### Abstract
Many studies have examined the relationship between 2000m rowing ergometer performance and physiological variables, often suggesting that rowing ergometer performance models can be used to predict on-water performance. While studies have examined the kinematic, oxygen consumption, and electromyography similarities between rowing ergometry and on-water rowing, this is the first study to examine the relationship between physiological variables measured on the ergometer and 2000m performance on the water.
Nineteen elite heavyweight male rowers (26.2 ± 3.6 years, 92.2 ± 4.3 kg, 192.2 ± 4.5 cm) participated in the study. All testing was done over a two-week period. A 2000m (2K) on-water pursuit time trial in single scull boats, where the athletes started 30 seconds apart and competed over a 2K course for time, and a 2K ergometer performance test, were conducted on consecutive days in the first week. A progressive continuous incremental ergometer VO2 max and a modified 45s rowing Wingate test, to measure peak (Peak 45) and average power (Ave 45), were performed a week later. Ventilatory threshold (VT) was determined from a plot of VE/VO2. All ergometer testing was done on a Concept II model C rowing ergometer.
While Pearson correlations showed that VO2 max (r = -0.55, p< 0.05), Peak 45 (r = -0.43, p< 0.05), and Power at VO2 max (r= -0.733, p< 0.05) were significantly related to 2K ergometer performance, there was no correlation between any of the measured variables and 2K on-water performance (Height, r=0.273; Weight, r=0.373; VO2 max, r=0.049; VT, r=0.043; Peak 45, r= -0.229; Ave 45, r= -0.200; Power at VO2 max, r= -0.292). Additionally, there was no correlation between 2K ergometer performance scores and 2K on water performance scores (r= 0.120). The data suggest that physiological and performance tests performed on a rowing ergometer are not good indicators of on water performance. While it is a common practice for many, rowing coaches and sport scientists should be cautious when using the rowing ergometer to predict on water performance or select rowing crews.
**Key Words:** Rowing, physiological tests, ergometer, on water
### Introduction
The rowing ergometer (erg) has become an important tool for training and physiological monitoring of rowers. The erg has allowed sport scientists and researchers to overcome environmental factors such as current, water temperature, and wind, that can make physiological monitoring and research on rowers difficult. As a result researchers have been able to study and describe the relationship between a variety of physiological variables and rowing ergometer performance (2,10,11,13,16 ).
The repeatability of results within a brand and between models of a brand of ergometer (12,15) is quite good but there are differences in the physiological responses to rowing on different ergometers (3,6,7). While the ergometer has aided in the advancement of the body of knowledge on rowing, rowers criticize the “feel” of the ergometer compared to rowing on water. During the recovery phase of the rowing stroke on water, the mass of the boat slides underneath the rower (7). On most brands of rowing ergometer the opposite occurs, the mass of the rower must move up and down the slide bar during recovery and leg drive (7).
Other than the single study where Urhausen, Weiler, and Kindermann (14) examined the differences in the heart rate-lactate relationship between on-water and ergometer rowing and found that, for a given level of lactate, heart rate values were significantly higher on the water, the relationship between rowing ergometer physiological data to on-water performance has not been studied extensively. The purpose of this study was to examine the relationship between rowing ergometer physiological and performance data to 2000m on-water rowing performance.
### Methods
#### Participants
Nineteen heavyweight male rowers (26.2 3.6 years, 92.2 4.3 kg, 192.2 4.5 cm) who were part of a Canadian National Team training camp participated in the study. Sixteen of the subjects went on to compete at the World Championships. All testing was done over a two-week period. All athletes agreed to participate in physiological monitoring as part of their training program. All procedures were approved by the Rowing Canada Sports Science and Medicine Committee.
#### Rowing time trial (2K row)
A 2000m time trial in single scull boats was performed on the first day of the investigation. Athletes reported to the lake at 7:00 am and were given 60 minutes to prepare for the time trial. Each athlete performed a self-selected warm-up similar to what they would use before racing. Following the warm-up, athletes reported to the start line and were sent off 30s apart. The time trial was also being used for ranking in the team selection process; increasing the athlete’s motivation to perform well. All athletes were familiar with the technique involved in sculling, having raced or trained in sculling boats in the previous six months.
#### Maximal 45s sprint test
A modified Wingate sprint test on the Concept II Model C rowing ergometer was performed 90 minutes following the maximal oxygen uptake test. Subjects performed a self-selected warm-up for 10 minutes. After the warm-up the ergometer was programmed for a 45s trial and the damper was set to provide a “drag factor” of 200, the maximum that is normally attained on a used ergometer. Because dust, worn parts and other factors can affect the amount of resistance provided by each stop in the resistance control dial, the “drag factor” is the method used on the Concept II Model C ergometer for standardizing the resistance setting between ergometers.
Participants performed an all-out 45s effort with verbal encouragement. Participants were asked to row full strokes on each stroke of the test rather than use the partial strokes that are often incorporated at the start of a race. Power (W) for every stroke was calculated and displayed on the Concept II computer and recorded by the investigators. Peak power (Peak 45) was the highest power obtained on any individual stroke. Mean power (Mean 45) was the average of the individual stroke power over the 45s trial as calculated by the Concept II computer.
#### Ergometer time trial (2K erg)
On the second day of the investigation all subjects completed a 2000m rowing ergometer time trial. Subjects reported to the ergometer centre at 7:00 am and were given 60 minutes to prepare for the time trial. They followed similar warm-up procedures to those they did for the on-water trial the previous day. All subjects started the time trial at the same time to create a competitive atmosphere. Prior to the start of the time trial each rowing ergometer was calibrated to a “drag factor” of 120, which is the drag factor that was in use for all selection-based testing.
#### Maximal oxygen uptake (VO2 max)
One week after the ergometer time trial, maximal oxygen uptake was measured using a Parvomedics True One metabolic cart (Parvomedics Inc, Park City Utah). Subjects performed a continuous incremental test on the Concept II Model C rowing ergometer. All subjects started at 290 watts, increasing wattage by 30 watts every three minutes throughout the test. The test was stopped when the subjects reached volitional fatigue or were unable to complete a stage within five watts of the intended wattage. Power at VO2 max (VO2 power) was determined as the average power for the final stage of the test as calculated by the Concept II computer.
#### Anaerobic Threshold (AT)
Anaerobic threshold (AT) in L/min of oxygen and power at anaerobic threshold (AT power) were determined from a plot of VE/VO2 using the procedure described by Caiozzo et al (1982).
#### Statistical Analysis
Pearson correlation coefficients (r) were calculated to establish the relationship between rowing ergometer physiological parameters and on-water rowing performance. A Student T-test was used to determine whether a difference existed between on-water rowing 2K time and 2K ergometer time. Statistical significance was determined using a probability level of p<0.05.
### Results
The 2K ergometer times (6:05.4± 5.5s) were significantly faster than the 2K row times (7:35.7 ±11.4s) (p<0.05). There was no correlation between 2K ergometer performance scores and 2K row performance scores (figure 1).
The mean VO2 max score was 5.9 ±0.4 L/min or 63.7±4.1 ml/kg/min. Power at VO2 max averaged 442.5 ±25.5 Watts. Both VO2 power (figure 2) and VO2 max (table 1) were significantly correlated to 2K erg but not to 2K row.
Anaerobic threshold (AT) occurred at 4.9 ± 0.3 L/min or oxygen which was 83.4± 4 percent of VO2 max. AT power ranged from 332-418 watts with a mean of 368.5 ± 21.3. AT power, but not AT, was correlated to 2K erg performance (Table 1). Neither was correlated with 2K water performance.
Peak 45 values ranged from 770-1134 Watts with a mean of 927.6 ± 95.3. There was a significant correlation between Peak 45 and 2K erg times but not with 2K row (table 1). Mean45 values ranged from 737-896 watts with a mean of 796.9 ± 74.2. There was no correlation between Mean45 and either 2K erg or 2K row.
### Discussion
Monitoring training and changes in physiological parameters is a challenge in rowing. Changes in weather and water conditions between tests can make it difficult to compare data and draw valid conclusions. Rowing ergometers were originally designed so that rowers in colder climates could continue to train in a fashion similar to their sport during periods when they could not be on the water. The ability to perform a rowing movement in a controlled environment made the rowing ergometer an attractive tool for monitoring changes in physiological variables. The findings of the present study are consistent with others (10,11) that have found significant correlations between rowing ergometer 2000m performance and VO2 max, AT and maximal power from a modified Wingate test performed on the same type of ergometers. The relationship between VO2 max and 2K erg performance in the current study is lower than that seen by Ingham et al. (4) and Cosgrove et al. (2), who found r = 0.88 and r = 0.85, respectively. Cosgrove (2) studied club level college-aged males of varying ability and while the Ingham study examined World Championship finalists both males and females as well as lightweight and heavyweight rowers were included in the analysis, creating a more heterogeneous group. The subjects in the current study were more homogenous in respect to their 2K erg times compared to other studies; 5:59-6:12 in the current study compared to ranges of 7:32.9- 8:07 in Riechman et al. (10), 6:20-7:26-in Russel et al. (11) and 6:30-7:45 in Cosgrove et al (2). Because a sport performance is multifaceted, with physiological, biomechanical, technical and psychological factors all playing roles in the final outcome, as the group performance becomes more homogenous it is less likely that any single physiological variable will be a strong distinguisher of performance.
One of the main purposes of determining the relationship between different physiological variables and rowing performance is to identify those variables that need to be trained to maximize performance (10). The current study demonstrates that although there may be a relationship between some physiological variables and rowing ergometer performance there is no relationship between physiological variables measured on a rowing ergometer and on-water performance in a group of elite heavyweight male rowers. This is the first study to directly compare the relationship between physiological variables determined on a rowing ergometer to on-water performance. Juimae et al. (5) examined the relationships among anthropometric variables, ergometer, and on-water performance, finding that only muscle mass correlated to on-water single scull performance while almost all anthropometric variables were related to ergometer performance. The lack of relationship between physiological or anthropometric predictors of ergometer performance and on-water performance is not surprising given that the relationship between on-water performance scores and ergometer performance scores can vary greatly across boat classes and levels of competition.
In two separate studies that examined the relationship between World Championship ranking or Junior World Championship ranking and 2000m ergometer performance Mikulic et al. (8) and Mikulic et al. (9) found significant correlations in 10 of 13 World Junior events and 17 of 23 World Championship events, but the standard errors were too large to establish accurate ranking predictions for any of the events. The highest correlations (r=0.92) were seen in the junior women’s single scull event, followed by the junior men’s single scull (r=0.80) and the junior women’s double scull (r=0.79). In contrast to the r = 0.12 of the current study, the senior men’s single scull had an r = 0.72. Some of the difference in results of the Mikulic studies (8,9) compared to the current study may be due to the nature of the variables correlated. In both Mikulic studies (8,9) the correlations were with final World or Junior World Championship rankings, whereas the present study looked at the relationship between actual times rowing on the water versus rowing the ergometer. In the Mikulic studies (8,9), the highest correlations were seen in sculling boats, particularly the singles. Athletes competing at a World Championship in sculling boats are normally specialists in that discipline. In the current study although all athletes were familiar with sculling and spent some of their time training in single sculls, 14 of the 19 were not sculling specialists. The lowest correlations in Mikulic’s work (8,9) were also seen in sweep rowers competing in larger boats r = 0.47 for the heavyweight men’s eight and r = 0.21 for the lightweight men’s eight. Ergometer rowing is technically more similar to sculling than to sweep rowing, the technical differences between the sculling specialists and non-specialists may explain the difference in correlations seen in the current study. This clearly supports the notion that there are differences between rowing ergometer performance and on-water rowing performance, particularly for sweep rowing athletes, and that physiological variables determined on the rowing ergometer may not be good predictors of performance on the water.
### Applications In Sport
This study reinforces what many coaches already know; there is more to a rowing performance than physiological test results or rowing ergometer performance scores. Ergometer rowing requires less skill than on-water rowing (10). Rowing technique on the water is a complex skill that requires balance, efficiency, and maintaining the boat speed during the recovery phase. These factors cannot be measured on an ergometer. This makes the rowing ergometer a good tool for studying and tracking physiological changes that occur during a rowing movement and can help coaches identify those athletes who have a large discrepancy between ergometer and on-water performances that may be technique related. However, caution needs to be exercised when trying to extrapolate rowing ergometer performance and physiological scores to on-water performance.
### Tables
#### Table 1
| Physiological Variable | 2K erg | 2K row |
|---|---|---|
| Peak 45 | -0.426* | -0.229 |
| Mean 45 | -0.321 | -0.200 |
| AT power | -0.470* | -0.267 |
| AT | -0.320 | 0.043 |
| VO2 max | -0.555* | 0.049 |
* p< 0.05
### References
1. Caiozzo, V.J., Davis, J.A., Ellis, J.F., Azus, J.L., Vandagriff, R., Prietto, C.A., & McMaster, W.C. (1982). A comparison of gas exchange indices used to detect the anaerobic threshold. J Appl Physiol. 53, 1184–1189.
2. Cosgrove, M., Wilson, J., Watt, D., & Grant, S. (1999). The relationship between selected physiological variables of rowers and rowing performance as determined by a 2000m ergometer test. J. Sports Sci. 17, 845-852.
3. Hahn, A.G., Tumilty, D.M., Shakespear, P., Rowe, P., & Telford, R.D. (1988). Physiological testing of oarswomen on Gjessing and Concept II rowing ergometers. Excel. 5, 19-22.
4. Ingham, SA., Whyte, GP., Jones, K., & Nevill, A.M. (2002). Determinants of 2000m rowing ergometer performance in elite rowers. Eur. J. Appl. Physiol. 88, 243-246.
5. Jurimae, J., Maestu J. Jurimae T. & Pihl, E. (2000) Prediction of rowing performance on single sculls from metabolic and anthropometric variables, J. Hum. Mov. Stud. 38, 123-36.
6. Lormes, W., Buckwitz, R., Rehbein, H., & Steinacker, J.M.(1993). Performance and blood lactate on Gjessing and Concept II rowing ergometers. Int J Sports Med. 14(Suppl 1), S29-S31.
7. Mahony, N., Donne, B., & O’Brien, M. (1999). A comparison of physiological responses to rowing on friction-loaded and air-braked ergometers. J Sports Sci. 17, 143-149.
8. Mikulic, P., Smoljanovic, T., Bojanic, I., Hannafin, JA., & Pedisic, Z, (2009). Does 2000m rowing ergometer performance correlate with final rankings at the World Junior Rowing Championship? A case study of 398 elite junior rowers. Journal of Sports Sciences. 27, 361-366.
9. Mikulic, P., Smoljanovic, T., Bojanic, I., Hannafin, JA., & Matkovic, B, (2009). Relationship between 2000m rowing ergometer performance times and World Rowing Championships rankings in elite standard rowers. Journal of Sports Sciences. 27, 907-913.
10. Reichman, S., Zoeller, R., Balasekaran, G., Goss, F., & Robertson, R. (2002). Prediction of 2000m indoor rowing performance using a 30s sprint and maximal oxygen uptake. J. Sport Sci. 20, 681-687.
11. Russell, A., Le Rossignol, P., & Sparrow, W. (1998). Prediction of elite schoolboy 2000m rowing ergometer performance from metabolic, anthropometric and strength variables. J. Sport Sci. 16, 749-754.
12. Soper, C., & Hume, P.A.(2004). Reliability of power output during rowing changes with ergometer type and race distance. Sports Biomech. 3, 237-224.
13. Tachinaba, K., Yashiro, K., Miyazaki, J., Ikegami, Y., & Higuchi, M. (2007). Muscle cross sectional area and performance power of limbs and trunk in the rowing motion. Sports Biomechanics. 6, 44-58.
14. Urhausen, A., Weiler, B., & Kinderman, W. (1993). Heart rate, blood lactate, and catecholamines during ergometer and on-water rowing. Int. J. Sports Med. 14, Suppl 1, 20-23.
15. Vogler, A., Rice, A., & Withers, R. (2007). Physiological responses to exercise on different models of the Concept II rowing ergometer. Int. J. Sports Physiol and Perf., 2, 360-370.
16. Womack CJ. Davis SE. Wood CM. Sauer, K., Alvarez, J., Weltman, A., & Gaesser, G. (1996). Effects of training on physiological correlates of rowing ergometry performance. J. Strength Cond. Res. 10, 234-238.
### Coresponding Author
Ed McNeely
560 Proudfoot Lane #1012
London, Ontario
N6H 5C9
Canada
613-371-8913
<e.mcneely@rogers.com>
### Author Bio
Ed McNeely is the senior physiologist at the Peak Centre for Human Performance and a partner in StrengthPro Inc. a Las Vegas based sport and fitness consulting company he is also a National Faculty member of the United States Sports Academy
