# Assessment of Population Exposure to Urban Flood at the Building Scale

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Study Area and Data

#### 2.1. Study Area

#### 2.2. Data Source

- (1)
- Basic geographic data were provided by the Lishui City Surveying and Mapping Center, including digital elevation model (DEM), water system, transportation, building, administrative division, and underground drainage pipe network datasets. DEM data with 5 m resolution were used for terrain analysis. Administrative area and building pattern data assisted in the spatialization of demographic data. Water system and underground pipeline data were used for flood simulations.
- (2)
- POIs were obtained through the API of Baidu Maps, which is one of the most popular electronic maps in China. The total number of POIs in the study area was 11,092. To facilitate data processing, POIs were divided into 10 types including medical facilities, market shopping, restaurants, government agencies, companies, life services, education and training, residential quarters, transportation facilities, and tourist attractions.
- (3)
- The WorldPop dataset (the spatial distribution of population in 2015, China) was selected as the population grid dataset, which is available online (https://www.worldpop.org/geodata/summary?id=5279). Compared with other population grid datasets (GPW, GRUMP, and WorldPop), WorldPop has high accuracy and resolution in China [29] and is the result of a global population distribution mapping project led by the Geographic Data Institute of the University of Southampton in the United Kingdom. It is generated using a random forest algorithm by integrating multi-source data from population censuses, satellites, mobile devices, etc., with a grid resolution of 3 arc (approximately 100 m at the equator). This high spatial resolution means it is widely used in disaster assessment, socio-economics, and other fields. The WorldPop dataset for the study area is shown in Figure 2. From the official website of the Lishui government (http://sq.inlishui.com), we collected census data of 18 communities (25 communities in total). The WorldPop dataset has certain accuracy relative to the census, with an R
^{2}value of 0.7673. However, the RMSE value is 3396.3060, which may be caused by the fact that the floating population is not included in the WorldPop data.

## 3. Methodology

#### 3.1. Mapping the Population at the Building Scale

_{i}; (2) decompose the layers according to POI categories; (3) use Thiessen polygons to split the POI points in each type layer, and assign population data to build patches with overlay analysis; (4) synthesize all layers to obtain the total population within the building region, and evaluate the accuracy of the redistribution results.

#### 3.1.1. Correlation Analysis between POIs and Population Grid

#### 3.1.2. Population Redistribution

_{i}denotes the number of people in the ith building, $Co{r}_{j}$ is the normalization value of ${\rho}_{PO{I}_{i},PD}$. and represents the weight of the jth type POI population distribution, POP

_{ij}is the population of the ith building under the influence of POI type j, $\Delta PO{P}_{jk}$ is the kth no-building Thiessen polygon in the layer of the jth type POI, and the distance (i,k) represents the European distance between building i and polygon k. According to the formula, the population calculation model of a building consists of two parts: initial allocation and supplementary correction. The initial allocation redistributes the Thiessen polygon formed by the gathering points of various POIs according to the proportion of the building area. Complementary corrections are then made to the population distribution based on the initial distribution in proportion to the inverse distance by the principle of geographic proximity. This ensures that the total population is not lost.

Algorithm 1. Population Redistribution | |

Input: POIs, Buildings B | |

Output: $po{p}_{b}$ | |

1: | For each type i |

2: | Create Thiessen Polygon TP_{i} |

3: | Overlay TP_{i} and Buildings B |

4: | For each polygon tp in TP_{i} |

5: | If tp contains Building b |

6: | Then $po{p}_{b}\leftarrow po{p}_{tp}\ast Area\left(b\right)/{\displaystyle \sum}Area\left({b}_{n}\right)$ |

7: | Else $\Delta pop\leftarrow po{p}_{tp}+\Delta pop$ |

8: | For each Building b |

9: | $po{p}_{b}\leftarrow po{p}_{tp}+{w}_{b}\Delta pop$ |

10: | return $po{p}_{tp}$; |

#### 3.1.3. Accuracy Assessment

^{2}and RMSE are often used to evaluate the accuracy of the calculation results [21,27,29,34]. R

^{2}describes the accuracy of the calculation results and RMSE is used to measure the standard error between the predicted population and the actual population. We used these two statistical indicators to evaluate the result of population redistribution by comparing the population of WorldPop grids and the population of buildings in the same block (the study area was manually divided into multiple blocks according to road network data).

#### 3.2. Flood Model

#### 3.3. Scenario Design

## 4. Results and Discussion

#### 4.1. Correlation Analysis between POIs and Population Grid

#### 4.2. Results of Downscaling Gridded Population Mapping

^{2}value between the two is 0.8623 and the RMSE value is 567.6667. The results of the regression analysis show that the building-scale population redistribution results are generally consistent with WorldPop data on a macroscopic basis (Figure 5b). Specifically, the closest is block 94, with a WorldPop population value of 1400 (after rounding) and a reassigned population assessment of 1404, representing a difference of approximately four people. The largest gap occurs in block 76, with a WorldPop population value of 951 (after rounding) and a reassigned population assessment of 382, representing a difference of approximately 569 people (Figure 6). Block 76 has fewer buildings, with an area-to-square-footage ratio of only 0.1766, whereas block 94 has more buildings and is evenly distributed, with a building-to-square-footage ratio of 0.3630. Thus, the method employed in this study has higher accuracy in urban areas where buildings are densely distributed.

#### 4.3. Results of Flooding Simulation

#### 4.4. Assessment of the Affected Population

## 5. Conclusions

## Author Contributions

## Funding

## Acknowledgments

## Conflicts of Interest

## References

- Edenhofer, O. Climate Change 2014: Mitigation of Climate Change; Cambridge University Press: Cambridge, UK, 2015; Volume 3, ISBN 1-107-05821-X. [Google Scholar]
- UNISDR. The Human Cost of Natural Disasters: A Global Perspective; Centre for Research on the Epidemiology of Disaster (CRED): Brussels, Belgium, 2015. [Google Scholar]
- CRED Crunch 58—Disaster Year in Review (2019); CRED: Brussels, Belgium, 2019.
- Hallegatte, S.; Green, C.; Nicholls, R.J.; Corfee-Morlot, J. Future flood losses in major coastal cities. Nat. Clim. Chang.
**2013**, 3, 802–806. [Google Scholar] [CrossRef] - Gain, A.K.; Mojtahed, V.; Biscaro, C.; Balbi, S.; Giupponi, C. An integrated approach of flood risk assessment in the eastern part of Dhaka City. Nat. Hazards
**2015**, 79, 1499–1530. [Google Scholar] [CrossRef] [Green Version] - Dai, Q.; Zhu, X.; Zhuo, L.; Han, D.; Liu, Z.; Zhang, S. A hazard-human coupled model (HazardCM) to assess city dynamic exposure to rainfall-triggered natural hazards. Environ. Model. Softw.
**2020**, 127, 104684. [Google Scholar] [CrossRef] - Freire, S.; Aubrecht, C. Integrating population dynamics into mapping human exposure to seismic hazard. Nat. Hazards Earth Syst. Sci.
**2012**, 12, 3533–3543. [Google Scholar] [CrossRef] - Dell’Acqua, F.; Gamba, P.; Jaiswal, K. Spatial aspects of building and population exposure data and their implications for global earthquake exposure modeling. Nat. Hazards
**2013**, 68, 1291–1309. [Google Scholar] [CrossRef] - Smith, A.; Bates, P.D.; Wing, O.; Sampson, C.; Quinn, N.; Neal, J. New estimates of flood exposure in developing countries using high-resolution population data. Nat. Commun.
**2019**, 10. [Google Scholar] [CrossRef] [Green Version] - Langford, M.; Higgs, G.; Radcliffe, J.; White, S. Urban population distribution models and service accessibility estimation. Comput. Environ. Urban Syst.
**2008**, 32, 66–80. [Google Scholar] [CrossRef] - Wu, C.; Murray, A.T. A cokriging method for estimating population density in urban areas. Comput. Environ. Urban Syst.
**2005**, 29, 558–579. [Google Scholar] [CrossRef] - Batty, M. Fifty Years of Urban Modeling: Macro-Statics to Micro-Dynamics. In The Dynamics of Complex Urban Systems; Physica-Verlag HD: Heidelberg, Germany, 2007; pp. 1–20. [Google Scholar]
- Mennis, J. Generating Surface Models of Population Using Dasymetric Mapping. Prof. Geogr.
**2003**, 55, 31–42. [Google Scholar] [CrossRef] - Deichmann, U.; Street, H.; Balk, D.; Yetman, G. Transforming Population Data for Interdisciplinary Usages: From Census to Grid; Center for International Earth Science Information Network: Washington, DC, USA, 2001. [Google Scholar]
- Balk, D.L.; Deichmann, U.; Yetman, G.; Pozzi, F.; Hay, S.I.; Nelson, A. Determining Global Population Distribution: Methods, Applications and Data. In Advances in Parasitology; Elsevier: Amsterdam, The Netherlands, 2006; Volume 62, pp. 119–156. ISBN 978-0-12-031762-2. [Google Scholar]
- LandScan: A global population database for estimating populations at risk. In Remotely-Sensed Cities; Mesev, V. (Ed.) CRC Press: Boca Raton, FL, USA, 2003; pp. 301–314. ISBN 978-0-429-18116-0. [Google Scholar]
- Gaughan, A.E.; Stevens, F.R.; Linard, C.; Jia, P.; Tatem, A.J. High Resolution Population Distribution Maps for Southeast Asia in 2010 and 2015. PLoS ONE
**2013**, 8, e55882. [Google Scholar] [CrossRef] - Zhu, X.; Dai, Q.; Han, D.; Zhuo, L.; Zhu, S.; Zhang, S. Modeling the high-resolution dynamic exposure to flooding in a city region. Hydrol. Earth Syst. Sci.
**2019**, 23, 3353–3372. [Google Scholar] [CrossRef] [Green Version] - Hossain, M.K.; Meng, Q. A fine-scale spatial analytics of the assessment and mapping of buildings and population at different risk levels of urban flood. Land Use Policy
**2020**, 99, 104829. [Google Scholar] [CrossRef] - Kounadi, O.; Ristea, A.; Leitner, M.; Langford, C. Population at risk: Using areal interpolation and Twitter messages to create population models for burglaries and robberies. Cartogr. Geogr. Inf. Sci.
**2018**, 45, 205–220. [Google Scholar] [CrossRef] [PubMed] - Yao, Y.; Liu, X.; Li, X.; Zhang, J.; Liang, Z.; Mai, K.; Zhang, Y. Mapping fine-scale population distributions at the building level by integrating multisource geospatial big data. Int. J. Geogr. Inf. Sci.
**2017**, 1–25. [Google Scholar] [CrossRef] - Yu, B.; Lian, T.; Huang, Y.; Yao, S.; Ye, X.; Chen, Z.; Yang, C.; Wu, J. Integration of nighttime light remote sensing images and taxi GPS tracking data for population surface enhancement. Int. J. Geogr. Inf. Sci.
**2019**, 33, 687–706. [Google Scholar] [CrossRef] - Song, Y.; Huang, B.; Cai, J.; Chen, B. Dynamic assessments of population exposure to urban greenspace using multi-source big data. Sci. Total Environ.
**2018**, 634, 1315–1325. [Google Scholar] [CrossRef] - Zhao; Li; Zhang; Du Improving the Accuracy of Fine-Grained Population Mapping Using Population-Sensitive POIs. Remote Sens.
**2019**, 11, 2502. [CrossRef] [Green Version] - Dong, P.; Ramesh, S.; Nepali, A. Evaluation of small-area population estimation using LiDAR, Landsat TM and parcel data. Int. J. Remote Sens.
**2010**, 31, 5571–5586. [Google Scholar] [CrossRef] - Silván-Cárdenas, J.L.; Wang, L.; Rogerson, P.; Wu, C.; Feng, T.; Kamphaus, B.D. Assessing fine-spatial-resolution remote sensing for small-area population estimation. Int. J. Remote Sens.
**2010**, 31, 5605–5634. [Google Scholar] [CrossRef] - Bakillah, M.; Liang, S.; Mobasheri, A.; Jokar Arsanjani, J.; Zipf, A. Fine-resolution population mapping using OpenStreetMap points-of-interest. Int. J. Geogr. Inf. Sci.
**2014**, 28, 1940–1963. [Google Scholar] [CrossRef] - de Ruig, L.T.; Haer, T.; de Moel, H.; Botzen, W.J.W.; Aerts, J.C.J.H. A micro-scale cost-benefit analysis of building-level flood risk adaptation measures in Los Angeles. Water Resour. Econ.
**2019**, 100147. [Google Scholar] [CrossRef] - Bai, Z.; Wang, J.; Wang, M.; Gao, M.; Sun, J. Accuracy Assessment of Multi-Source Gridded Population Distribution Datasets in China. Sustainability
**2018**, 10, 1363. [Google Scholar] [CrossRef] [Green Version] - Hu, S.; He, Z.; Wu, L.; Yin, L.; Xu, Y.; Cui, H. A framework for extracting urban functional regions based on multiprototype word embeddings using points-of-interest data. Comput. Environ. Urban Syst.
**2020**, 80, 101442. [Google Scholar] [CrossRef] - Gao, S.; Janowicz, K.; Couclelis, H. Extracting urban functional regions from points of interest and human activities on location-based social networks. Trans. GIS
**2017**, 21, 446–467. [Google Scholar] [CrossRef] - Niu, N.; Liu, X.; Jin, H.; Ye, X.; Liu, Y.; Li, X.; Chen, Y.; Li, S. Integrating multi-source big data to infer building functions. Int. J. Geogr. Inf. Sci.
**2017**, 31, 1871–1890. [Google Scholar] [CrossRef] - Yang, X.; Ye, T.; Zhao, N.; Chen, Q.; Yue, W.; Qi, J.; Zeng, B.; Jia, P. Population Mapping with Multisensor Remote Sensing Images and Point-Of-Interest Data. Remote Sens.
**2019**, 11, 574. [Google Scholar] [CrossRef] [Green Version] - Chen, Y.; Zhang, R.; Ge, Y.; Jin, Y.; Xia, Z. Downscaling Census Data for Gridded Population Mapping With Geographically Weighted Area-to-Point Regression Kriging. IEEE Access
**2019**, 7, 149132–149141. [Google Scholar] [CrossRef] - Bates, P.D.; De Roo, A.P.J. A simple raster-based model for flood inundation simulation. J. Hydrol.
**2000**, 236, 54–77. [Google Scholar] [CrossRef] - Horritt, M.S.; Bates, P.D. Evaluation of 1D and 2D numerical models for predicting river flood inundation. J. Hydrol.
**2002**, 268, 87–99. [Google Scholar] [CrossRef] - Wood, M.; Hostache, R.; Neal, J.; Wagener, T.; Giustarini, L.; Chini, M.; Corato, G.; Matgen, P.; Bates, P. Calibration of channel depth and friction parameters in the LISFLOOD-FP hydraulic model using medium-resolution SAR data and identifiability techniques. Hydrol. Earth Syst. Sci.
**2016**, 20, 4983–4997. [Google Scholar] [CrossRef] [Green Version] - Yang, Q.; Dai, Q.; Han, D.; Zhu, X.; Zhang, S. Impact of the storm sewer network complexity on flood simulations according to the stroke scaling method. Water
**2018**, 10, 645. [Google Scholar] [CrossRef] [Green Version]

**Figure 4.**Map of population distribution at the building scale. Map of population distribution at the building scale: (

**a**) population distribution in block A; (

**b**) population distribution in block B; (

**c**) population distribution in block C; (

**d**) population distribution in block D.

**Figure 5.**Accuracy evaluation: (

**a**) divided assessment unit; (

**b**) scatterplot between WorldPop and redistribution results.

**Figure 6.**Partial comparison of redistributed population and WorldPop estimation results: (

**a**) population distribution at block 76 at the building scale; (

**b**) WorldPop population distribution in block 76; (

**c**) population distribution at the building scale in block 94; (

**d**) WorldPop population distribution in block 94.

**Figure 7.**Flooding simulation results under different rainfall return periods (5, 10, 20, 50, and 100 yr).

**Figure 8.**Comparison of the number of people affected by urban flooding in multiple scenarios (5, 10, 20, 50, and 100 yr return periods).

**Figure 10.**Underestimation of the affected population: (

**a**) using WorldPop grid to assess the affected population and (

**b**) using the redistributed building population.

ID | Type | Quantity | Correlation | Weight |
---|---|---|---|---|

1 | Market shopping | 2220 | 0.7494 | 0.1245 |

2 | Restaurant | 2801 | 0.7010 | 0.1165 |

3 | Government institution | 584 | 0.6662 | 0.1107 |

4 | Domestic service | 2562 | 0.6056 | 0.1006 |

5 | Medical facility | 214 | 0.5892 | 0.0979 |

6 | Residential area | 1163 | 0.5762 | 0.0957 |

7 | Company | 462 | 0.5717 | 0.0950 |

8 | Tourist attraction | 422 | 0.5416 | 0.0900 |

9 | Education and training | 358 | 0.5354 | 0.0890 |

10 | Transportation facility | 306 | 0.4827 | 0.0802 |

Mild | Moderate | Serious | Critical | Total | |
---|---|---|---|---|---|

5-yr | 0.2998 | 0.1154 | 0.0296 | 0.0008 | 0.4456 |

10-yr | 0.5844 | 0.2632 | 0.0745 | 0.0043 | 0.9264 |

20-yr | 0.8450 | 0.4258 | 0.1065 | 0.0278 | 1.4051 |

50-yr | 0.9854 | 0.5595 | 0.1642 | 0.0560 | 1.7654 |

100-yr | 1.1457 | 0.6909 | 0.2342 | 0.0868 | 2.1575 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Zhu, S.; Dai, Q.; Zhao, B.; Shao, J.
Assessment of Population Exposure to Urban Flood at the Building Scale. *Water* **2020**, *12*, 3253.
https://doi.org/10.3390/w12113253

**AMA Style**

Zhu S, Dai Q, Zhao B, Shao J.
Assessment of Population Exposure to Urban Flood at the Building Scale. *Water*. 2020; 12(11):3253.
https://doi.org/10.3390/w12113253

**Chicago/Turabian Style**

Zhu, Shaonan, Qiang Dai, Binru Zhao, and Jiaqi Shao.
2020. "Assessment of Population Exposure to Urban Flood at the Building Scale" *Water* 12, no. 11: 3253.
https://doi.org/10.3390/w12113253