不同尺度空間插值法進行人口分佈推估的比較研究

A Comparison of Methods for Spatial Interpolation across Different Spatial Scales

林庭嘉(Lin Ting-Jia)

指導教授 ： 張國楨

國立臺灣師範大學/文學院/地理學系/碩士(2021年)

https://doi.org/10.6345/NTNU202101140

摘要

近年來，雲端運算及物聯網應用技術的進步對於社會經濟、醫療公衛、自然科學等領域都具有非常重要的幫助。隨著政府開放資料(open data)之推廣應用，各類型的資料透過加值與開放應用，創造出更多跨領域的應用成果。然而，伴隨豐富且多元的應用對於數據的解析度要求更高，常常會遇到小尺度數據缺少亦或者沒有更細緻的調查數據等問題。因此透過現有的數據經空間內插方式，進而得到更細緻的資料應用。
本研究以戶籍人口分佈資料作為驗證對象在不同尺度下進行空間內插推估分析，搭配與戶籍人口關聯之建蔽率、容積率、樓地板面積等作為推估的輔助資料以此建立人口推估的模型，進而得到更細緻的建築物層級之戶籍人口分佈。研究目標：建立不同尺度下對於人口分佈的推估模式，並找出與戶籍人口關聯之變數，並提供小尺度的單元在戶籍人口上的推估模式。
研究成果顯示：變數中樓地板面積、容積率對於戶籍人口有顯著的相關性，運用地理加權回歸對最小統計單元之戶籍人口推估出建築物層級的戶籍人口分佈，並經過交叉驗證在建築物尺度下推戶籍人口數其推估相對誤差的準確度平均值為3.554766(人)。

關鍵字

最小統計區 ； 地理資訊系統 ； 容積率 ； 建蔽率 ； 地理加權回歸

並列摘要

In recent years, advances in cloud computing and Internet of Things (IoT) applications have been of great importance to the fields of socio-economic, medical, and public health, and natural science. With the promotion of the application of open data, various types of data can be used to create more cross-disciplinary applications through value-added and open applications.
However, with the abundant and multi-dimensional applications requiring higher resolution of data, there are often problems such as the lack of small-scale data or the absence of more detailed survey data. Therefore penetrates the existing data after the spatial interpolation way, then obtains the most detailed information for application.
In this study, the spatial interpolation of the household register population distribution data is used as the validation object to conduct the spatial interpolation, analysis at different scales, and the building coverage rate, floor area ratio, and floor area associated with the household register population are used as the auxiliary data for the estimation to establish the population estimation model, and to obtain a more detailed distribution of the household register population at the building level. The aim of the study is to develop a model for estimating population distribution at different scales, to identify variables associated with the household register population, and to provide a model for estimating the household register population for small-scale units.
Research results show that: a variable in the floor area, floor area ratio for the population of a significant correlation, using geographically weighted regression to smallest statistical unit census register population to estimate the building level of census register population distribution, and scale through cross certification in building census register population estimate its average relative error of accuracy is 3.554 (people).

並列關鍵字

Basic Statistical Area ； GIS ； Floor Area Ratio ； Building Coverage Rate ； Geographically Weighted Regression

參考文獻

英文文獻

Comber, A., Proctor, C., & Anthony, S. (2008). The creation of a national agricultural land use dataset: Combining pycnophylactic interpolation with dasymetric mapping techniques. Transactions in GIS, 12(6), 775–791. https://doi.org/10.1111/j.1467-9671.2008.01130.x.

David J Briggs, Susan Collins , Paul Elliott , Paul Fischer , Simon kingham , Erik Lebret , Karel Pryl , Hans Van Reeuwijk , Kirsty , smallbone & Andre Van Der Veen(1997). Mapping urban air pollution using GIS: a regression-based approach. International Journal of Geographical Information Science, 11:7,699-718, DOI: 10.1080/136588197242158

Daniele Da Re ,Marius Gilbert,Celia Chaiban,Pierre Bourguignon,Weerapong Thanapongtharm,Timothy P. Robinson,Sophie O. Vanwambeke(2020). Downscaling livestock census data using multivariate predictive models: Sensitivity to modifiable areal unit problem.PLoS ONE. 2020;15(1):1–16. https://doi.org/10.1371/journal.pone.0221070

Goovaerts, P. (2000). Geostatistical approaches for ncorporating elevation into the spatial interpolation of rainfall. Journal of Hydrology, 228(1–2), 113–129. https://doi.org/10.1016/S0022-1694(00)00144-X.