Team-handball is a physically demanding sport that is characterized by fast-paced end-to-end actions in which two teams pass a ball with the aim of scoring goals. As with other team sports, such as football (Castellano and Casamichana, 2010), volleyball (Jimenez-Olmedo et al., 2016), tennis (Sanchez and Gomez-Marmol, 2013) or rugby (Martins and Gas, 2014), handball has also been adapted to the sandy surface of the beach. Beach handball has emerged as a new worldwide, popular discipline with the support of the International, European and National Handball Federations.
Beach handball was created with specific rules to be more spectacular than indoor handball. Among them, we can find a double score for goals scored in flight, 360 degrees throwing or the use of the goalkeeper to attack. Beach handball is played by four players, one of them being the goalkeeper, on a sandy rectangular field of 27meter length and 12-meter width. The match consists of two 10-minute sets with a shoot out to get a winner team in case of a tied match. Therefore, beach handball is characterized by the combination of high-intensity efforts, such as sudden accelerations with short recoveries, heterogeneously distributed across a match. As a result, players are exposed to both high- and low-level work rates which demand appropriate speed, sprint ability, strength and power (Lara-Cobos, 2011).
In order to improve performance, the demands of the sport must be known to design specific training. Therefore, physical and physiological analysis provides scientists and coaches with accurate information to set up routines and goals. Similar to analysis in other disciplines, such as beach soccer (Castellano and Casamichana, 2010), the assessment of running demands as an indicator of physiological load, together with Heart Rate (HR) measurements, are crucial to determine if there are differences between competition levels or genders. Kinematic variables such as distance, velocity, time sprinting and accelerations required by professional beach handball players during competitions at specific HR intensities would help coaches in optimizing the training process, since tailored training programs can be designed to match these demands (Campos-Vazquez et al., 2015).
A number of studies have been carried out on analyzing the physical profile of classic handball (Chaouachi et al., 2009; Gorostiaga et al., 2006; Manchado et al., 2013; Povoas et al., 2012) and other sport variations, such as small-sided handball (Corvino et al., 2014; 2016; Dello-Iacono et al., 2016). Beach handball has been studied through notational analysis (Gruic et al., 2011; Morillo-Baro and Hernandez-Mendo, 2015; Morillo-Baro et al., 2015) and psychology (Morillo-Baro et al., 2016). However, to the knowledge of the authors, there are currently no studies that analyze physical demands during a competition match in male and female players at national and international levels. Consequently, the aims of the present study were to investigate the physical demands by means of time-motion analysis of beach handball players with GPS technology, and the subsequent physiological response by means of HR responses.
Twelve male (mean age: 26.3 [+ or -] 4.8 yrs, range 20-33 yrs; body weight: 84.5 [+ or -] 12.1 kg; body height: 1.87 [+ or -] 0.09 m; training experience: 9.5 yrs, training: 24.5 h/week) and female (mean age: 23.7 [+ or -] 4.8 yrs, range 17-33 yrs; body weight: 62.4 [+ or -] 4.6 kg; body height: 1.68 [+ or -] 0.05 m; training experience: 7.1 yrs, training: 23.5 h/week) beach handball players of the Spanish National Team were recruited for this study. The female Spanish National Team became Champion of the 2016 Beach Handball World Championship held in Budapest, Hungary. Similarly, the male Spanish National Team ranked fifth in the same competition.
All the players were previously informed about the research aims, experimental protocol and procedures of the study and voluntary gave their informed written consent to participate. The Ethics Committee at the University of Alicante gave institutional approval to this study, in accordance with the Declaration of Helsinki.
The study was conducted during the pre-game meeting prior to the 2nd Annual Spanish Beach Handball Cup held in June 2016. Male and female teams played two matches of two 10-min periods within each team. Since recordings were captured each half, a total of 48 records were attained for each group. To ensure that time-motion analysis was considered only when each player was involved in the game, goalkeepers and resting or bench periods were excluded.
The player's running profile during the competitive matches was assessed using a portable GPS device (SPI Pro X; GPSports Systems, Canberra, Australia) worn in a special harness on each player's back (mass: 76 g, dimensions 48 x 20 x 87 mm). Speed and distance were recorded at a sampling frequency of 15 Hz, whereas acceleration was recorded at 100 Hz by means of a built-in triaxial accelerometer. In order to capture HR data, an HR transmitter belt was also worn by each player (Polar Electro, Kempele, Finland). Once raw data had been gathered, the software provided by the manufacturer was used to download and analyze data in detail (Team AMS R1 2015, GPSports, Australia). Studies on validity and reliability of GPS devices reported that GPS analysis is a valid and reliable method for time-motion tracking in athletes during field sports (Coutts and Duffield, 2010; Jennings et al., 2010; Koklu et al., 2015).
The running profile was evaluated according to their frequency and duration using the distance attained in six speed zones: zone 1: standing (0-0.4 km/h), zone 2: walking (0.5-4 km/h), zone 3: jogging (4.1-7 km/h), zone 4: cruising (7.1-13 km/h), zone 5: high-intensity running (13.1-18 km/h) and zone 6: sprinting (18 km/h). Speed zone labels were considered as similar to those used in other studies conducted in intermittent team sports (Cummins et al., 2013). However, speed zone intervals were corrected accordingly since beach handball is played on sandy surfaces which prevent players from performing quick movements. Therefore, the speed zones proposed by Castellano and Casamichana (2010) for beach soccer were used, together with the standing zone to account for lack of movement. The maximum speed achieved by males and females were 21.2 km/h and 18.5 km/h, respectively, which are in accordance with sprinting zones.
By means of the integrated triaxial accelerometer in each GPS device, accelerations and decelerations on three reference planes are measured as changes in gravitational forces, knowing that gravity of 9.8 m/[s.sup.2] equals 1 g. Player exposure to collisions was graded according to six impact zones provided by system manufacturer and used in previous studies (Cunniffe et al., 2009; McLellan and Lovell, 2012): zone 1: light impact, hard acceleration /deceleration/change of direction (5-6 g), zone 2: light-moderate impact, minor player collision, contact with the ground (6-6.5 g), zone 3: moderate-heavy impact, tackle (6.5-7 g), zone 4: heavy impact, tackle (7-8 g), zone 5: very heavy impact, high collision, tackle (8-10 g) and zone 6: severe impact/tackle/collision (10 g). Body load during exercise is a measure of high impact activities, computed by the system software as a weighted sum of intense accelerations, decelerations, changes in direction, rapid changes in posture and collisions. Acceleration was also analyzed in zones: zone 1: low (1-2 m/[s.sup.2]), zone 2: moderate (2-3 m/[s.sup.2]) and zone 3: high (3 m/[s.sup.2]), according to previous studies (Akenhead et al., 2013; Osgnach et al., 2010).
Finally, physiological demands were analyzed by means of HR data. In order to estimate the maximum HR (HRmax) of each player, the formula proposed by Tanaka et al. (2001) was used. However, for players with higher HRmax value during matches than the estimated one, the reference for calculating heart rate zones was the actual maximum value. Heart rate zones were determined as a function of each player's HRmax as follows: zone 1: 95%.
All results were analyzed using Statistical Package for Social Sciences (SPSS v.22 for Windows, SPSS Inc, Chicago, USA). Data is presented as mean and standard deviation and the alpha level of significance was set at p 1.2 (large) (Batterham and Hopkins, 2006). Due to the small sample in this study (n = 24), the bias-corrected effect size was used (Hedges and Olkin, 1985). Also, a two-way ANOVA (factor sex and factor half time) was performed to compare differences in half times, between male and female players and interaction between these factors, with significance at p
Table 1 presents the results of the time-motion analysis for male and female teams. For the match analysis, the zones where male players have traveled the most are walking (0.5-4 km/h) and jogging (4.1-7 km/h). However, the comparison between halves showed significant differences in mean distance in the walking category only (20.5% higher in the first half).
For female players, the walking (0.5-4 km/h) and jogging (4.1-7 km/h) zones were the most travelled, following the same trend as male team. The following variables showed significant differences between halves (corrected Cohen's ES between brackets): total distance (moderate), average speed (large), distances covered in the standing (moderate) and walking (large) speed categories and average time spent in sprint (small). Female players covered 21.7% more distance in the first half than in the second half but with lower average speed (78.6%).
Table 2 shows accelerometer data provided in speed changes and collisions through g force zones. Players' exposure to impacts is similar for male and female players. Both groups received a large number of very light impacts, consisting of accelerations or decelerations and changes of direction while running...