNN) model is trained at the client station where the flood has been expected. Flood forecasting module of local FFNN model predicts the expected water level by taking multiple regional parameters as input. The dataset of five different rivers and barrages has been collected from 2015 to 2021 considering four aspects including snow melting, rainfall-runoff, flow routing and hydrodynamics. The proposed flood forecasting model has successfully predicted previous floods happened in the selected zone during 2010 to 2015 with 84 % accuracy. INDEX TERMS hydraulic, meteorological, flood forecasting system, federated learning, feed-forward neural network.FIGURE 1. Natural hazards percentage as per occurrence in Pakistan during 2021 [6]

show abstract

Section: Related Workmentioning

confidence: 99%

FFM: Flood Forecasting Model Using Federated Learning

et al. 2023

View full text Add to dashboard Cite

show abstract

“…A New Feature of X-SLIP Platform for Real-Time Prediction through Rainfall Forecasts X-SLIP platform could be implemented in any LEWS if stability of the topsoil were analysed in real time; this is possible by integrating rain forecast in SLIP runs so that any critical situation can be predicted well in advance. In Europe, weather forecasts are based on the ECMWF (European Centre for Medium-Range Weather Forecasts) model, considered as one of the most reliable [66]. Some countries, among which Italy, adopt COSMO (COnsortium for Small-scale MOdeling); based on ECMWF model, COSMO integrates details with higher resolution and allows us to predict localised weather events [67].…”

Section: 2mentioning

confidence: 99%

Incorporating Rainfall Forecast Data in X-SLIP Platform to Predict the Triggering of Rainfall-Induced Shallow Landslides in Real Time

Gatto

2023

Geosciences

View full text Add to dashboard Cite

Extreme and prolonged rainfall resulting from global warming determines a growing need for reliable Landslide Early Warning Systems (LEWS) to manage the risk of rainfall-induced shallow landslides (also called soil slips). Regional LEWS are typically based on data-driven methods because of their greater computational effectiveness, which is greater than the ones of physically based models (PBMs); however, the latter reproduces the physical mechanism of the modelled phenomena, and their modelling is more accurate. The purpose of this research is to investigate the prediction quality of the simplified PBM SLIP (implemented in the X-SLIP platform) when applied on a regional scale by analysing the stability of rain forecasts. X-SLIP was updated to handle the GRIB files (format for weather forecast). Four real-time predictions were simulated on some towns of the Emilia Apennines (northern Italy) involved in widespread soil slips on 5 April 2013; specifically, maps of factors of safety related to this event were derived assuming that X-SLIP had run 72 h, 48 h, 24 h and 12 h in advance. The results indicated that the predictions with forecasts (depending on the forecast quality) are as accurate as the ones derived with rainfall recordings only (benchmark). Moreover, the proposed method provides a reduced number of false alarms when no landslide was reported to occur in the whole area. X-SLIP with rain forecasts can, therefore, represent an important tool to predict the occurrence of future soil slips at a regional scale.

show abstract

“…These systems differ in terms of the number of ensemble members and the approach to capturing the various sources of uncertainties. Despite providing useful forecast guidance, many operational centers and studies report an issue with under-dispersiveness of the ensemble (Buizza et al, 2005;Raftery et al, 2005;Hohenegger et al, 2008;Gebhardt et al, 2011;El-Ouartassy et al, 2022;Lakatos et al, 2023;Manikanta et al, 2023), where the spread of the ensemble members is too small to fully capture the forecast uncertainty.…”

Section: Introductionmentioning

confidence: 99%

Adaptive selection of members for convective-permitting regional ensemble prediction over the western Maritime Continent

Sharma,

Lee,

Porson

et al. 2023

Front. Environ. Sci.

View full text Add to dashboard Cite

A common issue faced by the downscaled regional ensemble prediction systems is the under-dispersiveness of the ensemble forecasts, often attributed to the lack of spread under the initial conditions from the global ensemble. In this study, a novel method that adopts an adaptive approach to selecting global ensemble members for regional downscaling has been developed. Instead of using a fixed set of pre-selected global ensemble members, the adaptive selection performs a sampling algorithm and selects the global ensemble members, which maximizes a fractions skill score (FSS)-based displacement between ensemble members. The method is applied to a convective-permitting ensemble prediction system over the western Maritime Continent, referred to as SINGV-EPS. SINGV-EPS has a grid spacing of 4.5 km and is a 12-member ensemble that is driven by the European Centre for Medium-Range Weather Forecasts (ECMWF) 51-member global ensemble. Month-long trials were conducted in June 2020 to assess the impact of adaptive selection on the ensemble forecast spread and rainfall verification scores. In both fixed pre-selection and adaptive selection experiments, SINGV-EPS was still under-dispersive. However, adaptive selection improved the ensemble spread and reduced the root-mean-square error (RMSE) of the ensemble mean in wind, temperature, and precipitation fields. Further verification of the rainfall forecasts showed that there was a reduction in the Brier score and a higher hit rate in the relative operating characteristic (ROC) curve for all rainfall thresholds when adaptive selection was applied. Additionally, the ensemble mean forecasts from adaptive selection experiments are more accurate beyond 24 h, with a higher FSS for all rainfall thresholds and neighborhood lengths. These results suggest that the adaptive selection is superior to the fixed pre-selection of global ensemble members for downscaled regional ensemble prediction.

show abstract

On the verification of ensemble precipitation forecasts over the Godavari River basin

Cited by 9 publications

References 45 publications

FFM: Flood Forecasting Model Using Federated Learning

FFM: Flood Forecasting Model Using Federated Learning

Incorporating Rainfall Forecast Data in X-SLIP Platform to Predict the Triggering of Rainfall-Induced Shallow Landslides in Real Time

Adaptive selection of members for convective-permitting regional ensemble prediction over the western Maritime Continent

Contact Info

Product

Resources

About