[Objectives] Hydraulic engineering structures, particularly dams, disrupt river connectivity and pose significant threats to fish populations that rely on downstream migration. While upstream fish passage facilities have received considerable attention, ensuring safe and efficient downstream passage remains a critical and less-solved challenge in eco-hydraulics. The high mortality rates associated with downstream migration, especially through turbines, spillways, and other infrastructure, jeopardize fish conservation and the ecological sustainability of hydropower. This review aims to systematically synthesize global research on the behavioral characteristics and hydrodynamic requirements of fish during downstream migration. The primary objectives are: (1) to consolidate the known hydrodynamic thresholds (e.g., velocity, turbulence, shear stress) for various fish species; (2) to evaluate the design and efficacy of different downstream passage facilities in relation to these hydrodynamic needs; and (3) to propose a integrated framework for future research and facility design that moves beyond single-parameter approaches to a holistic "fish-flow system" perspective. [Methods] This study constitutes a systematic review of the scientific and engineering literature concerning fish downstream migration. The methodological approach involved a multi-stage analytical process: Literature Synthesis and Categorization: A comprehensive body of international research, including peer-reviewed journals, technical reports, and conference proceedings, was systematically gathered and analyzed. The literature was categorized along three primary themes: (1) empirical studies on fish behavior (e.g., active vs. passive descent) in response to hydrodynamic factors (velocity, turbulence, shear, acceleration, pressure); (2) technological assessments of different downstream passage facilities (bypasses, turbines, fish collection systems, specialized channels); and (3) advancements in research methodologies, encompassing in-situ field monitoring, laboratory flume experiments, and numerical modeling techniques like Computational Fluid Dynamics (CFD) and individual-based models (IBMs). Comparative Analysis and Critical Evaluation: Within each category, findings were compared and evaluated. This involved synthesizing quantitative data (e.g., preference velocity thresholds, injury limits) into consolidated summaries (Table 1) and qualitatively assessing the strengths, weaknesses, and operational challenges of different passage technologies. Special attention was paid to inconsistencies, research gaps, and the applicability of findings across different fish species and geographies. Framework Development: Building on the synthesized evidence and identified gaps, a new conceptual framework was developed. This framework was designed to provide a systemic lens for understanding and designing downstream passage. [Results] Firstly, fish downstream behavior is systematically categorized into active descent (head-first, efficient) and passive descent (counter-current, inefficient), with transitions between these states triggered by specific hydrodynamic conditions. Preference and injury thresholds for critical hydrodynamic parameters have been quantified for several species. For instance, juvenile grass carp exhibit a preference velocity of 0.19-0.49 m/s, while Atlantic salmon smolts prefer 0.38-0.73 m/s. Turbulent kinetic energy (TKE) below 0.03 m²/s² is generally preferred, with higher TKE leading to disorientation and inefficient passage. Crucially, injury thresholds for shear strain rate vary significantly among species, from 500 s^-1 for trout to 2179 s^-1 for juvenile Chinese carps. Similarly, pressure change gradients exceeding 50 kPa/s during descent or 15 kPa/s during ascent can cause barotrauma. The critical evaluation of passage facilities reveals distinct performance profiles. Surface-oriented bypasses and specialized downstream channels (e.g., fish slides) generally offer a higher survival rate by leveraging fish surface-orientation and providing low-turbulence pathways. Despite improvements that can increase fish survival rates to over 97% in fish-friendly turbines, they still represent the highest-risk passage route compared to alternative passage routes. Collection and transportation systems are effective for high dams and complex terrain conditions, but require intensive operational maintenance and incur high costs. Furthermore, many fish passage facilities demonstrate poor attraction efficiency at their entrances. One of the most significant result of this review is the proposal of an innovative, three-dimensional framework for defining and designing hydrodynamic conditions for downstream passage. This framework posits that effective passage requires the simultaneous satisfaction of three interconnected criteria: Necessity: Hydrodynamic conditions (e.g., specific "preference velocities" and "low-turbulence windows") that actively attract fish and initiate downstream movement into the facility. Safety: Conditions that prevent injury or mortality, defined by rigid thresholds for damaging forces (e.g., shear strain rate, pressure gradient) throughout the entire passage route. Timeliness: Conditions that promote efficient and timely passage by minimizing behavioral delays (e.g., "non-direct descent" caused by adverse acceleration flows), ensuring fish are not energetically depleted or exposed to predators for extended periods. [Conclusions] This review concludes that the prevailing approach, which often focuses on isolated hydrodynamic parameters, is insufficient for designing highly effective downstream passage facilities. The path forward requires a paradigm shift towards an integrated "fish-flow system" approach, as embodied by the proposed Necessity-Safety-Timeliness (N-S-T) framework. The ideal downstream passage is not merely a conduit with non-lethal hydraulic conditions; it is a system where the entrance flow (Necessity) seamlessly connects with a guiding, efficient internal flow (Timeliness), all while rigorously excluding hazardous forces (Safety). Future research can prioritize: (1) multi-factorial studies that explore the synergistic effects of coupled hydrodynamic parameters on fish behavior; (2) expanded research on adult fish and a wider range of non-salmond species, particularly those prevalent in Asian rivers; (3) the systematic development of a comprehensive, open-access database of hydrodynamic requirements; and (4) the integration of advanced monitoring technologies (e.g., AI-powered sonar, drones) and predictive numerical models for both design optimization and post-construction evaluation. By adopting this holistic perspective, the goal of harmonizing hydropower generation with sustainable fish conservation becomes increasingly attainable.