| 摘要: |
| 步行是老年人日常的主要出行和活动方
式。近年来，高度城市化发展使城市环境和生活
方式发生转变，降低了老年人健康生活的质量。
为了提高老年人的生活质量，如何优化社区环境
以提高步行次数促进其身心健康的发展成为人
们关注的热点问题。基于山地城市贵阳市中心城
区老年人步行行为调查数据，利用梯度提升决策
树构建老年人步行次数模型，以探究建成环境
对老年人步行次数的非线性作用及阈值效应。研
究结果表明：第一，建成环境对老年人步行次数具有重要影响，且非线性影响与阈值效应明显，15 min生活圈的适老化建设应结合建成环境的有
效范围；第二，山地城市土地资源有限的缺陷下，提高人口密度与土地利用混合度以及较平坦的地
形（0%~5%）反而与老年人步行次数负相关；第三，地形起伏>15%负向影响老年人步行次数，而在
5%~15%间影响微弱；第四，提高居住密度、避免环境过度机动化以及保障老年人休闲与购物的
小尺度社区有利于其步行次数增多。针对研究结果提出相关建议，以期为山地城市15 min生活圈
的适老化建设提供科学参考。 |
| 关键词: 建成环境 老年人 步行次数 梯度提升决策树 地形起伏 |
| DOI：10.13791/j.cnki.hsfwest.20230209 |
| 分类号: |
| 基金项目:国家自然科学基金地区科学项目（71864008） |
|
| Study on the Influence of Built Environment in Mountain Cities on the Walking Times ofthe Elderly: A Case Study of the Central Urban Area of Guiyang |
|
LIU Simin,ZHAO Hang,HUANG Yong,WEI Wei,XIONG Renjiang,YUN Yihan
|
| Abstract: |
| Walking is the main daily travel and activity mode of the elderly. In recent years, high
levels of urbanization have led to changes in the urban environment and lifestyle, which have
reduced the quality of healthy living for older people. The single walking time and distance of the
elderly are affected by the physical function, which has the limit value and is not easy to change.
Increasing the walking times to reach a certain walking intensity has important health benefits to
improve the physical function and mental illness of the elderly. In addition, the walking times reflects
the walkability of the space environment. In order to improve the quality of life of the elderly, how
to optimize the community environment to increase the times of walking to promote their physical
and mental health development has become a hot issue of concern. Guiyang is a typical mountain
city in southwest China, and its population aging degree is deepening year by year. Combined with
the urban governance project of “one circle, two fields and three reforms” recently proposed by
Guiyang, it is necessary to further consider the suitability of walking environment for the elderly
in the construction of “15-minute life circle” suitable for aging. The urban space is affected by the
terrain, which is characterized by large non-linear coefficient of roads, poor connectivity of road
network, stepped layout of residential areas and uneven distribution of public service facilities. The
particularity of space environment can easily affect the elderly walking, reduce their willingness
to walk, and thus reduce the times of walking, which is not conducive to the development of
healthy aging. Secondly, mountainous cities are characterized by construction land shortage, dense
population and large topography, so the built environment and the elderly walking is not a simple
linear relationship. Gradient boosting decision tree is a data-driven algorithm, which does not need
to follow the strict prior assumptions of linear models on functional relations. It can eliminate
the interference of outliers, fit the influence of independent variables on dependent variables in
different ranges, and effectively quantify the influence degree of independent variables on dependent
variables. Therefore, based on the survey data of the walking behavior of the elderly in the central
city of Guiyang, a gradient boosting decision tree was used to construct a walking times model for
the elderly to test the relative importance of the built environment and explore the non-linear effect
and threshold effect of the built environment on the walking times of the elderly from the aspects ofthe built environment, the perception of the built environment, walking attitude, satisfaction and individual socioeconomic attributes. In addition, although
the topographic change of mountain cities can be represented by the size of the slope ratio, there are subjective differences in the elderly’s perception of
the terrain. By incorporating “slope ratio” reflecting relief degree into the built environment and “terrain condition perception” into the built environment
perception, the elderly walking times model con?歴浲???慥湤搠?べ???歡浤???湴?潢牯摯敳牴?瑮潧?灤牥潣癩楳摩敯?猠捴楲敥湥琠楣晡楮挠?牬敳景攠牴敥湳捴攠?晨潥爠?瑩桦敦????浣楥渠畢瑥整?汥楥普攠?换楪牥捣汴敩?潥映?浥潲畲湡瑩慮椠湣?捡楮瑧祥?獡畮楤琊慳扵汢敪?晣潴物?慥朠楴湥杲?捡潩湮猠瑰牥畲捣瑥楰潴湩?n on the walking times of the elderly. Firstly, the built environment has an important effect on the walking times of the elderly,
and the non-linear effect and threshold effect are obvious. The aging suitable construction of 15-minute life circle should be combined with the effective
range of built environment. Secondly, due to the limited land resources in mountain cities, increasing the mixing degree of population density and land use
as well as the flat terrain (0%~5%) is negatively correlated with the walking times of the elderly. Thirdly, topographic relief >15% had a negative effect on the
walking times of the elderly, while it had a weak effect between 5% and 15%, which was consistent with the relatively small importance of terrain perception
on the walking times of the elderly. Fourthly, small-scale communities that increase residential density, avoid over-motorization of the environment, and
ensure leisure and shopping for the elderly are conducive to increased walking times. Based on the research results, relevant suggestions are put forward.
Firstly, in commercial and central areas with a population density of 40,000 people /km 2 or a land use mixing degree greater than 0.4 or flat terrain (0%~5%),
the elderly should also be regarded as an important service group, and service facilities such as parks, squares and chess and card rooms should be reasonably
allocated based on the per capita share of the elderly. Secondly, in the area with large relief, especially when the slope ratio is greater than 15%, the
construction of small life circle should be perfected, and the density of leisure and entertainment and shopping stores should be ensured at 10/km 2 and 20/
km 2 . Finally, build a “people-oriented” small block, control the density of the road network and site within 7 km/km 2 and 3 /km 2 , and the residential density
and distance from the site are not less than 0.04 km 2 |
| Key words: Built Environment the Elderly Times of Walking Gradient Boosting Decision Tree Relief of Terrain |
