Improving Urban Population Distribution Models with Very-High Resolution Satellite Information

Taïs Grippa, Catherine Linard, Moritz Lennert, Stefanos Georganos, Nicholus Mboga, Sabine Vanhuysse, Assane Gadiaga, Eléonore Wolff

Research output: Contribution to journalArticle

3 Downloads (Pure)

Abstract

Built-up layers derived from medium resolution (MR) satellite information have proven their contribution to dasymetric mapping, but suffer from important limitations when working at the intra-urban level, mainly due to their difficulty in capturing the whole range of variation in terms of built-up densities. In this regard, very-high resolution (VHR) remote sensing is known for its ability to better capture small variations in built-up densities and to derive detailed urban land use, which plead in favor of its use when mapping urban populations. In this paper, we compare the added value of various combinations of VHR data sets, compared to a MR one. A top-down dasymetric mapping strategy is applied to reallocate population counts from administrative units into a regular 100 × 100 m grid, according to different weighting layers. These weighting layers are created from MR and/or VHR input data, using simple built-up proportion or reallocation “weights”, obtained from a set of multiple ancillary data used to train a Random Forest regression model. The results reveal that (1) a built-up mask derived from VHR can improve the accuracy of the reallocation by roughly 13%, compared to MR; (2) using VHR land-use information alone results in lower accuracy than using a MR built-up mask; and (3) there is a clear complementarity between VHR land cover and land use.
Original languageEnglish
Article number13
JournalData
Volume4
Issue number1
DOIs
Publication statusPublished - 1 Mar 2019

Fingerprint

population distribution
urban population
land use
complementarity
train
land cover
remote sensing

Keywords

  • African city
  • dasymetric mapping
  • population modelling
  • random forest
  • remote sensing
  • top-down approach
  • very-high resolution data
  • Very-high resolution data
  • Random forest
  • Remote sensing
  • Top-down approach
  • Population modelling
  • Dasymetric mapping

Cite this

Grippa, T., Linard, C., Lennert, M., Georganos, S., Mboga, N., Vanhuysse, S., ... Wolff, E. (2019). Improving Urban Population Distribution Models with Very-High Resolution Satellite Information. Data, 4(1), [13]. https://doi.org/10.3390/data4010013
Grippa, Taïs ; Linard, Catherine ; Lennert, Moritz ; Georganos, Stefanos ; Mboga, Nicholus ; Vanhuysse, Sabine ; Gadiaga, Assane ; Wolff, Eléonore. / Improving Urban Population Distribution Models with Very-High Resolution Satellite Information. In: Data. 2019 ; Vol. 4, No. 1.
@article{8c35ee50a11241c1aa06c4d2c3584e21,
title = "Improving Urban Population Distribution Models with Very-High Resolution Satellite Information",
abstract = "Built-up layers derived from medium resolution (MR) satellite information have proven their contribution to dasymetric mapping, but suffer from important limitations when working at the intra-urban level, mainly due to their difficulty in capturing the whole range of variation in terms of built-up densities. In this regard, very-high resolution (VHR) remote sensing is known for its ability to better capture small variations in built-up densities and to derive detailed urban land use, which plead in favor of its use when mapping urban populations. In this paper, we compare the added value of various combinations of VHR data sets, compared to a MR one. A top-down dasymetric mapping strategy is applied to reallocate population counts from administrative units into a regular 100 × 100 m grid, according to different weighting layers. These weighting layers are created from MR and/or VHR input data, using simple built-up proportion or reallocation “weights”, obtained from a set of multiple ancillary data used to train a Random Forest regression model. The results reveal that (1) a built-up mask derived from VHR can improve the accuracy of the reallocation by roughly 13{\%}, compared to MR; (2) using VHR land-use information alone results in lower accuracy than using a MR built-up mask; and (3) there is a clear complementarity between VHR land cover and land use.",
keywords = "African city, dasymetric mapping, population modelling, random forest, remote sensing, top-down approach, very-high resolution data, Very-high resolution data, Random forest, Remote sensing, Top-down approach, Population modelling, Dasymetric mapping",
author = "Ta{\"i}s Grippa and Catherine Linard and Moritz Lennert and Stefanos Georganos and Nicholus Mboga and Sabine Vanhuysse and Assane Gadiaga and El{\'e}onore Wolff",
year = "2019",
month = "3",
day = "1",
doi = "10.3390/data4010013",
language = "English",
volume = "4",
journal = "Data",
issn = "2306-5729",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "1",

}

Grippa, T, Linard, C, Lennert, M, Georganos, S, Mboga, N, Vanhuysse, S, Gadiaga, A & Wolff, E 2019, 'Improving Urban Population Distribution Models with Very-High Resolution Satellite Information', Data, vol. 4, no. 1, 13. https://doi.org/10.3390/data4010013

Improving Urban Population Distribution Models with Very-High Resolution Satellite Information. / Grippa, Taïs; Linard, Catherine; Lennert, Moritz; Georganos, Stefanos; Mboga, Nicholus; Vanhuysse, Sabine; Gadiaga, Assane; Wolff, Eléonore.

In: Data, Vol. 4, No. 1, 13, 01.03.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Improving Urban Population Distribution Models with Very-High Resolution Satellite Information

AU - Grippa, Taïs

AU - Linard, Catherine

AU - Lennert, Moritz

AU - Georganos, Stefanos

AU - Mboga, Nicholus

AU - Vanhuysse, Sabine

AU - Gadiaga, Assane

AU - Wolff, Eléonore

PY - 2019/3/1

Y1 - 2019/3/1

N2 - Built-up layers derived from medium resolution (MR) satellite information have proven their contribution to dasymetric mapping, but suffer from important limitations when working at the intra-urban level, mainly due to their difficulty in capturing the whole range of variation in terms of built-up densities. In this regard, very-high resolution (VHR) remote sensing is known for its ability to better capture small variations in built-up densities and to derive detailed urban land use, which plead in favor of its use when mapping urban populations. In this paper, we compare the added value of various combinations of VHR data sets, compared to a MR one. A top-down dasymetric mapping strategy is applied to reallocate population counts from administrative units into a regular 100 × 100 m grid, according to different weighting layers. These weighting layers are created from MR and/or VHR input data, using simple built-up proportion or reallocation “weights”, obtained from a set of multiple ancillary data used to train a Random Forest regression model. The results reveal that (1) a built-up mask derived from VHR can improve the accuracy of the reallocation by roughly 13%, compared to MR; (2) using VHR land-use information alone results in lower accuracy than using a MR built-up mask; and (3) there is a clear complementarity between VHR land cover and land use.

AB - Built-up layers derived from medium resolution (MR) satellite information have proven their contribution to dasymetric mapping, but suffer from important limitations when working at the intra-urban level, mainly due to their difficulty in capturing the whole range of variation in terms of built-up densities. In this regard, very-high resolution (VHR) remote sensing is known for its ability to better capture small variations in built-up densities and to derive detailed urban land use, which plead in favor of its use when mapping urban populations. In this paper, we compare the added value of various combinations of VHR data sets, compared to a MR one. A top-down dasymetric mapping strategy is applied to reallocate population counts from administrative units into a regular 100 × 100 m grid, according to different weighting layers. These weighting layers are created from MR and/or VHR input data, using simple built-up proportion or reallocation “weights”, obtained from a set of multiple ancillary data used to train a Random Forest regression model. The results reveal that (1) a built-up mask derived from VHR can improve the accuracy of the reallocation by roughly 13%, compared to MR; (2) using VHR land-use information alone results in lower accuracy than using a MR built-up mask; and (3) there is a clear complementarity between VHR land cover and land use.

KW - African city

KW - dasymetric mapping

KW - population modelling

KW - random forest

KW - remote sensing

KW - top-down approach

KW - very-high resolution data

KW - Very-high resolution data

KW - Random forest

KW - Remote sensing

KW - Top-down approach

KW - Population modelling

KW - Dasymetric mapping

UR - http://www.scopus.com/inward/record.url?scp=85067562591&partnerID=8YFLogxK

U2 - 10.3390/data4010013

DO - 10.3390/data4010013

M3 - Article

VL - 4

JO - Data

JF - Data

SN - 2306-5729

IS - 1

M1 - 13

ER -