Introduction

In international rowing competitions, the standard race distance is 2000 m, with stroke repetitions typically ranging between 220 and 260 strokes per race cycle1,2,3. Stroke rates average between 32 and 38 strokes per minute, with an average power output of 450–550 watts per stroke4,5. Rowing performance generally lasts 6–8 min and relies heavily on both aerobic (65–80%) and anaerobic (20–35%) energy metabolism to complete the race efficiently6,7. This sport is characterized by repetitive muscle cycles and requires a high degree of physical fitness in terms of cardiovascular endurance, muscular fitness, and maximal strength8. Due to the repetitive nature of rowing movements, rowers are prone to injury from accumulated muscle fatigue over time, which can impair performance. Research shows that rowers with higher physical fitness levels tend to achieve better performance outcomes, highlighting the importance of enhancing physical conditioning to improve rowing performance. There is ample evidence supporting strength training as an effective method to improve muscle strength (maximal strength and endurance), which benefits performance in rowers of varying ages, genders, and training statuses9,10,11,12.

Rowing success requires strength, endurance, and effective training modalities. Strength and conditioning programs provide an efficient means for enhancing muscular fitness and athletic performance in both trained and untrained individuals9,10. Rowing training is structured on an annual cycle, with elite rowers committing over 13 h (or approximately 11 sessions) of training per week13. Among the various conditioning methods, plyometric training has been recognized as a safe, accessible, and time-efficient high-intensity training approach to increase muscular strength, power, endurance, and overall athletic performance14,15. This type of training has demonstrated notable improvements in maximal jump height and lower limb coordination16. Furthermore, plyometric training has shown to significantly improve peripheral fatigue, and enhance muscle explosiveness and neuromuscular efficiency17. It is widely used in the training of sports that require power output and acceleration capabilities, which emphasizing the role of the stretch–shortening cycle (SSC) mechanism18. Even in short durations (under 10 weeks), training effects are often apparent, encompassing both open- and closed-chain kinetic exercises18,19. Previous studies have also demonstrated that four weeks of plyometric training can effectively improve athletes’ lower-limb reactive strength index and leg stiffness20,21.

Although rowing is primarily an endurance sport, in actual competitions, especially during the start and sprint phases, rapid bursts of movement and high force output are required, thus placing certain demands on muscle explosiveness and power performance22,23. Given that rowing demands repetitive flexion and extension of both upper and lower limbs, plyometric training aligns well with the physical conditioning requirements for rowers. Current research supports the positive effects of plyometric training on rowing-specific performance metrics, such as increased rowing power14,24. Especially for young athletes, such training at a stage when the neuromuscular system is still plastic can further strengthen neural drive and movement control, and promote the development of overall sports performance25.

In addition to biomechanical parameters, training effectiveness is often evaluated with the use of surface electromyography (sEMG), which identifies muscle activity patterns during cyclic locomotor exercises, such as running, cycling, and rowing26,27. This tool provides valuable data on muscle activation timing and fatigue levels28,29,30. Insights on muscle activation during rowing can inform technical modeling to enhance athletic performance, reduce injury risk, and optimize training program design26,31. In assessing muscle function, fatigue tests often involve sustained or repeated muscle contractions. Muscle fatigue index is quantified by calculating the slope of the median frequency (MDF) of sEMG data, with studies showing a positive correlation between the MDF slope and the incidence of muscle injury32,33.

Plyometric training, when incorporated into rowing programs, has been reported to enhance peak rowing power more effectively than traditional conditioning methods14,24. While its benefits have been well documented in sports that require rapid force production34,35its specific application in rowing particularly regarding muscle fatigue resistance and rowing-specific performance remains underexplored. Rowing is a high-intensity sport that demands both substantial aerobic and anaerobic capacity6,7. Although a limited number of studies have shown that plyometric training can improve rowing power and strength-related performance parameters14,24further research is needed to clarify its role in fatigue mitigation.

Plyometric exercises have been employed to reinforce muscle groups susceptible to fatigue, potentially delaying the onset of fatigue and enhancing athletic performance36. However, it remains uncertain whether such training can effectively reduce muscle fatigue in rowers. Therefore, the primary aim of this study was to investigate the effects of a four-week plyometric training intervention on muscle fatigue and rowing-specific performance in young rowers. We hypothesized that the plyometric training group would exhibit lower MDF slope and improved rowing ergometer performance compared to the control group.

Methods

Study design

This randomized two-group study aimed to evaluate the sport performance and effectiveness of rowers in delaying muscle fatigue after completing different four-week preseason training programs (plyometric training versus routine strength and conditioning training). Both training plans were conducted under the supervision of coaches. Each participant underwent assessments on the Saturday at 7:00 a.m. both before the start and after the completion of the training program. The evaluations included body composition measurement, countermovement jump (CMJ), isometric mid-thigh pull (IMTP), and a 2000-meter rowing ergometer muscle fatigue test with simultaneous monitoring of physiological parameters via sEMG and heart rate (HR) sensors. Prior to all testing, participants performed a general and specific warm-up consisting of 10 min of light to moderate running, dynamic stretching, and submaximal attempts of each test exercise. All athletes were previously familiarized with the testing procedures. The same testing protocol was repeated post-training at the end of the four-week intervention. All assessments were conducted by the same researcher on both occasions to ensure internal consistency and reliability of the measurements.

Participants

Twenty-three youth male trained rowers (age range: 16.0-23.9 years) were recruited for participation. Eligible participants were current rowing team members engaged in regular, sport-specific training with regional or national competition qualifications. Participants were randomly assigned using random codes generated by a random number generator to either a control group (n= 12, standard strength and conditioning training; age: 19.1 ± 3.3 years; height: 169.3 ± 7.3 cm; body weight: 73.4 ± 12.5 kg) or an experimental group (n = 11, plyometric training adapted from, Egan-Shuttler et al., 2017, 2019; age: 18.2 ± 2.6 years; height: 171.7 ± 4.5 cm; body weight: 71.3 ± 8.6 kg) for a four-week group training protocol (Table 1). Exclusion criteria included: (1) any muscle injury or surgery within the last six months, (2) any gait abnormalities, and (3) the use of any medication affecting muscle tone. Participants were instructed to avoid high-intensity exercise, caffeine, and alcohol for 24 h before testing and to ensure adequate sleep the night prior (≥ 7 h)37. This study was conducted in full conformance with the relevant guidelines and regulations, i.e., the principles of the Declaration of Helsinki guidelines. This research protocol was approved by the Institutional Review Board of Fu Jen Catholic University (http://irb.rdo.fju.edu.tw/) (reference number: C110108). All participants fully understood the study procedures and provided signed informed consent before the testing began. For minor participants, their legal representatives also provided consent.

Table 1 A four-week plyometric training schedule (Adjusted from Egan-Shuttler et al., 2017, 2019).
Full size table

Intervention

Participants in the control and experimental groups completed a four-week training program (Table 1). The control group followed the standard rowing team training, which consisted of off-water routine strength and conditioning training combined with on-water rowing practice. The experimental group followed a plyometric training program alongside the same on-water rowing practice, with exercises focusing on vertical explosive power, such as box jumps, depth jumps, multiple box-to-box jumps, and double-leg hops. To enhance explosive triple extension for rowing, the program also included backward and overhead medicine ball throws (5 kg). Both groups performed either resistance or plyometric training for 30 min, three days per week, with 48-hour intervals between sessions (e.g., Monday, Wednesday, and Friday). A complete training cycle consisted of continuous training for four weeks.

Body composition measurement

Body composition was measured according to standardized procedures38. Participants’ body weight (BW), skeletal muscle mass (SMM), body mass index (BMI), percentage of body fat (PBF), and resting metabolic rate (RMR) were assessed using the InBody® 570 Body Composition Analyzer (Biospace, Inc., Seoul, Korea), with age and height manually entered. Height was measured with the H900 mechanical stadiometer (NAGATA Scale Co., Ltd., Tainan, Taiwan). To reduce measurement error, all participants wore lightweight athletic clothing and removed their shoes, socks, and any metal accessories before measurement.

Countermovement jump (CMJ) test

Participants performed a single CMJ on a triaxial force platform (Kistler9260AA, Kistler Ltd., Switzerland) with both feet. The movement requirements were as follows: Each subject was instructed to keep his torso as vertically aligned as possible, with his hands on his hips, and no arm swinging to assist the jump was permitted. They quickly squatted to approximately 90 degrees of knee flexion and then immediately jumped vertically with maximal force. The jump was considered complete upon landing. The force platform sampling rate was set to 1000 Hz. Each subject performed the test three times, with a one-minute rest between each attempt. The rate of force development (0–30 milliseconds; RDF30ms), peak force of CMJ (PFCMJ), jump height (JH), and modified reactive strength index (RSImod) were calculated in MATLAB software (version R2021a, MathWorks, Inc., Natick, MA, USA) as per Gathercole et al. (2015)39. All force parameters were normalized to body weight, and the data from the three trials were averaged.

Isometric Mid-thigh Pull (IMTP) test

Participants were assessed in an IMTP rack (Kairos Strength, Murphy, NC, USA). A force platform (Kistler9260AA, Kistler Ltd., Switzerland) was placed at the center of the IMTP rack, with the barbell positioned at approximately mid-thigh height (50% of the distance between the greater trochanter and the lateral epicondyle of the knee), allowing participants to maintain a standardized pulling posture. The barbell height could be adjusted using an indicator. Participants were instructed to “pull as fast and as forcefully as possible” and maintain maximal effort for at least 3 s. The force platform sampling rate was set to 1000 Hz, and participants completed three trials with a one-minute rest between each. The rate of force development (0–200 milliseconds; RDFIMTP) and peak force of IMTP (PFIMTP) were calculated in MATLAB software (version R2021a, MathWorks, Inc., Natick, MA, USA), as per Comfort et al. (2019)40. All force parameters were normalized to body weight, and the data from the three trials were averaged.

2000-meter (2000 m) Rowing ergometer time trial test with sEMG collection

Prior to the rowing ergometer time trial, EMG electrodes (DelsysTrigno® Wireless EMG systems, Boston, MA, USA) were affixed to four muscle sites on the dominant side (erector spinae [ES], biceps femoris [BF], vastus medialis [VM], and gastrocnemius [GN]), which are known to contribute significantly to rowing performance31,41,42. Electrode placement followed the procedures outlined by Gee et al. (2023)25and a chest strap HR monitor (Polar H10, Polar Electro Oy, Finland) was used to collect data for each 2000 m trial. This EMG system had 16 channels, with a sampling rate set to 2000 Hz. Participants completed a five-minute self-paced warm-up on the rowing ergometer (RowErg model D, Concept2, US), rested for 2 min, and then completed a single 2000 m time trial.

To quantify muscle fatigue, the MDF slope was analyzed, as it is a widely used indicator in fatigue-related analyses43,44,45. MDF is defined as the frequency that divides the power spectrum of the EMG signal into two equal halves. A progressive decline in MDF over time reflects a shift toward lower frequency components, which is typically associated with decreased muscle fiber conduction velocity and the onset of neuromuscular fatigue45. Due to individual differences in 2000 m completion times, the MDF slopes of the sEMG signals were calculated at every 500 m (e.g., 0 m, 500 m, 1000 m, 1500 m, and 2000 m). Ten consecutive strokes were recorded at each 500 m interval; at the start (0 m) and end (2000 m), the initial and final ten strokes were recorded, respectively46. Raw sEMG signals were filtered with a zero-lag band-pass Butterworth filter at 10 and 500 Hz cut-off frequencies47. sEMG data were analyzed in MATLAB software (version R2021a, MathWorks, Inc., Natick, MA, USA).

Statistical analysis

The sample size was estimated with a software program (G*Power v3.1.9.7; Heinrich-Heine-Universität Dusseldorf) with an effect size of 0.5, alpha error of 0.05, and statistical power of 0.61 in the analysis of covariance (ANCOVA) statistical test. Thus, the minimum number of subjects required for the study was determined to be 23.

In this study, experimental data were analyzed in IBM SPSS, version 20 (IBM Corp., New York, NY, USA). The Shapiro–Wilks test was applied to verify the normality of data distribution. The ANCOVA statistical test was conducted to examine the differences in the effects of the two training methods on muscle fatigue and training outcomes. In this study, change scores (post-training minus pre-training values) were used as the dependent variables. Partial ƞ2 was calculated to assess effect sizes. The partial ƞ2 value of 0.01 indicates a small effect, 0.06 a medium effect, and 0.14 a large effect48. Statistical significance was set at α = 0.05, and all data are presented as mean ± standard deviation.

Results

All participants successfully completed the 4-week training program and performance tests. Table 2 shows participant characteristics at baseline and following the 4-week intervention. Shapiro–Wilks tests confirmed that all continuous data had normal distributions (p > 0.05), fulfilling the assumptions for parametric tests.

Table 2 The characteristics at baseline and following the 4-week intervention of the participant.
Full size table

For the exercise performance tests, ANCOVA results for pre- and post-test measurements in both the control and experimental groups (variables CMJ, IMTP, and 2000 m rowing test) are presented in Table 3. In the CMJ test, significant differences between the experimental and control groups after four weeks of training were found in RFD30ms (F(1,21) = 4.53; p = 0.046, ƞ2 = 0.19), RFD30ms/BW (F(1,21) = 4.59; p = 0.045, ƞ2 = 0.19), PFCMJ (F(1,21) = 4.53; p = 0.046, ƞ2 = 0.19), PFCMJ/BW (F(1,21) = 5.22; p = 0.033, ƞ2 = 0.21), JH (F(1,21) = 9.08; p = 0.007, ƞ2 = 0.31), and RSImod (F(1,21) = 6.27; p = 0.021, ƞ2 = 0.24). In the IMTP test, only PFIMTP (F(1,21) = 6.68; p = 0.018, ƞ2 = 0.25) showed significant differences between the experimental and control groups after four weeks of training, whereas no significant differences were observed for RFD200ms (F(1,21) = 1.76; p = 0.199, ƞ2 = 0.08), RFD200ms/BW (F(1,21) = 1.28; p = 0.271, ƞ2 = 0.06), or PFIMTP/BW (F(1,21) = 2.78; p = 0.111, ƞ2 = 0.12). In the 2000 m rowing test, significant differences between groups after four weeks were observed in rowing time (F(1,21) = 7.17; p = 0.014, ƞ2 = 0.26) and power output (F(1,21) = 4.90; p = 0.039, ƞ2 = 0.20), but not for %HRmax (F(1,21) = 0.12; p = 0.728, ƞ2 = 0.01).

Table 3 Results of an analysis of covariance (ANCOVA) with time as a covariate showing the significance of the effects of exercise performance during the time course of the experiments control group and experimental group before and after four-week training.
Full size table

Regarding muscle fatigue analysis, Fig. 1 illustrates the decline in MDF for each 500 m segment during the 2000 m rowing time trial, before and after the 4-week training period, in both the control and experimental groups. Table 4 lists the ANCOVA results and descriptive statistics for MDF slopes under different training conditions for the ES (F(1,21) = 6.48; p = 0.019, ƞ2 = 0.25), BF (F(1,21) = 10.73; p = 0.004, ƞ2 = 0.35), VM (F(1,21) = 12.14; p = 0.002, ƞ2 = 0.38), and GN (F(1,21) = 8.59; p = 0.008, ƞ2 = 0.30). These findings indicated that the MDF slopes for the ES, BF, VM, and GN in the experimental group were significantly lower in the experimental group than in the control group after four weeks of training.

Fig. 1
figure 1

Changes in MDF across 500 m intervals in a 2000 m rowing test: Pre- and Post-four-week training in control and experimental groups. CG = control group; EG = experimental group; MDF = median frequency. *Indicates that MDF slope of EG significantly difference from CG, p < 0.05.

Full size image
Table 4 Results of an analysis of covariance (ANCOVA) with rowing distance as a covariate showing the significance of the effects of MDF slope during the rowing distance course of the experiments control group and experimental group before and after four-week training.
Full size table

Discussion

Further understanding of how training affects muscular fitness, particularly under conditions of fatigue, can help in preventing injury and optimizing athletic performance. In the current study, we conducted a four-week intervention with rowers (plyometric training vs. routine strength and conditioning training). The results showed that after four weeks of training, the experimental group (plyometric training) demonstrated greater improvements in biomechanical parameters, including explosive power, maximal strength, and rowing performance, compared to the control group. Additionally, the experimental group exhibited a lower MDF slope, indicating enhanced fatigue resistance. These findings suggest that plyometric training effectively improved rowing performance.

Plyometric training has been suggested to enhance strength and power output across various age groups of athletes49,50. Consistent with previous studies, the rowers in our study displayed significant increases in (1) explosive performance metrics, such as RFD30ms, RFD30ms/BW, PFCMJ, PFCMJ/BW, JH, and RSImod; (2) maximum strength metrics, such as PFIMTP; and (3) rowing time trial performance metrics, including rowing time and power output. These results align with past findings, as plyometric training is commonly employed to enhance strength performance and sport performance14,51,52. From the perspective of the muscle SSC mechanisms and training science, smoother and faster reactive mechanisms aid explosive power and maximal strength performance. Regarding rowing-specific techniques, the rowing action involves repetitive muscle cycles and is physiologically demanding53. Increasing power output and reducing cycle time per stroke, while improving movement quality, ultimately lowers rowing time trial durations. Egan-Shuttler et al. (2017)14 further indicated that rowers with higher power output may exhibit greater technical proficiency, thus conserving energy during rowing.

On the other hand, plyometric training has also been shown to rapidly establish neuromuscular connections, thereby enhancing muscle strength, athletic performance, and the body’s capacity to absorb impact54,55,56. However, there remains limited evidence on whether it can delay the onset of muscle fatigue in rowers during time trials. Using MDF in sEMG as a parameter to quantify muscle fatigue, with the slope over time (MDF slope) as the muscle fatigue index47a steeper slope indicates increased muscle fatigue, and vice versa. A decline in the MDF slope during sustained or repeated contractions typically reflects a shift toward lower-frequency components in the sEMG signal, which is associated with motor unit fatigue, decreased conduction velocity, and increased recruitment of slower, more fatigue-resistant muscle fibers32,33. Our study results demonstrated that, after the four-week intervention, the MDF slope in sEMG from the 2000-meter rowing ergometer time trial showed significant reductions in fatigue in the ES, BF, VM, and GN muscles in the plyometric training group, whereas the control group, which only maintained routine training, did not achieve effective fatigue reduction. Plyometric training enhances neuromuscular coordination, motor unit recruitment efficiency, and the ability to maintain high firing rates, particularly in fast-twitch muscle fibers57. These adaptations may reduce the rate of MDF slope decline by improving the muscle’s ability to sustain power output under fatigue-inducing conditions58.

The biomechanics of rowing are complex, with numerous variables influencing boat speed. Coordination of both the upper and lower limbs is required to optimize boat propulsion59,60. In competition, rowers repeat stroke cycles over 220 times2heavily relying on aerobic (65–80%) and anaerobic (20–35%) metabolism to cover the race distance as quickly as possible6,7. As lower limb muscles play a key role in rowing, serving as the main source of propulsion in the drive phase, they are at higher risk of fatigue and injury due to the repetitive, high-load cyclic movements. Appropriate training methods can thus enhance muscular fitness in ways that target sport-specific demands, boosting explosive power, maximal strength, and technical skill, while effectively reducing or delaying muscle fatigue.

In conclusion, with the rapid development of technology in performance monitoring and computational capacity, the relationship between sports science and athletic training will continue to grow. Throughout training cycles, technological monitoring will have a profound impact on athletes during both training and competition preparation phases, enabling athletes, coaches, and sports physiologists to accurately assess training status, neuromuscular function, and fatigue levels in a timely manner to prevent injuries. However, certain limitations remain in this study. First, although the statistical analyses were appropriately conducted, the relatively small sample size may limit statistical power and the ability to detect smaller but potentially meaningful effects. Future studies with larger cohorts are warranted to validate and generalize these findings. Second, the study included only male high school rowers, which limits the generalizability of the results to female athletes or adult populations. Third, as with many training-based interventions, the potential influence of placebo effects or increased motivation due to participation in a supervised program cannot be entirely ruled out, and should be considered in future experimental designs. Future research will focus on periodized training impacts on muscular fitness and fatigue in hope of further improving training outcomes and preventing sports injuries.

Practical applications

Rowing is a competitive sport that demands exceptional cardiorespiratory endurance and muscular fitness. Given that rowers dedicate extensive time to training, often exceeding 13 h per week, it is essential to prioritize efficiency and safety in their training regimen. Based on the results of this study, four weeks of plyometric training effectively delayed/reduced the decline in MDF of the ES, BF, VM, and GN muscles, enhanced explosive power and maximal strength, and improved rowing performance. Plyometric training increased leg muscle fitness and coordination, facilitating faster and more complete movements in the sliding seat and thereby boosting total power output during neural transmission and muscle contraction. For rowers, executing powerful, high-velocity strokes on the water is a primary training goal. Mitigating muscle fatigue and enhancing athletic performance in training planning are essential, as confirmed by our findings on the positive effects of plyometric training in rowers.

Data Availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.