The purpose of the present paper is to offer an explorative contextual analysis of the relationship between aircraft noise exposure and behavioural reactions. The relationship between changed noise exposure and behavioural responses is analyzed in relation to visitor characteristics as well as other possible reasons for non-use. More specifically, the following questions are analyzed:

The two forest areas studied were Bygdøy, near the old main airport, and Romeriksåsen near the new airport. The forest area at Bygdøy is about 2.6 km², while Romeriksåsen is about 7,600 km². The areas were chosen on grounds of their location relative to the airports, and because they were much used by the local communities. Figures 1 and 2 show maps of the areas, and their location relative to the old and the new airport, respectively. Neither Bygdøy nor Romeriksåsen is situated more than about 11 kilometers from the homes of any of the respondents in the respective samples, which means that it is theoretically possible for them to use the area regularly. The air traffic over both areas was dominated by jet aircraft in route traffic both before and after the change. The change in exposure levels was largest at the old airport, since this airport was totally closed down. What remained after the change were overflights at high altitudes, and helicopter traffic. The new main airport was an existing airport, which was expanded. The areas and the aircraft noise levels are described at more detail elsewhere [10,11,53,54].

Telephone interviews were conducted with the same people at the same time of the year before and after the moving of the airport, in November 1997 and 1999 for Romeriksåsen, and in May–June 1998 and 1999 for Bygdøy. Throughout the paper we will use the term t₁ (time 1) whenever referring to data from the “before” situation in each of the areas, and t₂ (time 2) whenever referring to the “after” situation. Because the data collection period each time lasted for about four weeks, people interviewed during the first half of the period the first time would also be interviewed during the first half of the period the second time, and vice versa. This was done to secure that changes in frequency of use were not due to seasonal changes.

The main part of the questionnaire was presented only to the visitors of the areas [visitors should have visited the area at least once during the past three months, and at least two times during the past six months (Romeriksåsen), or two times during the past 12 months (Bygdøy)]. Another set of questions was posed to a sub-sample of non-visitors. Visitors were interviewed about their use of the area the past three months, while non-visitors were asked why they had not visited the area during the same period. At the start of the second interview the respondents were filtered with the same filter questions as before into one visitor- and one non-visitor-group, regardless of their status in the first interview. Details about the data collection are presented elsewhere [53].

Table 1 shows the number of interviews made at t₁ and t₂ as well as the number of dropouts (those who could not be interviewed twice, regardless of reason) at each point in time. Nine calls were made to reach the respondents at t₂. A comparison with field study samples from the same areas [10,11] indicated that the final visitor samples were representative of the users of the respective area in terms of gender, age, and educational level. To test whether the dropouts differed from the remaining samples of visitors from t₁ regarding the central variable annoyance with sound from aircraft, t-tests were conducted. The t-test assesses whether the means of two groups are statistically different from each other. No systematic differences were found. Multivariate logistic regression analyses were conducted to test whether the dropouts from the initial non-visitor group differed systematically from the remaining sample regarding reasons for non-use and earlier experience with the area (see the section “Analyses and Variables” for a description of the variables). Respondents who indicated that a reason for not visiting Bygdøy at t₁ was that they were not interested in outdoor recreational activities, or they had just moved in, or their health was too bad, were less likely to participate in the second survey than other non-visitors. Non-visitors who preferred other outdoor recreational areas at t₁, were less likely to drop out of the Bygdøy sample than other non-visitors. Lack of interest in outdoor recreational activities and bad health as reasons for being non-visitors at t₁ were also associated with a heightened probability to drop out of the Romeriksåsen sample. In addition, the non-visitors at t_1, who had visited the area earlier, were more likely to be in the final sample than subjects with no earlier experience with the area.

The analyses were conducted using the statistical package SPSS 9.0 for Windows. The analyses were restricted to the respondents who were interviewed both at t₁ and t₂. The samples from the two study areas were analyzed separately, to examine behavioural responses to either a decrease or an increase in noise exposure. The questionnaires for the first and the second interviews were, with one exception, identical until the conclusion of the second interview, which included some additional questions.

No significant differences were found in frequency of use among the respondents who visited either area both before and after the relocation of the main airport. There was no significant indication of a systematic relationship between the altered noise conditions and the overall transitions from non-use to use at Bygdøy, and from use to non-use in Romeriksåsen. The probability of a transition in both cases was positively influenced by the level of prior experience with the area. Earlier experience with the area was the only variable that significantly predicted behavioural change at Bygdøy, while in Romeriksåsen, in addition to use frequency, participation in the activities of picking berries or mushrooms, and running, jogging, also significantly decreased the probability of becoming a non-visitor at t₂.

When analyzing only the group of new visitors to Bygdøy, and the group who ceased to use Romeriksåsen, some indications of a relationship between altered noise exposure and altered use were found. The new visitors at t₂ for whom aircraft noise was a reason for none-use at t₁ tended to use Bygdøy more frequently after the moving of the airport than other new visitors. The visitors to Romeriksåsen most likely to cease using the area because of the increased aircraft noise exposure were visitors to the northern part of the area where the change in exposure was largest, the frequent visitors at t₁, and visitors for whom the nature experience was an important reason for participation in outdoor recreational activities. The last mentioned variable was the variable that most strongly predicted annoyance with sound from aircraft at t₁. No firm conclusions can be drawn when it comes to the question of different activities and effects of aircraft noise, since the results are not consistent across the analyses.

The time perspective of the study might, however, have influenced the behavioural effects that were measured. The effects that were measured are short-term effects. Both surveys at t₂ concerned the first season after the change corresponding to the time of the first interviews. An effect might have been more distinct the next season, because one could argue that people first would have to experience the new situation before it could affect their behavioural choices. Behavioural adjustments may happen gradually, since it may take some time for the recreationists to find new places to go, and to change their habits. On the other hand, in her study of behavioural consequences of management regulations on mountain camping, Vorkinn [48] found that a substantial number of campers seemed to base their behavioural response on expectations of how the regulations would affect them, rather than on actual experience. Another possibility is that people might habituate to the situation instead of changing their behaviour. These are empirical questions that cannot be answered through the present study.

The result indicated stability in frequency of use just as much as change in the group who visited Bygdøy both survey years. There were just as many who used the area less frequently as more frequently at t₂ than at t₁, which indicates that there was not a systematic relationship between use frequency and the changes in noise exposure in this group of visitors. One reason for this might simply be that the noise exposure did not interfere with the recreational experience of these visitors. But it was shown in another paper analyzing the same data [53] that the degree of annoyance with aircraft noise was high before the moving of the airport. Nearly 50 percent reported to have been rather or very annoyed by the sound from aircraft while visiting the area. The effect of the relocation of the airport on aircraft noise annoyance was marked, almost no one reporting to be rather or very annoyed after the change. Also, about half the sample rated Bygdøy as a better recreational area after the change than it was before. These results indicate that there was a perceived improvement in area qualities following the closing of the airport that potentially could influence use frequency. On the other hand, these visitors already used the area in the situation with high aircraft noise exposure in the area, despite being annoyed by the noise. Although aircraft noise was adversely influencing the recreational experience of these visitors, it seems not to have influenced their behaviour in a similar way. One of the most commonly reported constraints to increased leisure participation is lack of time [57]. Possibly, the visitors to Bygdøy at t₁ were already participating at the level they preferred or could, so that the improvement of the area quality could not mean much to their overall use frequency. Time constraints and accessibility might possibly also have influenced these visitors’ choice to use the area in the first place, despite being annoyed by the aircraft noise at t₁.

The results did not indicate a systematic relationship between reporting aircraft noise to be a reason for non-use at t₁ and transition to use at t₂. It seems that the noise exposure was overall not a decisive reason for non-use at t₁ in our sample. That is not to say that it was not one among several factors influencing behavioural choices. But the decision process regarding leisure participation is complex, with different types of constraints influencing the process at various stages [57]. If the aircraft noise exposure in the area was only one among several factors inhibiting or limiting the use of the area at t₁, a change in noise exposure alone would not be enough to prompt an overall change in behaviour in this heterogeneous group of non-visitors. But earlier experience with the area significantly increased the probability of a transition to use at t₂. We did not measure behavioural preference in our study, only reasons for non-use. The earlier use of the area might be interpreted as an indication of preference for using the area.

Also, some of the respondents categorized as non-visitors at t₁ may actually have used the area during the period asked for. To avoid the absolute accidental visitors, the criterion to be categorized as a visitor in our sample was set at more than one visit during the past three months. Robertson and Regula [51] found evidence of short term fluctuations in the use level that was not due to the social or environmental conditions in the area in their study of the impact of sedimentation on water-based recreation. Our result may also indicate fluctuations in use in the group of infrequent visitors, which were unrelated to the area conditions.

Aircraft noise as a reason for non-use at t₁ was related to frequency of use at t₂ when including only those who had become visitors at t₂. The result of this analysis indicates positive behavioural change following environmental change, which is a kind of change that has barely been studied, as pointed out by Vorkinn [48]. The validity of the result depends on the reasons for non-use being true expressions of the respondents’ motivations. The reasons were formulated by the researchers, not by the respondents. However, they covered a wide range of possibilities, both area related reasons other than aircraft noise, and common constraints to leisure in general [57]. The respondents were given the opportunity to answer “not sure”, but very few were not able to say whether the suggested reasons applied or did not apply to them. Only six out of the 700 non-visitors that were initially interviewed found none of the listed reasons to apply to them. This indicates that the listed reasons covered the respondents’ reasons pretty well. There was no special focus on noise in the introduction of the interview, or in the items presented. Thus, there should be no special reason to state aircraft noise to be a reason for non-use if it was not at all important. The non-significant relationships between use frequency and all other variables in the model were negative, suggesting that these variables limited use frequency for these respondents. While the results indicated behavioural change caused by the altered noise conditions, this supports the interpretation that at least some of the overall transition in use may be explained as short-term fluctuations in use in the group of infrequent visitors, whose use level may be limited by various factors other than noise.

Although the result is parallel to the result regarding decreased noise exposure, the situations facing the visitors are not the same, since the changes in aircraft noise exposure were the opposite. Other factors may explain stability in use in the face of area deterioration, than stability in use following improvement in area quality.

First, although there was no significant change in use frequency among those who used Romeriksåsen both before and after the increase in aircraft noise exposure, some kinds of behavioural adaptation in this group of visitors cannot be totally ruled out. A more sophisticated model could possibly have detected other types of behavioural adaptation. Our analysis could not detect temporal displacement, or intra-site displacement. Both temporal and intra-site displacement might have been feasible strategies in Romeriksåsen, and consistent with our findings of no systematic change in use frequency. Previous studies on coping with perceived adverse area factors have found that the use of behavioural adaptation within recreational areas may be pervasive [37,38,42,46]. For instance, Anderson and Brown [37] found that over time, a majority of the respondents either changed their entry points to the Boundary Waters Canoe Area Wilderness, or selected campsites differently, or entered on a different day of the week. Hall and Shelby [38] found that about half the visitors in their survey altered behaviour in response to crowding at the Lake Billy Chinook reservoir. They found temporal displacement to be the most common strategy, but intra-site displacement was also common. In a explorative study of coping in the Adirondack wilderness area Johnson and Dawson [42] also found that about half their respondents used some kind of coping strategies. In a cross sectional survey including local residents of the communities in and around the Acadia National Park [46] most community residents reported that they continued to use the carriage roads of the park at about the same level as before, despite increased use levels and problem behaviour. However, the authors [46] also found that almost all respondents used some kinds of coping strategies. The two last mentioned studies [42,46] examined the use of both behavioural and cognitive coping mechanisms. Temporal and intra-site displacement were the coping strategies reported most frequently in both studies. Both studies also found that a majority used more than one strategy in response to crowding or conflict [42,46]. Johnson and Dawson state that: “coping is more than just a singular decision, but a complex process that is part of a larger, flexible user decision-making strategy.” [42], p. 288.

Cognitive coping strategies were not examined in our study, but is another set of coping strategies that would be consistent with continued use in the face of perceived adverse changes in area conditions (e.g., [42,45–47,49,50]). The following description of the cognitive coping strategies of product shift and rationalization is given by Miller and McCool ([45], p. 262), with reference to Shelby et al. ([50], p. 276): “Product shift involves a “change in the definition of the experience and Standards for the importance of characteristics of that experience.” Product shift also represents an overall change in the definition of the area. The net result of this is that satisfaction remains high and recreationists are not obliged to remove themselves either physically or temporally from the area. Rationalization represents a process whereby recreationists re-evaluate an undesirable situation in a more favourable light.” Although the existence of cognitive adjustments cannot be ruled out in our study, there are some indications of the opposite. It was shown in another paper analyzing the same panel data that the respondents who visited both at t₁ and t₂ perceived the quality of the recreational area to have deteriorated after the change due to the increased aircraft noise exposure in the area [53]. That is, it seems that this group of visitors tends to continue to use the area as they used to, despite the perceived deterioration of the area quality caused by the increased aircraft noise exposure.

The finding is similar to results in Vorkinn’s panel study [48] of visitor response to management regulations of “wild camping”. She found that a relatively high percentage of the existing users who revisited the site after the regulations were dissatisfied. Despite that, the experienced users were more likely to adapt to the new situation, than to give up camping in the area. Vorkinn ([48], p. 744), suggests that “a possible explanation for this is that people develop emotional bonds through repeated use of an area, which gives the area/site a value or meaning that other areas/sites do not have. If this is the case, alternative areas with similar environmental attributes will not be experienced as real substitutes.” The existence and consequences of emotional bonds to, and dependence on, specific sites or areas have been examined under the label “place attachment” [48,60–64]. A relationship between place attachment and continued use in the face of real or potential adverse changes in area conditions has also been indicated by others [40,65,66]. More specifically, there are indications that place attachment may be an especially important factor in regard to behavioural responses to adverse changes in the quality of local recreational areas [66]. Attachment to outdoor recreational areas has among other things been found to be related to the proximity of the area to the homes of the visitors and use frequency [48,61–64,66]. Frequency of use at t₁ was adversely associated with the likelihood of becoming a non-visitor at t₂, which indicates that the people who visited both years tended to be the most frequent visitors in our sample. While place attachment was not directly measured in our study, the results indicate that this variable should be included in future studies of noise impacts in local recreational areas. Constraints related to distance from home and travel time may further limit the substitutability of local areas, despite perceived adverse area conditions [39,40].

The purpose of this analysis was to identify the visitors at t₁ most vulnerable to aircraft noise, to examine the possible relationship between vulnerability to aircraft noise and behavioural change at t₂, not to examine what variables predict noise annoyance in an outdoor recreational situation per se. The geographical variables are specific to this particular area. The noise situation was a particular low noise exposure situation. Also, both annoyance and visitors’ characteristics were measured in a rather broad sense, encompassing the experiences and behaviours of a three months period. The R² was low, which indicates that the variables in the model explained only a small amount of the variance in noise annoyance. One variable obviously lacking was individual noise exposure, which was not obtainable for this study. The result that annoyance with aircraft noise in the recreational area was positively related to the nature experience as a reason for participation in outdoor recreational activities is, however, in accordance with previous findings [10,11]. The result that the visitors to the northern part of the area were significantly less annoyed than other visitors is in accordance with what one would expect from the exposure situation at t₁, which validates the result.

The variables that predicted annoyance with aircraft noise at t₁ did not predict transition from use to non-use at t₂. Instead, frequency of use and participation in the activities “picking berries or mushrooms”, and “running, jogging” at t₁ were associated with a decreased probability of ceased use at t₂. Our result regarding frequency of use is in line with findings in Vorkinn’s study [48] of regulations on wild camping in a Norwegian mountain area. The more experience the respondents had in the area, the more likely they were to adapt to the new situation and stay at a mountain camp, instead of ceasing to camp in the area. Vorkinn suggests that place attachment may be a possible explanation for this finding, the emotional bonds giving the area a special value or meaning that makes it hard to find a substitute area. The possible role of place attachment in our study was discussed above. Frequency of use may indicate place attachment, and place attachment is shown to be related to a decreased probability of ceased use. Hall and Shelby [38] found that a higher percentage of the most experienced users of a large reservoir, measured in year since first visit to the area, used temporal displacement, or a combination of temporal and spatial displacement than the less experienced visitors. Spatial strategies alone where most often used by the least experienced visitors. Although “frequency of use” the past three months and “time since first visit” are measuring different aspects of experience, the results from our study and the study of Hall and Shelby [38] are in line, in that the more experienced visitors seem to be less likely to use the most extreme coping response, being displaced from the site altogether. “Picking berries or mushrooms” is a kind of activity that obviously is dependent on special area resources, which is a likely reason for the decreased probability of ceased use for those who participated in this activity. “Running, jogging” is an activity that one would assume to be less sensitive to the experiential qualities of the area, and more focused on practical aspects like accessibility and suitability for the activity.

The two variables that were shown to significantly increase the likelihood of reporting aircraft noise to be a reason for non-use at t₂ were also significantly related to aircraft noise annoyance at t₁. Visitors to the northern part of the area tended to be less annoyed by aircraft noise at t₁, but more likely than other visitors to report aircraft noise to be a reason for ceased use at t₂. The result is plausible, since the actual change in aircraft noise exposure was estimated to be largest in the northern part of Romeriksåsen. Apart from the larger increase in aircraft noise exposure in the northern part than in the other parts, interference from aircraft noise might also have been perceived as stronger in the northern part because this part of the area was the most pristine in other ways. The nature experience as a reason for participation in outdoor recreational activities was the variable indicated to have the largest effect on noise annoyance at t₁, and was also related to increased probability of reporting aircraft noise as a reason for non-use at t₂. Ceased use at t₂ was, overall, shown to be less likely for the more frequent visitors at t₂ than for the more infrequent visitors. But some of the frequent visitors did also become non-visitors. Too much aircraft noise was a more likely reason for ceased use at t₂ for the previously frequent visitors than for those who visited more infrequently at t₁. One of the previously frequent visitors expressed “deep deep sorrow because of all the aircraft noise that has destroyed my favourite outdoor recreational area.” The statement was given in response to an open-ended question where the non-visitors were asked to elaborate on “other reasons” for non-use at t₂. This particular person visited more than twice a week at t₁, but had not visited at all the past three months at t₂. The statement indicates place attachment and serious impact for this particular previous visitor. Although only the statement of one visitor, it supports the suggestion that place attachment may be an important variable to include in future studies of noise impacts in local outdoor recreational areas. Overall, the results show that although there was no systematic cessation in use due to aircraft noise, to some the character of the area was changed in a way that appears to hamper their goal attainment in the area.