Forecasting Global Seaweed Production Trends Using Autoregressive Time Series Models: A Machine Learning Approach

As the global population approaches 10 billion, the demand for sustainable biotic resources has repositioned seaweed aquaculture from a localized maritime practice to a critical pillar of the global blue economy. The surge is evident from a sixfold rise in global seaweed biomass production from 1992 to 2024, with a staggering 97% of this biomass originating from cultivation rather than wild capture. However, the inherent stochasticity of marine environments and the non-linear nature of biological growth pose significant challenges to traditional econometric forecasting. The current study presents an evaluation of the predictive efficacy of autoregressive (AR), moving average (MA), autoregressive moving average (ARMA), Prophet, and automatic autoregressive integrated moving average (Auto ARIMA) compared with non-linear architectures, including non-linear autoregressive (NAR) networks trained using Levenberg–Marquardt (TRAINLM), Bayesian regularization (TRAINBR), and scaled conjugate gradient (TRAINSCG). The results demonstrated that the TRAINSCG algorithm consistently outperforms traditional models, effectively capturing the complex, non-linear trajectories that linear approximations fail to resolve. Validation metrics confirm a superior model fit, with a mean absolute percentage error (MAPE) of 3.84% and an R2 of 0.9455, allowing for a robust projection of global production reaching approximately 55 million tonnes by 2030 with an estimated value of US$22.42 billion. This trajectory underscores a booming market for red and brown macroalgae, driven by industrial demand for hydrocolloids. Systematic data acquisition with integrated climate modeling studies may provide the empirical foundation necessary to accurately assess investment risk and attract better research fund allocation. AI-driven insights may empower farmers, entrepreneurs, and policy makers to make calculated decisions that strategically advance the global seaweed bioeconomy.