The Comparisons Of On-site And Off-site Applications In Surveys On Perception Of And Preference For Urban Green Spaces

With the rapid development of urbanization,urban infrastructures are increasingly perfect and the construction level of urban public green space is getting higher.Urban public open green space plays an important role in residents’ life,promoting residents’ physical and mental health and stress recovery.With the continuous improvement of residents’ living quality,the demand for urban public open space is also on the rise.In order to meet the increasing demands of urban residents for the high quality of urban public green space,scientific researchers have gradually shifted their research focus to the human-oriented design of urban green space.By investigating the landscape perception and preference of different stakeholders,such as urban residents,tourists,green space workers and designers,the landscape planning and design to meet the needs of different groups of people were discussed.Investigators of landscape perception and preference usually conduct landscape perception and preference research by means of on-site survey,photo elicitation and virtual reality according to their own preferences and conditions.However,different researchers used different visual approaches,which often led to inconsistent and even contradictory results in their research.Therefore,in order to identify the reliability and validity of different visual approaches in landscape perception and preference evaluation,this study selected the different types of urban public green space of Xi’an,using the on-site survey,photo elicitation,and the virtual reality technology,crossing the four seasons of spring,summer,autumn and winter,and using different selections of participants,to examine the urban green space landscape perception and preference of college students.Non-parametric tests(Kruskal-Wallis and Mann-Whitney tests)and generalized linear models are used to test the differences between different visual approaches.The research results showed that:(1)The effects of on-site survey and off-site survey(photo elicitation and virtual reality)in landscape perception and preference research are not exactly the same.This was particularly evident in the striking differences found between the photo elicitation and the on-site survey.In contrast,the effect of virtual reality was more consistent with on-site survey.Therefore,the virtual reality is more reliable than the photo elicitation in terms of the proximity to the on-site approach.(2)Season is the factor that influences the perception and preference for urban green space with different vegetation structure,while the method of participants selection is not the influential factor.Therefore,the reliability of visual approaches is affected by both seasonal factors and vegetation structure factors.(3)For open green space in winter and closed green space in autumn,both off-site approaches of photo elicitation and virtual reality are reliable.In other words,when the research object is green space with obvious spatial characteristics,such as open green space,photo elicitation can be used instead of on-site survey.When the object of study is more complex green space,such as semi-open green space,and the site environment has obvious identifiable atmosphere,virtual reality can be used instead of on-site survey.(4)The perception and preference of the landscape generated by photo citation and virtual reality is not as strong as that reflected in the on-site survey.Therefore,it is suggested in practice that appropriate visual approaches should be carefully chosen for landscapes with different characteristics.The results above indicated that in the practice of landscape perception and preference,different visual approaches should be adopted according to the vegetation structure of green space,the seasonal dynamic characteristics of biometrical factors,and the inherent differences among different approaches.Both on-site and off-site strategies can be applied to the type of green space with the most typical vegetation characteristics,while off-site approaches should be carefully applied to sites where it is not easy for participants to relate to real experience perception.In addition,in the green space with obvious spatial structure characteristics,photo elicitation could replace on-site survey.For more complex green space(such as semi-open green space)and green space where the ambient atmosphere can be easily identified,virtual reality is more suitable to replace the on-site survey.These results can provide scientific basis for obtaining more accurate landscape perception and preference evaluation in the future.