Hydrological drought analysis in response to climate change based on a novel hybrid machine learning algorithm

Abstract

For the first time, a combination of metaheuristic algorithms and machine learning is used for hydrological drought analysis under climate change conditions and applications. The new framework is used by a novel hybrid machine learning model named the least-squares support vector machine-African vulture optimization algorithm (LSSVM-AVOA). The performance of the proposed hybrid algorithm was compared with three algorithms, including artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS), and least squares support vector machines (LSSVM). Using the technique for order of preference by similarity to the ideal solution (TOPSIS) method showed that LSSVM-AVOA with a score of 0.98 could be employed to improve the LSSVM modeling results. Three global climate models (GCM), including ACCESS-ESM1-5 (A1), CanESM5 (C5), and MRI-ESM2-0 (M2) during the historical period 1985–2014 and four scenarios, including SSP126, SSP245, SSP245-cov-fossil (SCF), SSP585 in future from 2021 to 2050 was considered for predicting discharge at Karun basin (Sepid Dasht Sezar, Bakhtiari Tang Pang, Sezar Tang Pang, Taleh Zang and Gotvand stations). Using new hybrid algorithm, the prediction results showed that discharge in most scenarios and stations decreased between − 0.81% and − 7.83% (except Sezar Tang Pang and Gotvand station). The standardized runoff index (SRI) results of hydrological drought analysis showed by the SSP585 scenario it seems that for the next first 10 years, a mild drought period can occur in this basin. Also, in the future period for SSP126 scenario by the first five years in the next 30 years, none drought is predicted.

Relative CO2Column Height for CO2Geological Storage: A Non-Negligible Contribution from Reservoir Rock Characteristics

“…Carbon dioxide (CO 2 ) constitutes a large proportion of greenhouse gases, estimated to be 76% of all emissions . As suggested by the Intergovernmental Panel on Climate Change (IPCC) and concluded in the 2021 United Nations Climate Change Conference (COP26) in Glasgow, net-zero emission targeted by 2050 can only be achieved with CO 2 capture and storage (CCS) technology while some fossil fuels are to be “phased down” that still allowed for some developing countries. − CCS consists of two main processes: CO 2 capture and CO 2 storage. The former aims to capture CO 2 either from manufacturing industries before being released (i.e., cement and steel plants) or directly capture from air. , The latter is the following downstream to store CO 2 securely and permanently in subsurface structures via saline formations, salt caverns, or depleted petroleum reservoirs, namely, CO 2 geological storage (CGS) .…”