| 摘要: |
| 乡村景观空间形态定量分析对于探索乡村空间发展规律、提升乡村人居环境具有重要意义。面对乡村分类建设与空间规划实践需求，构建了乡村景观空
间形态四维定量指标体系。遴选江苏省南京市江宁区14个村落作为实验样区，采取系统聚类法描述长三角丘陵地区乡村景观空间特征与分类，使用主成分分析法
探讨形成乡村景观类型的空间主控因子。研究结论如下：1)研究区乡村景观空间形态可聚类为山地丘陵塘库型、缓岗坡地塘坝型、平原圩区水网型3类具有明显地
貌差异的组团，其中，山地丘陵乡村最突出的空间形态特征是土地利用强度低、空间可达性低、地表粗糙度高且综合土地利用动态度高；2)主成分分析将多指标
综合为3个主成分，包括四维综合主控因子PC1(42.53%)、乡村建成空间类因子PC2(16.74%)和半自然生态空间类因子PC3(14.51%)，共解释了73.78%的
乡村景观空间形态特征；3)经验证，相较传统定性分类描述，基于多维度指标的空间定量分析结果更加精细、系统，且可复制性更强 |
| 关键词: 风景园林 乡村景观 空间形态 长三角地区 多维量化 |
| DOI：10.19775/j.cla.2025.03.0068 |
| 投稿时间：2023-11-13修订日期：2024-01-23 |
| 基金项目:浙江理工大学科研启动基金(23052173-Y)；国家重点研发计划项目子课题(2019YFD11004032)；浙江省教育厅科研项目(Y202456259) |
|
| Rural Landscape Spatial Morphology Based on Four-dimensional Quantization and MultivariateStatistics |
| CHEN Siqi,ZHANG Yujun* |
| Abstract: |
| Quantitative analysis of the spatial morphology of rural landscapes is
of great significance for exploring the rules of rural space and improving the rural
living environment. Fourteen villages in Jiangning District, Nanjing City, Jiangsu
Province were selected as experimental sample areas. Facing the practical
needs of rural classification construction and spatial planning, a four-dimensional
quantitative index system of rural landscape spatial form was constructed.
It includes two-dimensional morphological indicators based on landscape
ecology, two-dimensional morphological indicators based on urban morphology,
three-dimensional surface measurement indicators, four-dimensional historical
dynamic indicators, and other spatial indicators. The spatial morphological
analysis in this study goes beyond the confines of the two-dimensional form of
rural residential space, but more finely evaluates the morphological differentiation
characteristics in terms of spatial accessibility, landscape heterogeneity, area
proportion of different components, three-dimensional landscape, and historical
dynamics. It also highlights the complex characteristics of rural areas under
the joint action of nature and humans through indicators such as the area
proportion, dynamics, and fractal index of semi-natural patches. Based on the
construction of a four-dimensional index system and the acquisition of index
data, Spearman's correlation coefficient and one-way ANOVA were used to
explore the correlation between the various landscape spatial morphological
indicators and the significance of the index differences among different village
samples. In order to further explain the similarities and differences in the types
of rural landscape morphology, cluster analysis was performed on the spatial
characteristic indicators, and then principal component analysis was used to
explore the spatial main controlling factors that formed the rural landscape types.
The key to the research is to answer: 1) How to construct a rural landscape
morphology index system that highlights semi-natural features and integrates
four-dimensional spatial information; 2) How to classify the morphology of rural
landscapes based on a multidimensional index system to explain the existing
adaptive rural landscape characteristics and provide useful quantitative guidance
for future rural space planning and classification construction optimization.
The research conclusions are as follows: 1) The spatial morphology of the rural
landscape in the hilly area of the Yangtze River Delta can be clustered into three
groups with obvious landform differences: hilly-pond type, slope-pond type,
and plain-polder type. The most prominent spatial morphology characteristics
of hilly villages are low land use intensity, low spatial accessibility, high surface
roughness, and large land use dynamics, while plain-polder villages are the
opposite. 2) Principal component analysis combined multiple indicators into
three principal components, which explained 73.78% of the rural landscape
spatial features, including the four-dimensional comprehensive main control
factor PC1 (42.53%), the rural built-up space factor PC2 (16.74%), seminatural
ecological space factor PC3 (14.51%). In future practical applications, it
is advisable to give priority to indicators such as the proportion of semi-natural
habitat area, spatial integration, surface roughness, and land use dynamics
(PC1). Secondly, it is necessary to consider the rural built-up space factors
(PC2) and semi-natural ecological space factors (PC3). The former mainly
includes indicators such as settlement boundary index, road density, cohesion,
and urban distance, while the latter mainly includes indicators such as seminatural
patch dynamics, forest area proportion, Shannon diversity, and water
area. Villages with high rural built-up space factors usually have more complex
settlement boundaries, higher road density, more concentrated rural patch
distribution, and closer distance to the city, while villages with high semi-natural
ecological space factors maintain a rural landscape with a higher proportion of
blue-green space, stronger heterogeneity, and longer time. 3) It has been verified
that compared with the traditional qualitative classification description, the spatial
quantitative analysis results based on multi-dimensional indicators are more
refined and systematic, and more reproducible. In subsequent studies, we can
try the whole process of machine recognition and analysis methods, and extend
the application of four-dimensional spatial morphological quantitative indicators
to the statistical analysis of landscape spatial morphology in larger sample sizes
and different geographical regions. This method can serve as an auxiliary tool
for rural planning. From the aspects of two-dimensional space planning, threedimensional
vertical planning, and four-dimensional governance timing, it can
guide the formulation of appropriate differentiated rural management strategies.
Planning tools that combine large amounts of ecological, social, and economic
data are slowly entering the field of rural research. Future research needs to
continuously integrate emerging technologies and multi-source data, based on
rural foundations, to intelligently promote rural green development, improve the
rural living environment with high quality, and solidly promote the comprehensive
revitalization of rural areas. |
| Key words: landscape architecture rural landscape spatial morphology Yangtze River Delta region multidimensional quantization |
