Browsing by Author "Makumbura, R.K"

Now showing 1 - 2 of 2

Open Access
QPred: A Lightweight Deep Learning-Based Web Pipeline for Accessible and Scalable Streamflow Forecasting
(Tech Science Press, 2026) Makumbura, R.K; Wijesundara, H; Sajindra, H; Rathnayake, U; Kumar, V; Duraibabu, D; Sen, S
Accurate streamflow prediction is essential for flood warning, reservoir operation, irrigation scheduling, hydropower planning, and sustainable water management, yet remains challenging due to the complexity of hydrological processes. Although data-driven models often outperform conventional physics-based hydrological modelling approaches, their real-world deployment is limited by cost, infrastructure demands, and the interdisciplinary expertise required. To bridge this gap, this study developed QPred, a regional, lightweight, cost-effective, web-delivered application for daily streamflow forecasting. The study executed an end-to-end workflow, from field data acquisition to accessible web-based deployment for on-demand forecasting. High-resolution rainfall data were recorded with tipping-bucket gauges and loggers, while river water depth in the Aglar and Paligaad watersheds was converted to discharge using site-specific rating curves, resulting in a daily dataset of precipitation, river water level and discharge. Four DL architectures were trained, including vanilla Long Short-Term Memory (LSTM), stacked LSTM, bidirectional LSTM, and Gated Recurrent Unit (GRU), and evaluated using Nash-Sutcliffe Efficiency (NSE), Coefficient of Determination (R2), Root-Mean-Square-Error-Standard-Deviation Ratio (RSR), and Percentage Bias (PBIAS) metrics. Performance was watershed-specific, as the vanilla LSTM demonstrated the best generalisation for the Aglar watershed (R2 = 0.88, NSE = 0.82, RMSE = 0.12 during validation), while the GRU achieved the highest validation accuracy in Paligaad (R2 = 0.88, NSE = 0.88, RMSE = 0.49). All models achieved satisfactory to excellent performance during calibration (R2 > 0.91, NSE > 0.91 for both watersheds), demonstrating strong capability to capture streamflow dynamics. The highest performing models were selected and embedded into the QPred application. QPred was developed as a lightweight web pipeline, utilising Google Colab as the primary execution environment, Flask as the backend inference framework, Google Drive for artefact storage, and Ngrok for secure HTTPS tunnelling. A user-friendly front end utilises range sliders (bounded by observed minima and maxima) to gather inputs and provides discharge data along with metadata, thereby enhancing transparency. This work demonstrates that accurate, context-aware deep learning models can be delivered through low-cost, web-based platforms, providing a reproducible and scalable pipeline for hydrological applications in other watersheds and for practitioners. Copyright
Embargo
Variation of Leaf Area Index (LAI) under Changing Climate: Kadolkele Mangrove Forest, Sri Lanka
(Hindawi, 2022-10-10) Makumbura, R.K; Rathnayake, U
Mangroves are an essential plant community in coastal ecosystems. While the importance of mangrove ecosystems is well acknowledged, climate change is expected to have a considerable negative impact on them, especially in terms of temperature, precipitation, sea level rise (SLR), ocean currents, and increasing storminess. Sri Lanka ranks near the bottom of the list of countries researching this problem, even though the scientific community's interest in examining the variation in mangrove health in response to climate change has gained significant attention. Consequently, this study illustrates how the leaf area index, a measure of mangrove health, fluctuates in response to varying precipitation, particularly during droughts in Sri Lanka's Kadolkele mangrove forest. The measurements of the normalized difference vegetation index (NDVI) were used to produce the leaf area index (LAI), which was then combined with the standard precipitation index (SPI) to estimate the health of the mangroves. The climate scenario, RCP8.5, was used to forecast future SPI (2021–2100), and LAI was modeled under the observed (1991–2019) and expected (2021–2100) drought events. The study reveals that the forecasted drought intensities modeled using the RCP8.5 scenario have no significant variations on LAI, even though some severe and extreme drought conditions exist. Nevertheless, the health of the mangrove ecosystem is predicted to deteriorate under drought conditions and rebound when drought intensity decreases. The extreme drought state (-2.05) was identified in 2064; therefore, LAI has showcased its lowest (0.04). LAI and SPI are projected to gradually increase from 2064 to 2100, while high fluctuations are observed from 2021 to 2064. Limited availability of LAI values with required details (measured date, time, and sample locations) and cloud-free Landsat images have affected the study results. This research presents a comprehensive understanding of Kadolkele mangrove forest under future droughts; thus, alarming relevant authorities to develop management plans to safeguard these critical ecosystems.