A field study of exposure to whole-body vibration due to agricultural machines in a full-time rice farmer over one year

Abstract

Objectives: The aims of this study were to clarify in detail the levels of whole-body vibration (WBV) exposure from a variety of agricultural machines in a rice farmer over one year, and to evaluate the daily level of exposure compared with European and Japanese threshold limits. Methods: The subject was a full-time, male rice farmer. We measured vibration accelerations on the seat pan and at the seat base of four tractors with various implements attached, one rice-planting machine, two combine harvesters, produced by the same manufacturer, and one truck used for transportation of agricultural machines. The position and velocity of the machines were recorded in parallel with WBV measurements. In addition, during the year starting in April 2010, the subject completed a questionnaire regarding his work (date, place, content, hours worked, machines used). We calculated the daily exposure to WBV, A(8), on all the days on which the subject used the agricultural machines. Results: The WBV magnitude in farm fields was relatively high during tasks with high velocity and heavy mechanical load on the machine, and had no dominant axis. The subject worked for 159 days using the agricultural machines during the year, and the proportion of days on which A(8) values exceeded the thresholds was 90% for the Japan occupational exposure limit and 24% for the EU exposure action value. Conclusions: Our findings emphasize the need for rice farmers to have health management strategies suited to the farming seasons and measures to reduce WBV exposure during each farm task.

(J Occup Health 2015; 57: 378–387)

Introduction

Although mechanization has recently advanced in agriculture, many tasks continue to depend on manual labor, and agricultural work remains physically demanding, often having to be performed in awkward postures¹⁾. Furthermore, agricultural machines and tools are not ergonomically designed^2–5). The poor design of agricultural machines with driving seats (AMDSs) results in workers being exposed to high levels of whole-body vibration (WBV)^{1, 6)}. There is consequently excess load on the musculoskeletal system of farmers, particularly in the lumbar region. One review paper focusing on farmers with musculoskeletal symptoms reported that the prevalence of low back pain is higher in these subjects than in nonagricultural workers⁷⁾.

The European Directive on Mechanical Vibration set down requirements for WBV in the form of a daily exposure action value⁸⁾ (EUEAV) in 2006. Once the EUEAV is exceeded, the employer must implement an action plan to prevent further exposure exceeding the value. The EUEAV is set at either an A(8) [8-hour energy-equivalent, frequency-weighted, root-mean-square (RMS) vibration acceleration] of 0.5 m/s² or a VDV [fourth power vibration dose value⁹⁾] of 9.1 m/s^1.75. The Japan Society for Occupational Health also established recommendations of occupational exposure limit¹⁰⁾ (JOEL) for WBV in 2013. The JOEL is set at an A_sum(8) (based on the vector sum of the frequency-weighted RMS vibration acceleration value in three orthogonal directions) of 0.35 m/s². These values are determined assuming that fluctuation in daily exposure to WBV is small. In addition, the frequency-weighted RMS vibration acceleration value, based on the maximum value in three orthogonal directions, is applied to determine the EUEAV according to the EU Directive⁸⁾.

Rice-farming families in Japan typically possess more than one AMDS, including tractors, rice-planting machines, and combine harvesters, and they work using a tractor with a variety of implements for each farming task. When farmers use agricultural machines, they drive either on a farm road or in fields, which have a variety of surfaces, such as hard, dry, muddy, undulating, flat, and smooth surfaces. Different road surfaces have different effects on WBV¹¹⁾. These effects also vary according to the different manufacturers and models of AMDS^{11, 12)}. In addition, the duration of exposure to WBV (T_Exp) fluctuates based on different seasons and different tasks performed. Therefore, in a full-time rice farmer, there are large fluctuations in the daily level of exposure to WBV because of multiple factors, including season, type of task, machine used, implement used, conditions of the driving surface, and hours worked, and the fluctuations are much greater than for workers involved in manufacturing or transportation.

Regarding the direction of vibration, previous studies have shown that the WBV magnitudes for two or more axes were comparable for machines operated on a farm^11–13). The use of only vertical vibration to estimate exposure to WBV for buses¹⁴⁾ and taxis¹⁵⁾, having the highest value in many cases, was appropriate, whereas both vertical and horizontal vibrations have to be considered for AMDSs.

The intensity and dose of exposure to WBV induced by AMDSs over one year, in association with seasonal changes, need to be investigated. There is no previous research on this subject. Therefore, the aims of this study were to clarify in detail the levels of exposure to WBV from AMDSs in a farmer throughout one working year and to evaluate the daily level of exposure compared with the European and Japanese threshold limits.

Subject and Methods

Subject

The subject recruited was a 39-year-old male, full-time rice, barley, and soybean farmer who had worked on his family farm, in the plain area of Japan, for 17 years. The work process of crop rotation for rice, barley, and soybean is shown in Fig. 1. In 2010, the year of the study, the family farm had a cultivation area of 16 hectares (ha), with 11 ha dedicated to rice alone and 5 ha dedicated to crop rotation.

Fig. 1.

Typical work process for rice, barley, and soybean cultivation over two years.

This study protocol was approved (approval number: 24–27) by the Human Research Ethics Committee of Shiga University of Medical Science, Japan. After the details of this study were explained, the subject provided informed consent in writing.

Machines

We conducted measurements on four tractors: T1 with an engine power of 46 ps and a caterpillar tread instead of a rear wheel, T2 with an engine power of 50 ps, T3 with an engine power of 33 ps, and T4 with an engine power of 33 ps and no cabin; each tractor was equipped with a variety of implements. We also conducted measurements on the following: one rice 8-row planting machine; two combine harvesters, a 5-row reaping-type harvester (H1) and a 4-row reaping-type harvester (H2); and a truck for transportation of agricultural machines. All these machines, except for the truck, were made by the same manufacturer (ISEKI & Co., Ltd., Matsuyama, Japan). Details and photographs of all the machines are shown in Table 1.

Table 1. Specifications and photos of the machines

Name	Type of machine	Power (ps)	Year manufactured	Specifications	Tire diameter (cm)
					Front	Rear
T1	Tractor “GEAS 46s”	46	2004	Caterpillar treads instead of rear wheels	77	—
T2	Tractor “GEAS 503”	50	2000		71	116
T3	Tractor “GEAS ATK 33”	33	2009		74	123
T4	Tractor “TK 33”	33	1998	No cabin	75	116
PM	Rice planting machine “HST 20”	20		8-row planting
H1	Combine harvester “Japan 590”		2005	5-row reaping	Caterpillar tread
H2	Combine harvester “Frontier Fighter 446”		2002	4-row reaping	Caterpillar tread
TK	Truck for transportation an agricultural machine		2000	Made by Isuzu Motors Ltd.

The above machines, except TK, were made by ISEKI & Co., Ltd.

Measurement of vibration and other indicators

Vibration acceleration at the interface of the seat surface and the operator was detected using a triaxial seat accelerometer (Model 356B41, PCB PIEZOTRONICS Inc., Depew, NY, USA) mounted on the seat pan. Seat-base vibration acceleration was detected using a triaxial accelerometer (Model 356A16, PCB PIEZOTRONICS Inc.) placed rigidly at the seat base. Signals from the accelerometers were acquired using two 3-channel vibration meters (VM-54, RION Co., Ltd., Tokyo, Japan), each with a built-in VX-54WB WBV analysis program card (RION Co., Ltd.). The frequency-weighted RMS vibration acceleration values in the fore and aft direction (x-axis), lateral direction (y-axis), and vertical direction (z-axis), on the seat pan [a_wi(t): i =x, y, z] and base were calculated and recorded every second in flash memory cards mounted in the vibration meters. W_k and W_d, as defined in ISO 2631-1:1997⁹⁾, were adopted for the frequency weightings for vertical and horizontal vibration, respectively.

The position and velocity of each machine were recorded every second using a GPSMAP 60CSx portable GPS navigation device (Garmin, Olathe, KS, USA). The approximately constant velocity of an agricultural machine in a farm field was calculated by averaging the velocity after extracting the running time without including the time needed to enter or exit the field or the time needed for U-turns around the edge of the field. Sitting and leaving time points were recorded using a sitting switch, which we had previously developed and validated in forklift operators¹⁶⁾, on the seat vibration sensor and a HOBO U9 State Data Logger (Onset Computer Corp., Boume, MA, Cape Cod, USA). Video was recorded to assess the movement of the machines and the timing of when the driver sat down or left.

WBV and other indicators during each farm task were measured in at least three fields for half of each day. These measurements were conducted from April 2010 to November 2011.

Statistical analysis

Using one-way repeated measures analysis of variance, multiple comparisons were made between measured WBV values in the three directions, employing values for horizontal direction multiplied by 1.4⁹⁾. Correlations between the measured WBV values of each direction and the velocity were tested with respect to each driving surface using Pearson's correlation analysis. The significance level was set at 0.05, and the statistical analyses were performed using IBM SPSS Statistics version 20.

Questionnaire survey

Using a questionnaire that we designed, the subject recorded variables on the days worked using AMDSs for one year, starting in April 2010. All were closed-ended questions, and the data collected included the date, weather, place, content of work, surface condition of the farm field, hours worked in or around the fields (T_WinF), machines and implements used, and method and time of transportation of the agricultural machinery between the garage and farm fields. The subject agreed to cooperate with our study and recorded the questionnaire data in conjunction with a work diary, in which he described the characteristics of his work and agricultural tasks performed without AMDSs.

Calculation of daily WBV exposure

For a year, every day that the subject worked with AMDSs, we extrapolated the A(8) (A_sum(8) or A_max(8)) value for each farm task and for the day using WBV values obtained from our WBV measurements and the working time collected in the questionnaire survey, according to the following procedures.

When WBV was measured, the recording time for each task, in or around the farm fields, was divided into “sitting” (=exposed to WBV) and “leaving” (when the subject got out of the machine) categories, based on the data provided by the seat switch. The sitting time was further divided into four driving categories: operating in the field, driving on a paved road, driving on an unpaved road and others (entering or exiting the field, boarding the agriculture machine on the truck, etc.). The frequency-weighted RMS vibration acceleration values a_wij (j indicates the driving condition) in the three directions of the respective driving condition were calculated on the basis of a_wi(t). The mean velocity of the driving conditions was also determined. The vector sum a_sumj and maximum a_maxj of the vibration values in three directions were obtained with formula (1) and formula (2), respectively. The ratio of time RT_j for the five conditions, including “leaving,” to the recording time was also determined.

  

Using the a_j (a_sumj or a_maxj), RT_j, and the T_WinF of the farm task (=T_task) based on the questionnaire survey, the vibration energy E_task, A(8) value, and T_Exp for each farm task were calculated with formula (3), formula (4), and formula (5), respectively. We adopted the a_j and RT_j for the farm task which was identical in type with the task performed by the subject. In the absence of the same type of farm task among our WBV measurement, we adopted them for the farm task which was most similar in mechanical load of the machine.

  

The A(8) value on each day was calculated with formula (6) using the a_truck (a_sum,truck or a_max,truck) obtained while driving the transport truck and the transportation time T_truck (unit: hour) each day. The T_Exp for each day was the total of T_Exp for the task during the day with twice the T_truck added.

  

Results

WBV magnitude

We conducted 18 WBV measurements and obtained 16 complete measurements for farm tasks and transport. The measurement for T1, during daubing mud on the dikes, was obtained in 2011 when it was repeated, because the GPS data could not be recorded in 2010. The vibration acceleration of T3, during digging ditches at the rim of the farm field, was recorded at the seat base but not on the seat pan due to a problem with the vibration meter. Therefore, WBV of T3 for that task was calculated on the basis of the vibration acceleration at the seat base (a_wx,Base=0.21, a_wy,Base=0.14, a_wz,Base=1.59) and the ratio of vibration at the seat pan/base for the task of digging inside the ditch, as the same tractor was used (1.25 for the x-axis, 1.67 for the y-axis, 0.93 for the z-axis). The values, which are not shown in Table 2, were as follows: a_wx,Pan=0.27, a_wy,Pan=0.23, and a_wz,Pan=1.48.

Table 2. WBV magnitude and velocity

ID*. Farm task [Machine] Date	Surface#	Analysis time (h:min:s)	awx (m/s2)	awy (m/s2)	awz (m/s2)	Dominant†	asum (m/s2)	amax (m/s2)	Mean velocity‡ (km/h)
A. Daubing on dikes [T1] Apr. 15, 2010	Farm	0:18:27	0.53	0.39	0.31	No	0.98	0.75	—
	Pv	0:03:51	0.41	0.28	0.69	No	0.98	0.69	—
	UnPv	0:03:20	0.45	0.33	0.80	No	1.11	0.80	—
A. Daubing on dikes [T1] Apr. 21, 2011	Farm	1:19:47	0.27	0.35	0.23	No	0.66	0.49	0.6
	Pv	0:05:27	0.29	0.31	0.61	No	0.86	0.61	7.2
	UnPv	0:01:45	0.21	0.27	1.02	Z	1.13	1.02	3.8
B. Listing in a field [T3] Apr. 26, 2010	Farm	2:35:53	0.56	0.43	0.57	No	1.14	0.79	2.8
	Pv	0:08:36	0.31	0.37	0.52	No	0.85	0.52	10.1
	UnPv	0:05:15	0.45	0.61	0.81	No	1.33	0.85	8.9
C. Breaking ridges [T2] May 11, 2010	Farm	1:07:37	0.31	0.26	0.32	No	0.65	0.43	2.3
	Pv	0:18:21	0.36	0.41	0.94	Z	1.21	0.94	16.1
E. Transplanting [PM] May 13, 2010	Farm	0:32:39	0.46	0.47	0.29	No	0.97	0.65	3.1
	Pv	0:06:36	0.38	0.36	1.10	Z	1.32	1.10	9.1
F. Harvesting barley [H1] June 23, 2011	Farm	0:49:14	0.57	0.45	0.57	No	1.16	0.79	5.9
	Pv	0:01:34	0.54	0.35	0.72	No	1.15	0.75	10.0
G. Seeding soybeans [T4] July 11, 2011	Farm	0:58:48	0.56	0.50	0.74	No	1.29	0.79	2.2
	Pv	0:00:57	0.29	0.29	0.71	Z	0.92	0.71	9.1
I. Plowing [T2] Aug. 11, 2011	Farm	1:06:59	0.56	0.51	0.55	No	1.20	0.79	2.7
	Pv	0:01:08	0.33	0.41	0.58	No	0.94	0.58	8.1
J. Harvesting rice [H1] Sep. 15, 2010	Farm	0:37:07	0.51	0.48	0.59	No	1.15	0.71	4.8
	Pv	0:01:00	0.40	0.40	0.99	Z	1.27	0.99	5.8
J. Harvesting rice [H2] Sep. 26, 2011	Farm	1:03:36	0.50	0.37	0.61	No	1.06	0.69	3.7
	Pv	0:00:19	0.39	0.31	1.04	Z	1.26	1.04	4.6
K. Digging at field rim [T1] Oct. 25, 2011	Farm	2:07:14	0.30	0.30	0.29	No	0.65	0.41	0.6
	Pv	0:04:30	0.32	0.32	0.64	No	0.90	0.64	5.9
L. Making underdrains [T1] Oct. 26, 2010	Farm	1:00:06	0.60	0.38	0.60	No	1.16	0.83	3.6
	Pv	0:02:27	0.48	0.47	0.77	No	1.21	0.77	4.4
M. Seeding barley [T1] Nov. 8, 2010	Farm	1:13:28	0.35	0.32	0.44	No	0.79	0.49	2.8
	Pv	0:01:42	0.34	0.41	0.77	No	1.08	0.77	4.4
H. Digging inside a field [T3] Nov. 9, 2010	Farm	1:22:35	0.32	0.30	0.39	No	0.73	0.45	1.8
	Pv	0:07:07	0.25	0.29	0.75	Z	0.92	0.75	8.5
O. Fertilizer distribution [T2] Jan. 27, 2011	Farm	0:08:32	0.66	0.70	1.04	No	1.71	1.04	13.4
	Pv	0:07:40	0.42	0.40	0.90	Z	1.22	0.90	12.8
P. Plowing up soil [T1] Feb. 1, 2011	Farm	2:09:41	0.54	0.42	0.67	No	1.17	0.76	4.1
	Pv	0:07:35	0.37	0.30	0.58	No	0.88	0.58	10.6
Transportation [TK] Apr. 15, 2010	Pv	0:17:20	0.19	0.23	0.31	No	0.52	0.32

*  ID label for farm tasks in Figure 1.

^#  Pv, paved road; UnPv, unpaved road.

^†  A direction can be regarded as dominant when the a_wi (i=x, y, z) in each of the other directions (multiplied by 1.4 in the case of a_wi for the x and y directions) is less than 66% of that in the dominant direction.

^‡  The values for the farm fields are shown as the approximately constant velocities obtained by averaging after extracting the time needed to enter or exit the field and the time needed for U-turns around the edge of the field.

The results of the measurements during operation of the machines are shown in Table 2. An axis can be determined “dominant” when the a_wi in each of the other axes (in the case of a_wi, multiplied by 1.4 for the x- and y-axes) is less than 66% of that in the dominant axis¹⁷⁾. The arithmetic averages of a_wx, a_wy, a_wz, and a_sum on the farm fields were 0.47 m/s² (min 0.27–max 0.66 m/s²), 0.41 (0.26–0.70), 0.51 (0.23–1.04), and 1.03 (0.65–1.71), respectively, whereas those on paved roads, except for the transport truck, were 0.37 (0.25–0.54), 0.36 (0.28–0.47), 0.77 (0.52–1.10), and 1.06 (0.85–1.32), respectively. The arithmetic average of approximately constant velocities while operating AMDS in the farm field was 2.9 km/h (0.6–5.9 km/h), excluding the case of fertilizer distribution, the velocity of which was extraordinarily high (13.4 km/h).

1) Comparison of the magnitudes of the three axes

The a_wz values of seven AMDSs on paved roads were dominant, but there were no dominant axes for the WBV values of all AMDSs, obtained on the farm fields. Multiple comparisons of the WBV values showed significant differences for the three axes on the farm fields (p<0.01), with which 1.4 × a_wx value was the highest, and also on the paved roads (p<0.01), with which a_wz value was the highest.

2) Correlation between the WBV magnitude and velocity

Analysis of the WBV value of each axis and the approximately constant velocity in the farm fields showed a significant correlation, with a coefficient of 0.66 for the x-axis, 0.79 for the y-axis, and 0.80 for the z-axis (p<0.01, for all) (Fig. 2). After excluding the fertilizer distribution task, the correlation coefficient was 0.69 (p<0.01) for the x-axis, 0.48 (p=0.08) for the y-axis, and 0.59 (p=0.03) for the z-axis. The values for the paved roads showed no significant correlations, with a coefficient of −0.018 for the x-axis, −0.005 for the y-axis, and 0.014 for the z-axis.

Fig. 2.

Correlation between WBV magnitude and approximately constant velocity in the farm fields.

Daily WBV exposure for the farm tasks

The subject performed farm tasks using AMDSs 191 times throughout the study year, except for the task of “Sirokaki” (task ID=“D” in Fig. 1) and harvesting of soybeans (task ID=“N”). The farm tasks performed using AMDSs but without WBV measurements were as follows: distributing herbicides, rice husks, and poultry manure using the same implements as those used to distribute fertilizer; transporting soil to the farm field and distributing the soil using a shovel; treading barley plants with a roller implement; cleaning the inner sides of concrete dikes; and cutting grass using a flail mower.

The farm tasks were sorted according to content into groups of plowing, transplanting, seeding, harvesting, breaking subsoil (including digging ditches at the rim of the farm field and making drains), distribution, and others (including daubing mud on dikes, digging ditches inside the farm field, transporting soil, treading barley plants, cleaning the inner sides of dikes and mowing). All A_sum(8) and A_max(8) values for the farm tasks performed by the subject over the year were classified into two categories: less than or equal to 0.35 m/s² or higher for A_sum(8) and less than or equal to 0.5 m/s² or higher for A_max(8). The working times classified according to A(8) in each category, the mean A(8), and the total/mean of T_WinF and T_Exp for each task group are shown in Table 3.

Table 3. Working times and number of days classified according to A(8), mean A(8), total/mean of hours worked (T_WinF), and duration of exposure to WBV (T_Exp) with respect to each task group and month

*  The values in the case of the category “month” include the transportation time between parking the agricultural machine in the garage and transporting it to farm fields.

The numbers of tasks performed on the farm, except for the other task group, in descending order were 59 (30.9%) for plowing, 37 (19.4%) for harvesting, and 23 (12.0%) for transplanting. The proportion of tasks in which A_sum(8) exceeded the JOEL was 100% for breaking subsoil and seeding, 95.7% for transplanting, and 89.8% for plowing. The proportion of tasks in which A_max(8) exceeded the EUEAV was 63.6% for breaking subsoil, 28.8% for plowing, and 20.0% for seeding.

The mean T_WinF for each task in descending order was 256 minutes for seeding, 221 for transplanting, and 179 for plowing. The mean T_Exp was 230 minutes for seeding, 167 for plowing and 150 for transplanting. The rate of T_Exp against T_WinF was 93.3% for plowing and 89.9% for seeding, and those for distribution, harvesting, and transplanting were 65.2%, 66.5, and 67.9%, respectively. On the basis of the video footage, time-consuming tasks that did not involve operating AMDSs included placing fertilizer in the applicator for distribution, moving harvested grain from the tank to bags for transport during harvesting, and placing rice plants in the planting machine for transplanting.

Daily WBV exposure throughout the year

The subject sometimes performed multiple tasks using different machines on a single day and consequently worked using AMDSs on 159 days (43.6% of a year) during the study year. The distribution of the daily exposure, A(8) and T_Exp, on all days is shown in Fig. 3. The mean T_Exp per day was 202 minutes (min 10–max 556 minutes). The mean A_sum(8) was 0.61 m/s² (0.16–1.30 m/s²), and the number of days on which A_sum(8) exceeded the JOEL was 143 (89.9%). The mean A_max(8) was 0.39 (0.10–1.02), and the number of days on which A_max(8) exceeded the EUEAV was 38 (23.9%).

Fig. 3.

Distribution of the daily WBV exposure A(8) and duration of exposure to WBV (T_Exp) throughout the study year.

Of note, the subject worked for 32 consecutive days beginning on May 4^th, during which all the A_sum(8) values exceeded the JOEL. In the hot and humid season, June, July, and August, the A_sum(8) values on 23 days (=all days) exceeded the JOEL, in particular, the A_sum(8) values on three days in July were the highest for the entire study year.

The numbers of days classified according to A(8), mean A(8), total/mean T_WinF, and T_Exp on a monthly basis are shown in Table 3. In May, the subject worked for 30 days, with only one day of rest, and transplanted rice seedlings for half of these days. During May, the total T_Exp exceeded 100 hours, the number of days on which the A_sum(8) values exceeded the JOEL was 28 (93.3%) and the number of days on which the A_max(8) values exceeded the EUEAV was six (20.0%). In September, the subject worked for 24 days, harvesting rice on all those days, and the number of days on which the A_sum(8) values exceeded the JOEL was 23 (95.8%); the number of days on which the A_max(8) values exceeded the EUEAV was one (4.2%). In addition, the subject reported having to work for a long time without AMDSs to perform postharvest work, including removing, drying, packing, and transporting rice husks for handling.

Discussion

WBV magnitude

Our measurements of a_sum values in the fields ranged from 0.65 to 1.71 m/s², depending on the type of agricultural machine and farm task. The range and variety for the type of AMDS and driving surface conditions were similar to previous findings^{1, 11–13)}. Regarding the direction of vibration, there were no dominant axes for the magnitudes of vibration in the field. The magnitude of horizontal vibration, with a multiplication coefficient of 1.4, was higher than for vertical vibration, in all machines except for that used in fertilizer distribution, which had an extraordinarily high velocity. This finding indicated that vibration in the vertical direction is hard to generate by agricultural machines because of soft surface soil and low velocity and that a tractor has a mechanical fore-and-aft load caused by breaking up the soil while moving forward in the field. In particular, the horizontal vibration acceleration of the planting machine, even without the multiplication coefficient of 1.4, was 1.6 times higher than the vertical vibration acceleration. The lack of horizontal stability, caused by the large and narrow wheels of the planting machine, and lack of vertical vibration in the wet paddy field are suggested explanations of these findings. The adoption of vibration acceleration in a single axis most likely underestimates the risk of WBV exposure from AMDSs.

Validity of the WBV measurements

European Standard EN 1032 states that “the integration time for measurements shall be as long as normal, not less than three minutes¹⁷⁾”. We obtained an analysis time of more than eight minutes in the fields, even for fertilizer distribution tasks, which are relatively short, and over half an hour for the other tasks. The total T_WinF for the year was 543 hours, as shown in Table 3, and the total analysis time in the farm fields for the 17 measurements was 18.5 hours. This is more than sufficient time to obtain valid measurements compared with the standard three minutes for every eight hours worked. Our measurements of the differences in surface conditions are also valid, as we obtained the measurements in more than three fields.

Daily WBV exposure for the farm tasks

For plowing, both the number of days on which the subject worked using AMDSs and the A_sum(8) values exceeded the JOEL and were the highest among all the task groups. The mean WBV to which the subject was exposed while plowing was 1.08 m/s² [calculated from the mean A_sum(8) and the mean T_Exp for the group in Table 3] for an approximate three-hour stretch based on the mean T_Exp (167 minutes) and the rate of the mean T_Exp (167/179=93.3%) to the mean T_WinF. Since plowing carries such a high level of exposure to WBV, it is a high-priority task, and appropriate measures, such as using anti-vibration seat cushions, should be applied.

For transplanting, the number of days on which the A_max(8) values exceeded the EUEAV was zero, whereas the number of days on which the A_sum(8) values exceeded the JOEL was 22 out of 23 work days, depending on the characteristics of the vibration acceleration for the three directions of the planting machine. To clarify why estimations using A_sum(8) and A_max(8) differed so widely, additional data on exposure to WBV in planting machines needs to be obtained.

The A_sum(8) values for the task of breaking subsoil exceeded the JOEL on all the days worked, although this task was only performed on 11 days during the year. The AMDSs vibrated strongly under the heavy loads they were subjected to when digging up subsoil, and as a result, the level of WBV was high (a_sum of 1.16 m/s² for making drains using T1; a_wz,Base of 1.59 for digging at the rim of the farm field using T3). Agricultural workers should take measures to reduce their exposure to WBV during such tasks and take care of their body before commencing work.

The distribution tasks were performed at an extraordinarily high velocity in our WBV measurements. In these and other tasks, the magnitude of WBV was directly proportional to the approximately constant velocity in the field, and decreasing the running velocity of the AMDS was effective in reducing exposure to WBV. However, decreasing the running velocity also decreased work efficiency and increased T_Exp.

Seasonal daily WBV exposure

In May and September, the subject was very busy and was exposed to high levels of WBV. The number of days on which the A_sum(8) values exceeded the JOEL was 28 in May and 23 in September. In the industrial field, it is almost impossible to work a normal five-day week. The subject worked on AMDSs for 32 consecutive days beginning on May 4^th and worked for a long time without an AMDS when performing postharvest work in September. This indicates that during these two months, the busiest months of the year, it is necessary to manage the health of rice farmers carefully in order to prevent fatigue and musculoskeletal symptoms resulting from long work hours and excessive exposure to WBV.

Subject and machines

The exposure to WBV over one year is determined by the total area of cultivation, the number of workers, the AMDSs used, and the environment of the farmland, such as the area of each field and basic regional slope. The average area of cultivation for Japanese rice farming is approximately 1 ha¹⁸⁾, and many farmers cultivate in hilly and mountainous areas. In contrast, farmers in Hokkaido, which is in northern Japan, cultivate on a much larger scale compared with the family farm evaluated in this study. Thus, the results of one rice-farming family with a cultivation area of 16 ha in a plain region do not sufficiently represent the exposure of farmers to WBV from AMDSs in the farmland environments of Japan as a whole. In particular, using the same manufacturer for all the AMDSs had an influence on the overall magnitude of WBV, as previous studies have shown a wide range of values for machines made by different manufacturers^{1, 11, 12)}. It is therefore necessary to carry out future investigations, which should include farmers involved in large-scale farming with heavy machinery and small-scale farmers in hilly and mountainous areas.

Conclusion

In the study year, the full-time rice farmer worked for 159 days using AMDSs. The proportion of days on which the A(8) values exceeded the threshold limits for Japanese occupational exposure was 90%, and the proportion was 24% for the EU exposure action value. Unlike other industries, A(8) values for farming vary seasonally. When using AMDSs, the WBV levels are relatively high for tasks with a high velocity and heavy mechanical load. None of the AMDSs exhibited a dominant axis for WBV in farm fields. Therefore, evaluation of the degree of WBV exposure from AMDSs using the vibration acceleration for a single axis is questionable. These findings emphasize the need for rice farmers to utilize health management strategies suited to the farming seasons and measures to reduce exposure to WBV for each farm task.

Acknowledgment: We are grateful to the rice farmer and his family for participation in the vibration measurements and questionnaire survey in this study. Additionally, we would like to specially thank Mr. Masato Katoh for assisting in the data analysis and Dr. Mamoru Hirata for giving us a great deal of worthy advice.

References

• 1)   Bovenzi  M,  Betta  A. Low-back disorders in agricultural tractor drivers exposed to whole-body vibration and postural stress. Appl Ergonomics 1994; 25: 231-41.
• 2)   Fulmer  S,  Punnett  L,  Slingerland  DT,  Earle-Richardson  G. Ergonomic exposures in apple harvesting: preliminary observations. Am J Ind Med 2002; 2: 3-9.
• 3)   Earle-Richardson  G,  Jenkins  P,  Fulmer  S,  Mason  C,  Burdick  P,  May  J. An ergonomic intervention to reduce back strain among apple harvest workers in New York State. Appl Ergonomics 2005; 36: 327-34.
• 4)   Clementson  CL,  Hansen  AC. Pilot study of manual sugarcane harvesting using biomechanical analysis. J Agric Saf Health 2008; 14: 309-20.
• 5)   Mallick  Z. Evaluation and optimization of handle design parameters of a grass trimming machine. Int J Occup Saf Ergonomics 2008; 14: 343-9.
• 6)   Burdorf  A,  Hulshof  CTJ. Modelling the effects of exposure to whole-body vibration on low-back pain and its long-term consequences for sickness absence and associated work disability. J Sound Vib 2006; 298: 480-91.
• 7)   Osborne  A,  Blake  C,  Fullen  BM, et al. Prevalence of musculoskeletal disorders among farmers: a systematic review. Am J Ind Med 2012; 55: 143-58.
• 8)  The European Parliament and the Council of the European Union. Directive 2002/44/EC on the minimum health and safety requirements regarding the exposure of workers to the risks arising from physical agents (vibration). Official Journal of the European Communities 2002; L177: 13-9.
• 9)  International Organization for Standardization. Mechanical vibration and shock—Evaluation of human exposure to whole-body vibration—Part 1: General requirements. ISO 2631-1 1997.
• 10)  The Japan Society for Occupational Health. Recommendation of occupational exposure limits (2013–2014). J Occup Health 2013; 55: 422-40.
• 11)   Tsujimura  H,  Taoda  K,  Kitahara  T,  Nishiyama  K. Construction of a measuring system for an exposure to whole-body vibration and the vehicle's behaviour synchronously. Sangyo Eiseigaku Zasshi 2009; 51 (Suppl): 250 (in Japanese).
• 12)   Futatsuka  M,  Maeda  S,  Inaoka  T,  Nagano  M,  Shono  M,  Miyakita  T. Whole-body vibration and health effects in the agricultural machinery drivers. Ind Health 1998; 36: 127-32.
• 13)   Boshuizen  HC,  Bonges  PM,  Hulshof  TJ. Self-reported back pain in tractor drivers exposed to whole body vibration. Int Arch Occup Environ Health 1990; 62: 109-15.
• 14)   Bovenzi  M,  Zadini  A. Self-reported low back symptoms in urban bus drivers exposed to whole-body vibration. Spine 1992; 17: 1048-59.
• 15)   Funakoshi  M,  Taoda  K,  Tsujimura  H,  Nishiyama  K. Measurement of whole-body vibration in taxi drivers. J Occup Health 2004; 46: 119-24.
• 16)   Tsujimura  H,  Taoda  K,  Nishiyama  K. Exposure to whole-body vibration of forklift truck operators in dockyards—actual exposure in Japan and evaluation by EN 13059. Sangyo Eiseigaku Zasshi 2006; 48: 157-68 (in Japanese).
• 17)  European Committee for Standardization. Mechanical vibration—Testing of mobile machinery in order to determine the vibration emission value. European Standard EN 1032:2003+A1: 2008.
• 18)  Japanese Ministry of Agriculture, Forestry and Fisheries. Report on results of 2010 world census of agriculture and forestry in Japan: Summary documentation of surveillance report in agriculture/forestry management unit. [Online]. 2013 [cited 2013 Sep 17]; Available from: URL: http://www.maff.go.jp/j/tokei/census/afc/2010/houkokusyo.html (in Japanese).