The review was performed utilizing the SCOPUS and Web of Science (WOS) databases to assess relevant studies published between 1974 and 2023.

The initial search was conducted using the Boolean operators “AND” and “OR,” combined with keywords related to the subject area, such as “channel morphology (river morphology, channel cross-section, channel erosion, channel aggradation and riverbank erosion, channel degradation, incision, narrowing, widening, bed level),” “dam (reservoir, hydroelectric power plant),” and “hydrology (stream flow, peak flow, discharge, sediment yield, sediment load)” etc. to identify and select papers for review.

Eligibility of references was determined based on predefined inclusion and exclusion criteria. Only original research articles published in English in peer-reviewed journals focusing on river morphological responses to dam construction were included. Books, conference papers, review articles, and presentations about channel changes were excluded from the study. In the initial research, after removing duplicate papers using EndNote X6, based on the title assessment, some papers were excluded. Subsequently, papers with unrelated topic, duplicate content, and manuscripts with lacking clear methodology, evaluation criteria, or results were also excluded. Importantly, this study focuses exclusively on the effects of dams on river morphology; other anthropogenic influences such as land use changes, mining, or urbanization were not considered.

The preliminary search results were imported into Endnote X6, and duplicates were removed. Two independent authors then screened the titles and abstracts of all the papers and omitted some irrelevant papers based on the eligibility criteria. Afterward, the full texts of the remaining papers were downloaded and thoroughly reviewed to ensure that they meet the criteria for selection. Finally, eligible papers were systematically analyzed to extract relevant data. The following information was collected from each article: fist author, year of publication, river name, geographical location, timeframe of observed changes, dam name, location relative to dam (upstream or downstream), climatic zone, river pattern, data collection method, and observed morphological changes.

Finally, the papers were classified according to type of dam-induced morphological change (e.g., widening, narrowing, deepening, shallowing), position relative to the dam (upstream or downstream), river pattern (single-thread, multi-thread), location of the study area (continent, climatic zone [Köppen-Geiger (https://koeppen-geiger.vu-wien.ac.at/)] and method used for quantifying morphological changes (e.g., topographic measurements, remote sensing material, hydrological data) (see Supplementary Table S1).

A combined search of titles, abstracts, and keywords was conducted. First, the literature search was conducted using the SCOPUS database, which resulted in 172 articles found. Then, an additional search was carried out using the Web of Science database, resulting in 118 articles indicated. In total, 290 unique records were identified across both databases (n = 290). Subsequently, the database was reviewed to remove duplicates (n = 47) and articles not directly related to the study topic (n = 62). During the subsequent phase, a total of 181 records were assessed based on the title and abstract content, taking into account the exclusion criteria such as irrelevant keywords (e.g., check dam, other human intervention, experimental and flume studies). As a result, 59 articles were eliminated. The whole texts of the remaining records (n = 122) were evaluated to determine their eligibility. From 122 records, the articles which did not have enough data on river morphological adjustments (n = 21) and those have the effect of dam combined with other human interventions (n = 6) were excluded (n = 27). Finally, 95 articles met all inclusion criteria and were included in the systematic review (Figure 1).

An analysis of the existing literature revealed a notable increase in the number of publications addressing dam-induced morphological changes between 1974 and 2023 (Figure 2). This is parallel to the increase in dam construction, so that according to Zhang and Gu (2023), over the past three decades, dams have increased especially in developing countries in Asia and South America. The geographic distribution of the 95 reviewed studies is presented in Figure 3. Among these, 57 studies (60%) were conducted in Asia, followed by 16 (17%) in Europe and North America, 4 (4%) in South America, and 2 (2%) in Africa (Table 1). Also, number of the rivers and dams located in countries is shown in Table 1. The most frequently cases are in Asia from China (38 papers), primarily related to the Yangtze and Yellow Rivers (Table 1; Supplementary Table S1). The Yangtze River basin has area of 1.8 × 10⁶ km² with 6,300 km river length and the Yellow River basin is 752,443 km² with 3,471 km length (Supplementary Table S2). These rivers are among longest river in the world which have several dams built on them and their tributaries (Supplementary Table S1). Regarding climatic distribution, 59 studies (62%) were conducted in temperate climates, followed by 21 (22%) in semi-arid, 11 (12%) in continental, and 4 (4%) in tropical climates (Figure 4; Supplementary Table S1). Therefore, these studies are scattered across different climatic regions. The Köppen climate classification scheme divides climates into five main climate groups: A (tropical), B (arid), C (temperate), D (continental), and E (polar). Each group and subgroup is represented by a letter. The second letter indicates the seasonal precipitation type, while the third letter indicates the level of heat. According to Supplementary Table S1, the climate of study locations define as: Aw: Tropical wet and dry (savanna climate); BSh: Hot semi-arid climate; BSk: Cold semi-arid climate; Cfa: Humid subtropical climate; Cfb: Temperate oceanic climate or subtropical highland climate; Csa: Hot-summer Mediterranean climate; Csb: Warm-summer Mediterranean climate; Cwa: Monsoon; Dfa: Hot-summer humid continental climate; Dfb: Warm-summer humid continental climate; Dwb: Monsoon-influenced warm-summer humid continental climate.

A significant proportion of the reviewed studies (76 papers, or 80%) focused on the downstream effects of dams on river morphology, while a significantly smaller number of papers investigated the upstream effects (13 papers, or 14%) or both upstream and downstream effects (6 papers, or 6%) (Table 1). In terms of river types, most studies (71; 75%) were conducted on single-thread channels, while 11 studies (12%) focused on multi-thread systems, and 13 studies (13%) included both types. Most of the studies have been carried out in reaches located in alluvial (76.84%), alluvial and mountainous (10.53%), mountainous (7.37%), alluvial and coastal (3.16%), and coastal plains (2.11%), respectively. The shortest reach studied was 1.5 km upstream of a dam in Smolnik, Poland, while the longest extended 1,300 km downstream along the Lower Yangtze River, China (Supplementary Table S1).

In terms of morphological adjustments, channel incision (48; 56%), narrowing (24; 28%), have been most frequently reported in the river reaches downstream from dams. Narrowing ranged between 10% and 75%. The highest amount of channel narrowing (75%) have been reported by del Tanago et al. (2015) in downstream of Guadalete river in Spain (Supplementary Table S1). This study reach is single-thread in morphology and located at temperate climate (Csa). Also, Wu et al. (2023) stated that channel narrowing (60%) has been occurred in downstream reach of Lower Yellow River as a multi-thread river in temperate climate (Cwa). About bed-level changes, the highest incision over 5 m reported in downstream of the single-thread and multi-thread rivers by Kale and Ataol (2021) in Yeşilırmak (Turkey) and Kondolf and Swanson (1993) in Stony Creek (California). In contrast with downstream, channel widening (6; 32%) and aggradation (9; 47%) have been most frequent reported in the river reaches upstream from dams (Figures 5, 7). The highest channel widening up to 600% has occurred in upstream of Magdalena River in Colombia with tropical climate (Alvarado et al., 2023). Only just few papers about channel aggradation were stated but these papers didn’t report amount of aggradation.

In the case of initially multi-thread rivers located upstream of dams, long-term channel narrowing and pattern simplification have been observed (e.g., Xu, 1990; Xu, 2001; Xu and Shi, 1997), although some channel width fluctuations may occur during the initial phase of adjustment (Liro, 2016). For example, Liro (2015), Liro (2016) reconstructed that in a gravel-bed river, a rise in the downstream base level and the onset (∼first 20 years) of backwater fluctuations in upstream channel sections led to channel widening, driven by accelerated in-channel deposition, bar development, and associated bank erosion.

From river pattern, in the case of downstream-dam reaches a few papers reported a decrease in sinuosity (6 papers) (e.g., Xu, 1990; Morais et al., 2016; Yousefi, et al., 2016; Kale and Ataol, 2021; Alvarado, et al., 2023; Clavijo-Rivera, et al., 2023) and braiding index (4 papers) (e.g., Surian, 1999; Liro, 2017; Kumar, et al., 2019; Kong, et al., 2020), but only one study reported these effects in upstream-dam sections (Supplementary Table S1). Eight papers noted a decrease in bar areas downstream of dams in just a few case studies, including transitions from wandering to single-thread (1 paper), braided to single-thread (1 paper), and braided to wandering (1 paper). For instance, Kumar et al. (2019) found that the Wazirabad Dam in India caused a decline in bar area, a reduced braiding index, and a narrower channel belt downstream of the Yamuna River.

After operation of dam, reducing peak discharges and sediment may lead to changes in lateral channel migration in downstream of dams (e.g., Phillips et al., 2005; Gordon and Meentemeyer, 2006; Musselman, 2011; Smith and Mohrig, 2017; Kong et al., 2020; Liu et al., 2021; He et al., 2022; Goudie, 2022). Also coarsening of channel sediment in downstream of dam was reported in few papers (e.g., Phillips et al., 2005; Smith and Mohrig, 2017; Gao et al., 2021; Yang et al., 2023). In this regard, only one paper by Maselli et al. (2016) stated that in Isola Serafini dam in Italy, lateral migration rate of the channel is up to 45 m/yr upstream of the influence of backwater flow and ca.10 m/yr at the transition from normal to backwater flow conditions (30 km from the dam) and deposition of coarse-grained sediment.

It should be noted that by alteration of hydrology, sediment and morphology, vegetation cover changes. In some cases, sedimentation and encroachment of vegetation in single-thread reaches have been reported as a result of reduced flow downstream of dams (e.g., Choi et al., 2005; Asaeda and Rashid, 2012; del Tanago et al., 2015; Yang et al., 2023). Only Liro et al. (2022) demonstrated that backwater fluctuations of Ro˙znów dam on upstream of Smolnik River substantially disturb the hydrodynamics, which is lead to alterations in sedimentology, morphology, and riparian vegetation pattern. Liro (2019) suggested that changes in the river hydrodynamics and the deposition of fine sediments in upstream of dam, may be the factors triggering vegetation expansion. However, the same factors may eliminate plants that are not resistant to prolonged inundation and high sedimentation rate. Up to now there have been no works presenting a spatial–temporal reconstruction of this process and its relation to the geomorphic characteristics of a given site in the backwater.

These morphological adjustments were most commonly reconstructed using remote sensing materials, including satellite images, aerial photos, and maps (37 papers; 39%), data from field measurements (19; 20%), and historical hydrological data (8; 8%). A high number of works (31; 33%) utilized more than one data collection method (Figure 6). The use of remote sensing techniques has increased significantly in studies from 1992 to 2023. According to Chong et al. (2021), these remote techniques are increasingly replacing field-based approaches to studying geomorphic processes and changes. After gathering data on channel morphology pre and post-dam periods, the researchers have used the statistical analysis and/or hydrological modeling for distingutiong channel changes due to dam construction.

Finally, it is possible that channel changes have environmental and ecosystem effects, but such effects have not been evaluated broadly in the existing studies. However, few of the reviewed papers mentioned ecological and management consequences of channel adjustments in downstream of dams and only one paper mentioned these in upstream. The work of Kondolf and Swanson (1993) in Stony Creek, showed that the change from a braided channel pattern to a more sinuous and meandering channel, probably reflecting an adjustment to abruptly reduced rates of bedload sediment supply from the watershed, to the decrease in flood peaks and sediment transport capacity, and possibly to prolonged bankfull flows. Phillips et al. (2005) has reported intensive bank erosion downstream of dam in Lower Trinity. Sanchis Ibor1 et al. (2018) reported that transformation from wandering to single thread channel pattern in Serpis river was followed by a slow vegetation encroachment, and culminated with the stabilization of channel migration. The role of vegetation has been critical in controlling floods' effectiveness, reducing river mobility and shifting, and consolidating a channel planform model adapted to the post-dam flow conditions. Wu et al. (2018) demonstrated that after construction of Zhikong Reservoir on Lhasa River, the number and area of central bars in the braided reach closest to Zhikong Dam (RS1) were increased because of main channel incision and water level drop. Kale and Ataol (2021) mentioned that river degradation due to the effects of dams on Yeşilırmak River has a major environmental problem at the historical Carşamba Bridge and also along the riverbed of the Yeşilırmak River after dam construction. Agarwal et al. (2022) said that the ability of the Srinagar dam to adjust the sediment load of the event in its upstream is an immediate short-term but it creates a geomorphic disconnect in the channel between upstream and downstream reaches of the reservoir. This disconnect may have a negative effect on sediment–water routing of the river and sediment storage. About upstream, Liro et al., (2022) mentioned that the fluctuation effects of reservoir backwater have caused the change of channel pattern from the multi-thread to the single-thread. This increased the riparian forest expansion in the stream valley and decreased the floodplain utilization potential and caused the development of homogenous and simplified morphological and vegetation patterns.