PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
March 6, 2026HydrologyOpen Access

Short-Term Streamflow Forecasting for River Management, Using ARIMA Models and Recurrent Neural Networks

View Full Paper
Ask AI
Bookmark
Share

Authors

NSNicolai SîrbuARAndrei-Mihai Rugină

Discussion

Loading...

Member takes

Overview

Controlled comparison assesses water level forecasting accuracy in river management, suggesting model choice impacts prediction.

Key Points

  • The aim is to compare the forecasting capabilities of ARIMA/SARIMA models with LSTM networks for short-term river water-level predictions.
  • Conducted a controlled comparison between ARIMA/SARIMA and LSTM models.
  • Utilized synthetic daily hydrographs for normal, drought, and flood conditions.
  • Applied a rolling-origin forecasting strategy for bias reduction.
  • Evaluated forecast skill using standard error metrics and Global Forecast Skill Index.
  • Median performance between SARIMA and LSTM models was comparable across various conditions.
  • No statistically significant differences were detected based on nonparametric tests.
  • Differences observed under flood conditions require cautious interpretation due to sample limitations.

Cite This Study

Sîrbu et al. (2026) studied this question.

synapsesocial.com/papers/69aa70b8531e4c4a9ff5ab59https://doi.org/10.3390/hydrology13030082
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Streamflow Forecasting: A Comparative Analysis of ARIMAX, Rolling Forecasting LSTM Neural Network and Physical-Based Models in an Italian Natural Catchment2024
  2. 2Forecasting Water Demand With the Long Short-Term Memory Deep Learning Mode2024 · 4 citations
  3. 3Long Short-Term Memory Networks for Real-time Flood Forecast Correction: A Case Study for an Underperforming Hydrologic Model2024 · 2 citations
  4. 4Streamflow Forecasting: A Comparative Analysis of ARIMAX, Rolling Forecasting LSTM Neural Network and Physically Based Models in a Pristine Catchment2025 · 6 citations
  5. 5Comparative insights into short-term streamflow forecasting in navigating urban flood risks2026