ANALYSIS OF THE BEGINNING OF THE EARLY FLIGHT PHASE OF THE SKI JUMP IN ATHLETES WITH DIFFERENT PERFORMANCE LEVELS

Nordic combined (NC) is a winter sport divided into two portions: ski jumping (SJ) and cross country ski racing, which diff er in their optimum biomechanics, physiology, and body parameters (Burke, 2007). There are fewer studies on NC than on SJ, and the scientifi c literature related to kinematic analyses of basic jump phases in NC is limited (Janura, Lehnert, Elfmark, & Vaverka, 1999). While some studies on NC have been published in the biomechanical literature (Jošt & Pustovrh, 1995), others have been published as part of the literature on muscle physiology and lower limb strength (Bösl, Schwirtz, Rott, & Großgebauer, 2007; Ereline, 2006; Pääsuke, Ereline, & Gapeyeva, 2001). The body movement of the competitor at the jumping hill edge (take off and early fl ight) is considered the most important factor for good jump execution (Schwameder, 2008). At the end of the take off , the ski jumper engages in a forward rotating body angular momentum, which serves to retain the necessary velocity and the best initial conditions for execution of the early fl ight position (Schwameder, Müller, Lindenhofer, DeMonte, Potthast, Brüggemann, Virmavirta, Isolehto, & Komi, 2005). The kinematics of the execution of the ski jump pha ses have been analyzed by several investigators. For example, Janura, Elfmark, Seidl, and Vaverka (2001) found that in NC the best juniors compared to the worst juniors had the center of mass (CoM) more forward behind the edge of the hill and took the flight po sition earlier, with more eff ective aerodynamic conditions (smaller drag force). Vaverka, Janura, Elfmark, McPherson, and Puumala (1996) determined that there is a small relationship between the take off kinematic parameters and the length of a jump, which contribu tes to the large diff erences among the competitors. The investigators found, however, a larger correlation between the measured angular parameters of the trunk and lower extremities and the length of a jump during a fl ight (r = .81). Brüggemann, DeMonte, Komi, Isolehto, Müller, Pott hast, Schwameder, and Virmavirta ANALYSIS OF THE BEGINNING OF THE EARLY FLIGHT PHASE OF THE SKI JUMP IN ATHLETES WITH DIFFERENT PERFORMANCE LEVELS


INTRODUCTION
Nordic combined (NC) is a winter sport divided into two portions: ski jumping (SJ) and cross country ski racing, which diff er in their optimum biomechanics, physiology, and body parameters (Burke, 2007).There are fewer studies on NC than on SJ, and the scientifi c literature related to kinematic analyses of basic jump phases in NC is limited (Janura, Lehnert, Elfmark, & Vaverka, 1999).While some studies on NC have been published in the biomechanical literature (Jošt & Pustovrh, 1995), others have been published as part of the literature on muscle physiology and lower limb strength (Bösl, Schwirtz, Rott, & Großgebauer, 2007;Ereline, 2006;Pääsuke, Ereline, & Gapeyeva, 2001).
The body movement of the competitor at the jumping hill edge (take off and early fl ight) is considered the most important factor for good jump execution (Schwameder, 2008).At the end of the take off , the ski jumper engages in a forward rotating body angular momentum, which serves to retain the necessary velocity and the best initial conditions for execution of the early fl ight position (Schwameder, Müller, Lindenhofer, DeMonte, Potthast, Brüggemann, Virmavirta, Isolehto, & Komi, 2005).
The kinematics of the execution of the ski jump pha ses have been analyzed by several investigators.For example, Janura, Elfmark, Seidl, and Vaverka (2001) found that in NC the best juniors compared to the worst juniors had the center of mass (CoM) more forward behind the edge of the hill and took the flight po sition earlier, with more eff ective aerodynamic conditions (smaller drag force).Vaverka, Janura, Elfmark, McPherson, and Puumala (1996) determined that there is a small relationship between the take off kinematic parameters and the length of a jump, which contribu tes to the large diff erences among the competitors.The investigators found, however, a larger correlation between the measured angular parameters of the trunk and lower extremities and the length of a jump during a fl ight (r = .81).Brüggemann, DeMonte, Komi, Isolehto, Müller, Pott hast, Schwameder, and Virmavirta (2002) found only small diff erences in competitors' CoM velocities between the best and poor competitors, but did fi nd statistically signifi cant diff erences between the above-noted groups in knee angular velocity and so mersault angle at take off .Jošt, Čoh, Pustovrh, and Ulaga (2000) concluded that the best SJs had a large knee angle extension on and behind the hill's edge as well as a tendency for a more pronounced transfer of the hip and the CoM in the anterior direction.
The high requirements of the body at take off (maximum take off force during a short time period) and the complexity of the movement structure result in a high variability of the execution, not only between groups, but also within groups within given performance levels (Janura, Svoboda, & Uhlář, 2007;Vaverka, Janura, Elfmark, Salinger, & McPherson, 1997).Thus, it is necessary to consider these diff erences in the comparison of the ski jump execution of the competitors at diff erent performance levels (Arndt, Brügemann, Virmavirta, & Komi, 1995;Schmölzer & Müller, 2005).
The purpose of this study was to assess the execution of the start of the early fl ight phase in the NC among three groups of competitors, representing diff erent skill levels, and to compare them with a group of ski jumpers.
Video image data were obtained from two digital video cameras (50 Hz, Sony DCR-TRV 900, Sony, Tokyo, Japan), which were located perpendicular to the sagittal plane of the SJs' movement at early fl ight (Fig. 1).Because the conditions on the jumping hill did not facilitate the taking of a three dimensional (3D) measurement, we opted for a 2D kinematic analysis in the sagittal plane.The image space was calibrated by a 1 m arm cross-calibration frame placed in the plane of the movement at the beginning and at the end of the observed section.The length of a recorded sector was 7 m, and the frame had a resolution of 640 × 480 pixels, i.e., a shift of the cursor by 1 pixel resulting in a linear resolution of 0.011 m.The accuracy of the body angular values had been quantifi ed in a previous study, in which the magnitude of error for the recorded sector of appro xi mately 1.4 m was .51%,and the absolute error was .22%,respectively (Janura & Vaverka, 1997).

Fig. 1
The cameras' setup The data were manually digitized (APAS, Ariel Dynamics, Inc., Trabuco Canyon, USA) by an experienced researcher.The origin of the global coordinate system was placed in the middle of the in-run tracks at the jump ing hill edge.We assumed a symmetrical jump skier's body position in the sagittal plane.A seven link bilateral model was created based on nine points -the tip and tail of the ski, ankle, knee, hip, top of the head, shoulder, elbow, and wrist.The model included the following segments: foot and shank, thigh, trunk and head, arm, forearm, and jump ski.Data on the segments' relative mass (Karas, Otáhal, & Sušanka, 1990) were added to the relative mass of the jump skis and ski jumping equipment for the calculation of the CoM.
The average in-run velocities and lengths of jumps were taken from the competitors' offi cial results (18-8 m before the hill's edge).The basic angle parameters of the segments with respect to the horizontal were calculated from the coordinates of the selected points on the SJs' body and the skis in the area 1 m in front of and 6 m behind the jumping hill edge (Fig. 2 shows 0 and 5 m behind the jumping hill edge).The mean joint angular velocity was determined as a ratio of the mean joint angle change and time between the beginning (jumping hill edge) and end (5 m behind the jumping hill edge) of the observed section.The CoM angle was calculated as an angle of the line passing the ankle joint and CoM relative to horizontal.
Two way ANOVA (independent variables are the discipline and performance level) with a Fisher's post-hoc test were performed (Statistica, Version 8.0, Stat-Soft, Inc., Tulsa, USA).The desired power of the test was .80 (Ellis, 2010) for the diff erences in the variables noted in the study.P-values of less than .05were deemed to be signifi cant throughout.

RESULTS
The results of the dependent parameters for the NC and SJ groups in the section from the edge of the jumping hill to 5 m behind the edge are presented in TABLES 1 and 2.

Comparison of the NC competitors with diff erent performance levels
With the exception of the average in run velocity, we did not fi nd any signifi cant diff erences among the diff erent level groups.Specifi cally, group E had a higher average in run velocity than did group P (p < .01)or group M (p < .05).

TABLE 2
Observed parameters of the SJ competitors on the edge and 5 m behind the jumping hill edge Legend: LJ -length of jump, v -in-run velocity, M TR -trunk angle to horizontal, M TH -thigh angle to the horizontal, M SH -shank angle to horizontal, M SK -ski angle to horizontal, M CoM -angle between CoM and ankle connection and horizontal, Z TR -trunk angular velocity, Z TH -thigh angular velocity, Z SH -shank angular velocity, Z SK -ski angular velocity, Z CoM -CoM angular velocity, 0 -jumping hill edge, 5 -m behind the jumping hill edge, E -elite, M -mediocre, P -poor

Comparison of the NC and SJ competitors
The average in-run velocity was higher in NC group (p < .01).The SJ group experienced a smaller shank angle to the horizontal (p < .05) at the edge of the hill and a higher knee extension angular velocity (p < .05) 5 m behind the edge of the hill when compared to the NC group.A comparison of the observed NC and SJ groups is presented in Fig. 3.

Fig. 3
Graphic comparisons of the analyzed groups of NC and SJ competitors at selected points of the analysed area Group E -ski jumpers had a smaller shank angle to the horizontal (p < .05),smaller CoM angle (p < .01),and larger magnitude of the knee extension angular velocity characterized by the shift of the thigh anteriorly (p < .05) in the section 0-5 m behind the edge.
Group M -the SJ group exhibited a faster knee extension angular velocity (p < .01) in the section between 0 and 5 m behind the hill's edge.
Group P -the signifi cant diff erences were not found between NC and SJ groups.

Comparison of the NC competitors of diff erent performance levels
With the exception of the magnitude of the in-run velocity, which was larger among the E competitors, we did not fi nd any signifi cant diff erences in angular and linear parameters.This lack of a signifi cant diff erence might have been caused by the elite level athletes' correct in-run body position, which is one of the factors that maximize the in-run velocity (Komi, Nelson, & Pulli, 1974;Vaverka, 1987).The signifi cant infl uence of the in-run velocity on the length of jump was found in the fi nal round of the 2006 Olympic ski jumping competition in Torino, Italy (Virmavirta, Isolehto, Komi, Schwameder, Pigozzi, & Massazza, 2009).Dissimilarities in the execution of the jump at the early fl ight phase were found among the jumpers with similar lengths of jumps; therefore, it is necessary to study individual differences within these groups.The correlations between the measured body parameters and the length of each jump suggest the optimal solution for each jumper (Virmavirta, Isolehto, Komi, Brüggemann, Müller, & Schwameder, 2005).

Comparison of the NC competitors and the SJ group
The SJ group had a smaller ankle angle to the horizontal at the jumping hill edge, which might have given the ski jumpers a better body position, minimizing air resistance, for the following rotation and fl ight body position.This group also exhibited higher explosiveness in the knee joint, supported by the higher knee extension angular velocity 5 m behind the edge of the hill, which is comparable to the fi ndings of Arndt et al. (1995), who reported similar results as measured at the jumping hill edge.Janura et al. (1999), in contrast, conducted a study with the Czech Republicʼs teams to compare competitors in diff erent age categories and found that the SJ group had a large ankle angle to the horizontal along with a small knee extension of the angular velocity at the jumping hill edge.Their fi ndings, however, might have been infl uenced by diff erences in performance between groups at varying levels.
In run velocity is the key parameter for comparing both ski jumping disciplines.The NC competitors, with high average in run velocities, showed body movement patterns similar to those observed in SJ groups.This higher in-run velocity eliminates diff erences between both disciplines in training methods and in somatic parameters of competitors.
The fact that the NC competitors had a higher in-run velocity provided them with the potential to jump as far as the SJ group could.Smaller ankle angle to the horizontal by ski jumpers moved the shank more anteriorly and provided the opportunity for a better aerodynamic ski jumping body position.The higher knee extension of the angular velocity executed by the SJ group could have been caused by the fact that ski jumpers concentrate on muscular strength and explosive power training more than Nordic Combined skiiers, who focus on endurance training, which decreases muscle explosiveness (Bösl et al., 2007;Shephard & Astrand, 1992;Zatsiorsky, 1995).Therefore, it is necessary to utilize a set of internal preconditions for competitors in ski jumping analyses (e.g.anthropometric segment parameters and morphological entries).

Limitations
The ski jump events in NC and SJ were held on diff erent days, but the weather conditions were comparable, i.e., the air temperature (between 1 °C and 2 °C), snow temperature (-8 °C in both events), wind velocity (0.2-2.7 m.s -1 vs. 0.3-2.5 m.s -1 ), and wind direction.A possible negative infl uence of the external conditions was also eliminated by the fact that the competitors, who were, over the long term, characterized by good per for mance, were excluded from the measurement when they jumped badly.
In run velocity was signifi cantly infl uenced by a different starting gate in the NC (gate 37) and SJ (gate 24) events.

CONCLUSIONS
• Among the NC groups no signifi cant diff erences were found, with the exception of the average in run velocity.The in run velocity of the elite competitors was signifi cantly higher than in both other groups.
• Ski jumpers took a better aerodynamic position in Section 0-5 m behind the edge of the jumping hill, characterized by a large shift of the shank and center of mass anteriorly.
• In order to compensate for their muscular power and anthro pometric disadvantages, Nordic combi n ed com petitors have to use higher in-run velocity to achieve ski jump length comparable to those of ski jumpers.

TABLE 1
Observed parameters of the NC competitors on the edge and 5 m behind the jumping hill edge Legend: LJ -length of jump, v -in-run velocity, M TR -trunk angle to horizontal, M TH -thigh angle to the horizontal, M SH -shank angle to horizontal, M SK -ski angle to horizontal, M CoM -angle between CoM and ankle connection and horizontal, Z TR -trunk angular velocity, Z TH -thigh angular velocity, Z SH -shank angular velocity, Z SK -ski angular velocity, Z CoM -CoM angular velocity, 0 -jumping hill edge, 5 -m behind the jumping hill edge, E -elite, M -mediocre, P -poor . Science and practice of strength training.Champaign, IL: Human Kinetics.