Quantifying Urban Flood Extent Using Satellite Imagery and Random Forest: A Case Study in Southeastern Pennsylvania

doi:10.21203/rs.3.rs-3882712/v1

Download PDF

Research Article

Quantifying Urban Flood Extent Using Satellite Imagery and Random Forest: A Case Study in Southeastern Pennsylvania

https://doi.org/10.21203/rs.3.rs-3882712/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 27 Jul, 2024

Read the published version in Natural Hazards →

You are reading this latest preprint version

The risk of floods from tropical storms is increasing due to climate change and human development. Maps of past flood extents can aid in planning and mitigation efforts to decrease flood risk. In 2021, Hurricane Ida slowed over the Mid-Atlantic and Northeast United States and released unprecedented rainfall. Satellite imagery and the Random Forest algorithm are a reliable combination to map flood extents. However, this combination is not usually applied to urban areas. We used Sentinel-2 imagery (10 m), along with derived indices, elevation, and land cover data, as inputs to a Random Forest model to make a new flood extent for southeastern Pennsylvania. The model was trained and validated with a dataset created with input from PlanetScope imagery (3 m) and social media posts related to the flood event. The overall accuracy of the model is 99%, and the flood class had a user’s and producer’s accuracy each over 99%. We then compared the flood extent to the Federal Emergency Management Agency (FEMA) flood zones at the county and tract level and found that more flooding occurred in the Minimal Hazard zone than in the 500-year flood zone. Our Random Forest model relies on publicly available data and software to efficiently and accurately make a flood extent map that can be deployed to other urban areas. Flood extent maps like the one developed here can help decision-makers focus efforts on recovery and resilience.

Machine Learning

Emergency Management

Flooding

Surface Water

Hurricane Ida

Climate Change

1.1 Floods Cause Harm

Flooding has become a devastating and destructive hazard due to human development in high-risk areas (CRED, 2015) and climate change. The World Bank estimated in 2022 that 1.81 billion people, or 23% of the world’s population, are at risk of intense floods (Rentschler et al., 2022). Floods cause many types of harm (de Bruijn et al., 2019; Rosser et al., 2017), including economic losses (Pinos & Quesada-Román, 2022), damages to private homes/assets, damages to public infrastructure (Goffi et al., 2020), involuntary displacement, impacts on mental health (Markhvida et al., 2020), and disruptions to daily life and traffic flow (Hosseiny et al., 2020). In the most dangerous circumstances, floods can lead to loss of life (Goffi et al., 2020), taking 146 lives in 2021 in the U.S. (US Department of Commerce, 2021). In the U.S., floods cause major economic losses, with an average yearly cost of $4.5B (Smith, 2023). Among flooding, the average cost of each tropical storm, including hurricanes, is $22.2B (which includes damages caused by floods and winds) (Smith, 2023). The economic and social costs of floods are high, and the risk of floods from hurricanes is expected to increase due in part to trends in climate change (Trenberth et al., 2018) and human development.

1.2 Hurricanes risk increasing

Climate change is heating the atmosphere, allowing it to hold more moisture and increasing precipitation frequency and intensity (Van Oldenborgh et al., 2017), leading to higher flood risk (Ireland et al., 2015; Van Oldenborgh et al., 2017). In addition to more extreme precipitation, the U.S. is facing more intense hurricanes (Kossin et al., 2007). At the global scale, climate change is causing the speed of tropical cyclones to decrease and precipitation rates to increase (Kossin, 2018). In the U.S., this same trend applies with tropical storms stalling more often along the coast and increasing precipitation rates (Hall & Kossin, 2019). The intensity (Holland & Bruyère, 2014; Knutson et al., 2015), storm surge (Lin et al., 2012), and precipitation of these hurricane events are also expected to worsen due to climate change.

1.3 Unequal distribution of risk

People in poverty are disproportionately at risk of floods around the globe (Garbutt et al., 2015; Kawasaki et al., 2020; Mtapuri et al., 2018; Winsemius et al., 2018). This trend signifies that flood risk is not equally distributed (Wing et al., 2022). This imbalance occurs globally because people in poverty are more likely to live in a floodplain due to the concentration of jobs and transportation (Mtapuri et al., 2018). In the U.S., the same trend of inequitable flood risk applies, where flood risk disproportionately impacts poorer communities (Wing et al., 2022). At the city level, one case study of Los Angeles found that poorer communities have disproportionately higher flood risk, but this trend varied by flood type (Sanders et al., 2022). These findings at the national and local scale show that it is important to investigate the distribution of flood impacts to help inform emergency response and recovery.

1.4 Hurricane Ida 2021

Hurricane Ida was a category 4 hurricane that landed in the U.S. on August 29, 2021, bringing catastrophic damage (Beven II et al., 2022). In the first days of September, Hurricane Ida stalled, becoming extratropical and bringing heavy rains with rates around 3 inches per hour to states in the Mid-Atlantic and Northeast (Beven II et al., 2022). The storm caused dozens of fatalities and damaged homes, businesses, vehicles and infrastructure (Smith, 2023). The National Oceanic and Atmospheric Association (NOAA) National Centers for Environmental Information (NCEI) estimates that the cost of Hurricane Ida is $80.2 Billion (Consumer Price Index-Adjusted) (Smith, 2023), the costliest hazard of 2021.

In Pennsylvania, Hurricane Ida brought precipitation and floods that caused damage (Beven II et al., 2022). In the aftermath of Hurricane Ida, individuals and households in Pennsylvania received $124M in funding from the Federal Emergency Management Agency (FEMA) to cover damages (Cooper et al., 2022). Even with this influx in funding, many counties in the state are still recovering from the damages of Hurricane Ida, including Philadelphia, Montgomery, Delaware and Bucks counties (Cooper et al., 2022). Given the trends in intensifying hurricanes, it is imperative to plan for future events using insights from past flood events (Brandt et al., 2021).

1.5 Flood extent from satellite imagery

Satellite imagery is a reliable data source (Hermas et al., 2021) that is regularly used to map surface water and flood dynamics across various scales (Ayanu et al., 2012; Jones, 2019; Pekel et al., 2016; Tulbure et al., 2016). Machine learning is an effective method for classifying floods in satellite imagery (Tulbure et al., 2022). It has higher classification accuracy than parametric strategies (Maxwell et al., 2018), such as with a single water index (Goffi et al., 2020).

Supervised machine learning classification of satellite imagery relies on accurate training and validation data (Olofsson et al., 2014). A known flood extent from a reliable dataset (Hondula et al., 2021) or created from aerial photography (Rosser et al., 2017; Schnebele et al., 2014) is ideal training data. However, it is not available for every flood event. In the case of Hurricane Ida, aerial imagery was not collected in the study area of southeastern Pennsylvania (US Department of Commerce, 2022). Since an accurate flood extent was unavailable as training and validation data, we created one from satellite imagery and social media (Ireland et al., 2015; Perin et al., 2022).

Urban environments have complex waterways, shallow and ephemeral flooding, and ponding, meaning the flood extent is discontinuous (Tanim et al., 2022; Woznicki et al., 2019). Flood models can generate flood maps in urban areas (Knighton et al., 2021; Liu et al., 2015), however, flood maps generated from satellite imagery are considered more aligned with ground conditions. Synthetic Aperture Radar (SAR) data can collect data through clouds, but is limited in urban areas due to its side-looking nature (Mason et al., 2014) and can have gaps due to tall buildings causing radar shadowing or layover (Clement et al., 2018). Optical imagery cannot permeate through clouds, but does not face the same challenges with tall buildings. For our study area after Hurricane Ida, there was optical imagery, Sentinel-2, collected, but no freely available SAR data collected. In this study we used optical data and combined it with several datasets in our machine learning model to address the complexity of urban environments.

1.6 Gaps and objectives

There is currently no Hurricane Ida flood extent in Pennsylvania, including the city of Philadelphia and surrounding counties. The floods resulting from Hurricane Ida dissipated slowly and coincidentally overlapped with Sentinel-2 imagery collection, providing a unique opportunity for testing flood detection methods in an urban environment. Previous research has predominantly used satellite imagery and machine learning to detect floods using a vetted flood extent (from satellite imagery or a flood model) and rarely applies these methods to an urban area. Therefore we focused on the urban area of Philadelphia and surrounding counties after Hurricane Ida, an event without a vetted flood extent available, to fill these gaps.

The objectives of this research were to: (1) combine Sentinel-2 imagery and other data in a Random Forest (RF) machine learning algorithm to create a novel flood extent in southeastern Pennsylvania after Hurricane Ida; (2) compare the flood extent to FEMA’s (Federal Emergency Management Agency) flood zones; (3) use the flood extent to calculate flood exposure and determine the equality of its distribution. Since hurricanes and floods are expected to increase, methods for accurate and timely flood extent maps in urban areas are fundamental for improving recovery and mitigation efforts.

2.1 Study site

Philadelphia, located in southeastern Pennsylvania, is the sixth largest city by population in the U.S., with over 1.5 million people (Census, 2021). Over the past hundreds of years, the city has grown and developed, building over the existing streams. Floods occur in Philadelphia 12 days per year, and this frequency is expected to increase (Sweet et al., 2019). While the city is regularly flooded, it has made considerable efforts to reduce floods through multiple avenues, including its stormwater management program and strict design requirements (Hosseiny et al., 2020).

On September 1, 2021, Philadelphia was directly in the path of Hurricane Ida (Fig. 1) and received 2.37 inches of precipitation, significantly more than its annual monthly mean precipitation of 0.12 inches (NOAA, 2021). The impacts of Hurricane Ida in Philadelphia and the resulting floods were wide-ranging from disrupting daily life to damaging personal assets (Pulcinella et al., 2021). In addition to Philadelphia, surrounding less urban counties, including Delaware, Bucks, and Montgomery counties, were impacted by Hurricane Ida flooding and incorporated into our study area (Cooper et al., 2022). According to the U.S. 2020 Census, Philadelphia (both the name of the city and county) is the most urban at 100%, then Delaware and Montgomery counties at 88% and 76%, respectively, and lastly, Bucks County at 44% (U.S. Census Bureau, 2020).

2.2 Datasets

2.2.1 Satellite Imagery

We selected Sentinel-2 imagery because it aligned with peak Hurricane Ida flooding in our study area and was the highest spatial resolution of publicly available imagery. Sentinel-2 is a mission run by the European Space Agency and produces publicly available satellite imagery of the globe at a 10–60 m spatial resolution and a temporal resolution of ~ 5 days. We used Sentinel-2 imagery that was collected a day after Hurricane Ida passed through the study area (September 2, 2021) with little (< 1%) cloud cover, making it an optimal data source. In Google Earth Engine (GEE), we obtained the imagery, filtered it temporally and spatially, and removed cloud and cirrus pixels using the quality assessment band (QA60) (Tiwari et al., 2024).

Other imagery that was considered included Synthetic Aperture Radar (SAR) data, PlanetScope imagery and aerial imagery. The Sentinel-1 imagery dates did not align with the peak flooding in the study area. We considered PlanetScope imagery as the basis for the flood extent but instead used it to create the training and validation data for the RF model since it is best practice to use a higher resolution image for training and validation data than the model input (Olofsson et al., 2014). We did not consider aerial imagery because it was not collected by the National Geodetic Survey after Hurricane Ida in the study area (US Department of Commerce, 2022).

2.2.2 Sentinel-2 Indices

The RF inputs included all Sentinel-2 surface reflectance bands, two vegetation indices and six water indices, all previously shown to be important when mapping floods with satellite data (Goffi et al., 2020; Tulbure et al., 2016, 2022) (Table 1). We produced all the indices in GEE. The vegetation indices are used to help classify NotWater pixels by identifying areas of vegetation. We used several water indices, each with different strengths, to help categorize Water (permanent) and Flood pixels in the model.

The Normalized Difference Water Index (NDWI) is the standard for classifying water using green and near-infrared (NIR) bands (McFeeters, 1996). The Modified Normalized Difference Water Index (MNDWI) is a variation of NDWI that uses shortwave infrared (SWIR) instead of NIR and is more suitable in built-up areas than NDWI (Xu, 2006). The Automated Water Extraction Index (AWEI) uses five spectral bands to improve water classification by decreasing the environmental noise of shadows and dark surfaces (Feyisa et al., 2014). Two variations of the AWEI formulas (AWEI_nsh and AWEI_sh) have different effectiveness in urban areas. The AWEI_nsh formula is more equipped for urban areas because it effectively eliminates built surfaces. The AWEI_sh formula is more equipped for filtering out shadows, but is less equipped for urban areas because it tends to misclassify reflective roofs as water.

We also used Linear Spectral Unmixing (LSU) to produce three inputs, each with the percent of three different “endmembers” (water, urban and vegetation) or classes for each pixel. Pixels have mixed spectral signatures because the underlying land cover is mixed and highly variable (C. Yang et al., 2007). LSU addresses this heterogeneity by using all bands to estimate each pixel's “endmember” percent (C. Yang et al., 2007). LSU is helpful in the context of floods because it can be used to determine the fraction of water in each pixel and produce flood maps (Bangira et al., 2017; Gómez-Palacios et al., 2017).

Table 1

Sentinel-2 derived indices used in our RF classification.
Category	Index Name	Formula	Reference
Vegetation	Normalized Difference Vegetation Index (NDVI)	NDVI = (NIR - RED / NIR + RED)	Kriegler et al., 1969
Vegetation	Enhanced Vegetation Index (EVI)	EVI = 2.5 * ( NIR - RED) / (NIR + 6 * RED − 7.5 * BLUE + 1)	Huete, 1997
Water	Normalized Difference Water Index (NDWI)	NDWI = (GREEN - NIR) / (GREEN + NIR)	McFeeters, 1996
	Modified Normalized Difference Water Index (MNDWI)	MNDWI = (GREEN - MIR) / (GREEN + MIR)	Xu, 2006
	Automated Water Extraction Index (AWEI_nsh )	AWEI_nsh = 4 * (GREEN - SWIR1) - (0.25 * NIR + 2.75 * SWIR2)	Feyisa et al., 2014
	Automated Water Extraction Index (AWEI_sh )	AWEI_nsh = BLUE + 2.5 * GREEN − 1.5 * (NIR + SWIR1) = 0.25 * SWIR2	Feyisa et al., 2014
	Water Ratio Index (WRI)	WRI = (GREEN + RED) / (NIR + SWIR1)	Shen & Li, 2010
	Normalized Difference Moisture Index (NDMI)	NDMI = (NIR - SWIR1) / (NIR + SWIR1)	Gao, 1996
Flood	Normalized Difference Flow Index (NDFI)	NDFI = (RED - SWIR2) / (RED + SWIR2)	Boschetti et al., 2014
Multiple	Fractional Coefficients (water, urban and vegetation)	R_mix = (F_water * R_water) + (F_urban * R_urban) + (F_vegetation * R_vegetation)	Settle & Drake, 1993

2.2.3 Additional Inputs

In addition to Sentinel-2 surface reflectance bands and derived indices, several other datasets readily available in GEE were incorporated into the RF model (Table 2). A digital elevation model (DEM) can be used to derive data (e.g., slope) that influences where floods occur (Tulbure et al., 2016). In our model, we used the United States Geological Survey (USGS) 3DEP 10 m National Map (U.S. Geological Survey, 2023) in GEE to calculate slope, aspect, and hillshade. We also used the USGS National Land Cover Database at 30 m resolution in the model resampled to 10 m, because land cover and impervious surface contribute to flood extent (Apel et al., 2016; Blum et al., 2020).

In GEE we also incorporated datasets of 30 m resolution from the European Commission’s Joint Research Centre (JRC), including surface water occurrence and surface water classification (water, seasonal, permanent) (Pekel et al., 2016). We downscaled the JRC datasets and the USGS National Land Cover Database in GEE using the resample function and bilinear method to match the other RF inputs at 10 m resolution. In GEE, we combined the non-Sentinel-2 inputs and reprojected them to match the Sentinel-2 inputs. Next, all the inputs were combined, and a 3x3 window was created for each band.

Table 2

The RF model used 62 features derived from satellite imagery, digital elevation, land cover and water datasets that were preprocessed in Google Earth Engine (GEE).
Category	Source	Year	Original Resolution	Name
Reflectance, Vegetation, Water, Moisture, Linear Spectral Unmixing (LSU)	Sentinel-2	2021	10–60 m	12 Reflectance Bands
				9 Derived Indices
				3 Fractional coefficients (water, urban, vegetation)
DEM	USGS 3DEP 10m National Map Seamless (1/3 Arc-Second)	2017	10 m	Slope
				Aspect
				Hillshade
Land Cover	USGS National Land Cover Database (NLCD)	2019	30 m	Landcover
Land Cover	USGS National Land Cover Database (NLCD)	2019	30 m	Impervious
Water	European Commission Joint Research Centre (JRC)	2021	30 m	Global Surface Water - Occurrence
Water	European Commission Joint Research Centre (JRC)	2021	30 m	Yearly Water Classification History
Window	All above rows	Varies	10–60 m	3x3 Window

2.2.4 Training and Validation Data

In QGIS, we created the training and validation dataset by hand using several datasets (Fig. 2) (Perin et al., 2022). We used PlanetScope (3 m) imagery, which is higher resolution than our model inputs (Maxwell et al., 2018; Olofsson et al., 2014), and the National Hydrography Dataset (NHD), as reference to create Flood and Water polygons at the resolution of Sentinel-2 (10 m) imagery (Ireland et al., 2015). We also used social media posts from the Global Food Monitor project, a publicly available database of flood-related tweets (de Bruijn et al., 2019). We used ~ 3,000 tweets, including text and photos, and ~ 140 unique points to guide the creation of Flood polygons (Akhtar et al., 2021; Schnebele et al., 2014).

The Google basemaps in QGIS provided high-resolution imagery for drawing NotWater polygons (Perin et al., 2022). Random points in each county were used to guide the location of NotWater polygons. Before drawing the NotWater polygons, we ensured that the polygon was outside the NHD and that the basemap imagery was clear of pools and other surface water. For all three classes (NotWater, Water, Flood), every layer of data was checked to verify the accuracy of the polygon class. The training and validation dataset was 433 polygons consisting of 164 NotWater, 131 Water and 138 Flood polygons.

Next, we uploaded the training and validation dataset into GEE and converted it into 10 m pixels. We randomly oversampled the rarer classes of Water and Flood and used all NotWater polygons (Maxwell et al., 2018; Olofsson et al., 2014). Then, we split the dataset into two stratified random samples, with 70% for training and 30% for validation (Table 3).

Table 3

Training and validation dataset (pixels 10 m) consists of three classes, oversampling the rarer Water and Flood classes, split into two stratified random samples with ~ 48,000 pixels to train and ~ 20,000 pixels to validate the RF model.
Type	Percent	Total Pixels	Pixels
Type	Percent	Total Pixels	NotWater	Water	Flood
Training	70%	48,352	29,473	14,721	4,158
Validation	30%	20,648	12,527	6,279	1,842
Total	100%	69,000	42,000	21,000	6,000

2.2.5 FEMA Flood Zones

The National Flood Hazard Layer (NFHL) is a database of flood zones and flood insurance requirements maintained by FEMA in support of the National Flood Insurance Program (NFIP) (FEMA, 2023). The NFHL is available for the entire study area, therefore, every flooded pixel will occur in a FEMA flood zone. The flood zones we focus on in this study are the 100-yr (100-year), 500-yr (500-year), and Minimal Hazard zones because they are the primary risk classifications. The 100-yr and 500-yr flood zones have a 1% and 0.2% likelihood of flooding yearly (FEMA, 2020). We combined all other FEMA flood zones (floodway, 1% annual chance flood hazard contained in channel, area with reduced flood risk due to levee, and 1% depth less than 1 foot) into an “Other” category that encapsulates areas that are less common. The Minimal Hazard zone is outside the 500-yr flood zone and at higher elevations. Once we created the Hurricane Ida flood extent, we used the FEMA flood zones to determine the area and percent area of the Hurricane Ida flood that occurred in the different zones at the county and tract level.

2.2.6 Vulnerability and Demographic Datasets

The Centers for Disease Control and Prevention (CDC) and Agency for Toxic Substances and Disease Registry (ATSDR) Social Vulnerability Index (SVI, hereafter) is a vulnerability index at the tract and county level in the United States. Vulnerability is a community’s ability to prevent suffering and financial loss due to a disaster (Fielding, 2018). The CDC’s SVI dataset estimates an overall vulnerability score using four themes (socioeconomic status, household characteristics, racial & ethnic minority status, and housing type & transportation) (CDC, 2020). From the CDC’s SVI 2020 dataset, we used the sum of the four themes as the SVI index and the number of people below the 150% federal poverty level. The poverty data in the CDC’s SVI dataset came from the 2016–2020 American Community Survey (ACS). These two attributes were used to determine if Hurricane Ida floods disproportionately impacted people in poverty and vulnerable populations.

2.3 Random Forest Model

The RF machine learning algorithm is an ensemble classifier that uses a large number of decision trees that each use different random samples and a subset of features to assign a class, then the majority vote of all the trees classifies the data (Breiman, 2001; Maxwell et al., 2018). The RF algorithm effectively classifies surface water (Phan et al., 2020; Tulbure & Broich, 2013) and floods (Tulbure et al., 2016, 2022). While the algorithm is a ‘black box’, meaning you cannot visualize all trees, it still has higher accuracy than parametric techniques, and it has the added benefit of being robust to smaller and lower quality training datasets (Maxwell et al., 2018).

We executed the RF algorithm in GEE with the steps outlined in Fig. 3. We used the ee.Classifier.smileRandomForest function in GEE to train the model on the training data, then classify the entire study area and determine feature importance. The number of pixels used to train and validate are outlined in Table 3. We used the validation data to create a confusion matrix. After the first run of the algorithm, we ran the model with the Sentinel-2 features and indices and different combinations of additional features and parameters in order to select the features and parameters (number of trees, number of features at each split) that produce the highest overall accuracy. The final parameters chosen for the algorithm were the default numbers, 100 trees and eight features per split. These parameters align with those chosen in other research using RF classification in GEE (Phan et al., 2020). After these parameters were selected, the algorithm was run again with the optimal parameters and all the features.

2.4 Assessment of flood impact

After we created a flood extent with our RF model, we exported it from GEE and converted it from a raster to a vector file to compare to the vector file of FEMA flood zones. Then, the flood polygons were clipped to different FEMA flood zones. In this study, we calculated the area and percent flood in four different FEMA flood zones: 100-yr, 500-yr, Minimal Hazards, and Other. The Other category consists of areas, including floodways, that are likely to be flooded; therefore, we expected most flooding to occur in the 100-yr and Other zone. After the flood extent was clipped to the different zones, the area (km²) and the percent of the flood that occurred in each FEMA flood zone were calculated at the county and census tract level.

We assessed the flood exposure equality by plotting Lorenz Curves and calculating the Gini index. The Gini index is typically used to study income inequality (Lorenz, 1905), but can also be applied to studying flood exposure inequalities (Sanders et al., 2022). The Gini index ranges from − 1 to 1, 0 representing perfect equality. To determine the equality of flooding, we first merged the CDC’s SVI data and flood data by tract, then calculated flood exposure per tract. In this instance, the flood exposure is the tract population multiplied by the percent of the tract flooded. To measure equality in the different FEMA flood zones, flood exposure is the tract population multiplied by the percent of a given FEMA zone, then multiplied by the percent of the zone flooded. Then, we sorted the table by desired variables (population below the poverty line, SVI score) and plotted the Lorenz Curve with the cumulative percent of flood exposure on the y axis and the cumulative percent of population on the x axis. Then, we calculated the Gini index to determine if there was a disproportionate impact on people in poverty and vulnerable populations, and if this impact varied by flood zone.

We quantified the flood extent after Hurricane Ida in southeastern Pennsylvania using Sentinel-2 satellite imagery, derived indices, linear unmixing, land cover and surface water data in a RF model trained with polygons of three classes: NotWater, Water and Flood. The training data used PlanetScope imagery and incorporated crowdsourced social media and permanent water data. When creating a flood extent, this approach proved highly accurate (> 99% overall accuracy). The resulting flood extent compared to the FEMA flood zones also reveals that, in this event, there was more than double the area of floods in the Minimal Hazard zones than in the 500-yr flood zone.

3.1 Flood Extent Generated with Random Forest Model

Our methods produced a new flood extent map after Hurricane Ida of three classes: NotWater, Water and Flood for the study area of four counties in southeastern Pennsylvania, including the urban area of Philadelphia County, where no prior flood extent existed. The result is a map of flood extent for September 2, 2021, a day after Hurricane Ida passed through the study area and the day the Sentinel-2 mission captured data (Fig. 4). When zooming into different land uses in the study area, the RF model Flood classification visually aligns with Sentinel-2 false color imagery (Fig. 5).

We created a confusion matrix for our RF Model using the validation dataset. The overall accuracy was 99.86%, and for the Flood class, the producer’s and the user’s accuracy were over 99% (Table 4).

Table 4

Confusion Matrix (pixel count) of Random Forest (RF) model with an overall accuracy of 99.87%.
		Reference Data
		NotWater	Water	Flood	Total	User’s accuracy (%)
Classified Data	NotWater	12,527	0	0	12,527	> 99%
	Water	4	6,261	14	6,279	> 99%
	Flood	6	3	1,833	1,842	> 99%
	Total	12,537	6,264	1,847	20,648
	Producer’s accuracy (%)	> 99%	> 99%	> 99%

We found the relative feature importance using the explain function in GEE (Table 5). The most important feature for classification came from the Sentinel-2 Band 1, aerosols 3x3 window and aerosols; the next important feature was slope. The most important water index was NDFI, although in previous runs, it was MNDWI 3x3 window. The 3x3 window of the features tended to have lower importance in our RF model than the regular features. Consistently, the least important features were Water and Water 3x3 window from the JRC dataset.

Table 5

Relative feature importance for the RF model.
Top Features		Middle Features		Lower Features
Aerosols	2.74	NDMI	1.76	MNDWI	1.50
Aerosols (3x3)	2.35	NDVI (3x3)	1.75	JRC Water Occurrence	1.47
Slope	2.25	Blue (3x3)	1.73	WRI	1.46
Vegetation fraction (3x3)	2.19	Red Edge 3	1.71	Impervious	1.46
Red Edge 1	2.17	Red Edge 2 (3x3)	1.65	Aspect (3x3)	1.45
Landcover (3x3)	2.10	Blue	1.65	EVI	1.44
SWIR 2	1.99	NDWI	1.64	Aspect	1.43
Green	1.98	NIR	1.64	AWEI_sh (3x3)	1.37
Landcover	1.95	JRC Water Occurrence (3x3)	1.60	NDMI (3x3)	1.36
SWIR 2 (3x3)	1.91	NDFI (3x3)	1.58	EVI (3x3)	1.30
NDFI	1.91	AWEI_nsh	1.58	SWIR 1 (3x3)	1.28
Red (3x3)	1.90	SWIR 1	1.55	Impervious (3x3)	1.17
Water vapor (3x3)	1.90	Water fraction (3x3)	1.54	NIR (3x3)	1.16
NDVI	1.84	AWEI_nsh (3x3)	1.54	Hillshade	1.14
Vegetation fraction	1.83	WRI (3x3)	1.52	Red Edge 4 (3x3)	1.12
Red Edge 1 (3x3)	1.82	NDWI (3x3)	1.52	Red Edge 4	1.07
Water vapor	1.81	Urban fraction	1.52	Hillshade (3x3)	1.02
Green (3x3)	1.80	Red Edge 2	1.51	JRC Water (3x3)	0.92
Water fraction	1.79	AWEI_sh	1.51	Slope (3x3)	0.82
Urban fraction (3x3)	1.79	MNDWI (3x3)	1.50	JRC Water	0.81
Red	1.76	Red Edge 3 (3x3)	1.50
*The (3x3) signifies it is the mean value of the variable created with a 3 by 3 window

3.2 Comparison to FEMA Flood zones

After we created a flood extent for Hurricane Ida using our RF model, we calculated the area and percent of floods that occurred in the different FEMA flood zones (Table 6). The FEMA flood zones we investigated were the 100-yr, 500-yr, Minimal Hazard zones and Other (a combination of rarer zones). All counties experienced less than half a percent of the total county area being flooded. Bucks County had the most flooding by area, while Philadelphia and Bucks counties had the highest percentage of the area flooded, with 0.51% and 0.43%, respectively.

Table 6

The total classified flood, both area (km²) and percent, per county in the study area and within FEMA flood zones. Every area calculation comes with a margin of error of at least +/-1%, given the uncertainty in the RF model.
County	Total Floods		Floods within FEMA Flood Zones
	Total Floods		Other		100-yr		500-yr		Minimal Hazard
	%	km²	%	km²	%	km²	%	km²	%	km²
Bucks	0.43	6.94	48.0	3.33	34.7	2.41	2.4	0.17	14.6	1.01
Delaware	0.20	0.98	30.3	0.30	58.1	0.57	0.9	0.01	7.8	0.08
Montgomery	0.41	5.15	51.5	2.65	40.1	2.06	1.5	0.08	6.6	0.34
Philadelphia	0.51	1.89	44.9	0.85	51.5	0.97	0.4	0.01	3.2	0.06

Most flooding occurred in the 100-yr and Other FEMA flood zones. All counties received some amount of flooding in Minimal Hazard zones and more flooding by area and percent in the Minimal Hazard zone than the 500-yr zone. In Bucks County, ~ 14% +/- 1.4% of total floods occurred in the Minimal Hazard zone. The percentage of FEMA zones flooded told a slightly different story with the Other zone receiving the highest proportion of flooding, then the 100-yr, then 500-yr, then the Minimal Hazard zone (Table 7).

Table 7

Percentage of FEMA flood zones flooded per county.
County	FEMA Flood Zones % Flooded
County	Other	100-yr	500-yr	Minimal Hazard
Bucks	8.23	2.18	0.58	0.58
Delaware	3.53	1.27	0.10	0.10
Montgomery	7.88	3.68	0.16	0.16
Philadelphia	10.24	2.36	0.05	0.05

Tract-level flood information can reveal more local patterns in flooding. Figure 6 shows the percentage of floods that occurred in a subset of FEMA flood zones (100-yr, 500-yr, Minimal Hazard). The tract-level shows a similar pattern to the county information, showing that the highest percent of floods occurs in the 100-yr and Minimal Hazard zones and the smallest percent of floods occurs in the 500-yr zone.

3.3 Vulnerability and Poverty Distribution

In addition to investigating the distribution of Hurricane Ida floods across FEMA flood zones, we also investigated the distribution of floods across tract poverty and SVI scores. We found that flood exposure did not disproportionately impact people below the poverty line or vulnerable populations (higher SVI scores) (Fig. 7). People below the poverty line were underexposed to floods overall (G = -0.205), as well as in FEMA’s Other (G = -0.218) and Minimal Hazard zone (G = -0.363). Vulnerable populations (i.e. higher SVI scores) were underexposed to floods overall (G = -0.274), as well as in FEMA’s Other (G = -0.287), 100-yr (G = -0.248) and Minimal Hazard zones (G = -0.389). For both variables studied, there was a weak underexposure to floods for people below the poverty line and vulnerable populations in FEMA’s 500-yr zones (-0.2 < G < -0.1).

We developed reproducible methods for creating a flood extent after a major flood event in an urban area using the RF algorithm and mostly freely available data and software. While the RF algorithm and satellite imagery has been previously used to detect floods (Schaffer-Smith et al., 2020; Tulbure et al., 2018, 2022), these methods are rarely applied in an urban environment. We successfully applied these methods to the dense urban county of Philadelphia and three surrounding, less urban counties. We then used this novel flood extent of Hurricane Ida to investigate its distribution in FEMA flood zones and across the population.

4.1 Flood Extent

The flood extent for the study area had an overall accuracy of 99% and the Flood class had a users and producer's accuracy both over 99%. These metrics demonstrate that our methods are accurate when producing a flood extent in urban and suburban areas. A visual inspection also showed that in Philadelphia our RF model accurately classified floods in the frequently flooded neighborhood of Manayunk, the highway Interstate-676, and along a popular path, the Schuylkill Banks.

Our RF model accuracy is relatively high, compared to the accuracies of other studies that used remotely sensed imagery to classify water. Feng et al. (2015) and Rosser et al. (2017) focused on urban flooding in China (87.3% accuracy) and the United Kingdom (95% accuracy), respectively. Studies in non-urban areas also had relatively high accuracies. A study mapping hurricane flooding in North Carolina, USA had an accuracy of 91% (Schaffer-Smith et al., 2020) and a study mapping surface water in a semi-arid river basin in Australia had an overall accuracy of 99% with 80% as the user’s accuracy of the water class (Tulbure et al., 2022). Some of these variations in accuracy likely come from the classification method and type of remote sensing imagery used. Most of the studies used RF methods, similar to our study (Feng et al., 2015; Schaffer-Smith et al., 2020; Tulbure et al., 2022), but others used the Otsu method (Rosser et al., 2017). Additionally, each of these studies used a different type of remote sensing imagery, including Unmanned Aerial Vehicle (Feng et al., 2015), Landsat-8 (Rosser et al., 2017), SAR (Schaffer-Smith et al., 2020), and Harmonized Landsat Sentinel-2 (Tulbure et al., 2022). One possibility of the higher accuracy of this RF model is the relatively small study area and use of several datasets in addition to the satellite imagery.

While the overall accuracies of the flood extent and the Flood class were high, there were undetected Water and Flood areas. Since optical imagery cannot permeate obstructions (e.g., bridges, trees), the RF model did not detect water or floods under these areas. Optical imagery also cannot permeate clouds, therefore to create a flood map for an event with heavy cloud cover, these methods can be applied with SAR imagery. The RF model can be adapted for a cloudy flood event and use SAR imagery, which can permeate clouds, as the main input along with supporting DEM and land cover data, to create a flood extent map.

The RF model had difficulty detecting narrow rivers or creeks. When this occurred, the creek was usually categorized as NotWater or if the river’s banks flooded, then it was categorized as Flood. This was the case for most creeks, for example, Pennypack Creek and Wissahickon Creek, both ~ 20 miles long crossing multiple counties. This study used Sentinel-2 imagery in part because it is publicly accessible, but applying this RF model with higher-resolution imagery from the private sector (e.g. PlanetScope data) may improve classification, especially for smaller water bodies (Cooley et al., 2017).

Our RF model's feature importance varied from other models detecting surface water. For instance, in our RF model the highest-ranked feature was aerosols (Band 1 of Sentinel-2, central wavelength 443 nm, 60 m resolution), and we have not found other models with a similar pattern. Slope was also a highly ranked feature, likely because it is a good predictor of floods since it influences water pooling. The AWEI indices (AWEI_sh and AWEI_nsh) are useful for mapping surface water, along with MNDWI and NDWI (Feyisa et al., 2014; Pickens et al., 2020; Tulbure et al., 2016). In our RF model, the highest-ranked water index was NDFI, although on earlier model runs, MNDWI ranked in the top ten of features. While AWEI_sh is usually helpful for mapping surface water, it may have been less important in this model because it tends to misclassify highly reflective surfaces (such as skyscrapers) as floods (Feyisa et al., 2014).

The JRC Water input, which is a dataset created using Landsat imagery and classifying pixels into permanent, seasonal or not water, (Pekel et al., 2016) was consistently the least important feature. It is likely the least important because the JRC Yearly Water Classification History dataset does not include bodies of water smaller than 30 m by 30 m (Pekel et al., 2016) and the spatial resolution of our study is 10 m. Despite this drawback, when experimenting with different feature combinations, this dataset still slightly increased accuracy and was included in the final RF model.

Overall, it is difficult to compare our RF model and other RF models detecting floods and surface water because each uses different satellite imagery and input features. Our study area also tends to be smaller and more urban than other studies, which is potentially the reason our feature importance differs from other research. While the indices and datasets for this RF model were carefully selected, there is room for model simplification and decreasing the number of features. In addition to decreasing the number of features, experimentation can be done to determine which features are more effective in urban areas versus suburban areas to tailor the RF model to each county.

These methods created an accurate flood extent for an urban area with GEE and free imagery as inputs. While the training data took time to compile, the methods are accessible. They can be quickly deployed to find the flood extent in another urban area when satellite imagery is available. One limitation in making the training and validation dataset is that the curated flood-related tweets obtained from the Global Flood Monitor project are unavailable after February 2023 due to a change in Twitter’s web scraping policies (Calma, 2023).

These accurate flood extents can be used to improve and validate flood models and calculate flood depth (Bangira et al., 2017; Woznicki et al., 2019). Outside of modeling, accurate flood maps created with satellite imagery can help governments at the local and state level with preparedness and mitigation efforts (Akhtar et al., 2021). It can also help governments address the impacts of floods through adaptation projects and fixing zoning regulations (Wing et al., 2022).

4.2 FEMA Flood Zones

In our study area, the Hurricane Ida’s flood extent, by area and percent, predominantly occurred in the 100-yr and Other FEMA flood zones. These flood zones also had the highest proportion of floods compared to the other zones. This result aligns with FEMA’s flood zone descriptions because 100-yr flood zones have a 1% chance of flooding every year and the Other category includes zones that are designed to flood. Since our study compared a singular event to the FEMA flood zones, which represent the probability of yearly flooding, no conclusions can be drawn about the effectiveness of the FEMA zones.

In every county twice the amount of floods, by area and percent, occurred in the Minimal Hazard zone than the 500-yr zone. Since we did not quantify the flood depth or damages, this is not necessarily cause for concern, more a reflection of where water pooled. One county that stood out with a high proportion of flooding in the Minimal Hazard zone was Bucks County. There may be more floods in the Minimal Hazard zone because the 100-yr and 500-yr zones are substantially smaller. When investigating the percentage of the FEMA zones flooded, a higher percentage of the 500-yr flood zone was flooded (0.05% − 0.58%) than the Minimal Hazard zone (0.02% − 0.07%) in all counties. There are two patterns occurring: by area, more floods occurred in the Minimal Hazard zone than the 500-yr zone, whereas by percentage, the 500-yr flood zone had more flooding than the Minimal Hazard zone.

Our research shows that flooding can and does occur in the Minimal Hazard zone, which is important given the misconception that living outside the FEMA flood zone means there is no flood risk (Billings et al., 2022; Wing et al., 2022). FEMA flood zones underestimate flood exposure (Wing et al., 2018) and do not account for pluvial floods (U.S. Government Accountability Office, 2021). Floods and resulting damages regularly occur outside the FEMA 100-yr flood zone (Collins et al., 2022). Our Hurricane Ida flood map can reveal areas with a high proportion of floods in minimal-risk areas that may benefit from recovery assistance and future mitigation. Future research could expand the time scale and determine the rate of floods and the overall effectiveness of different FEMA flood zones in this study area.

4.3 Flood Distribution

We used the flood extent of Hurricane Ida to investigate if flood exposure was equally distributed in the study area. We found that total flood exposure did not disproportionately impact people in poverty and vulnerable populations. While this finding does not align with previous research showing that vulnerable populations are more exposed to floods (Tate et al., 2021), there are a few caveats. Firstly, we looked at a subset of counties impacted by Hurricane Ida. Secondly, we used flood exposure at the Census tract level, when block group level or land parcel level can capture more detailed spatial heterogeneity (Brelsford et al., 2017). Thirdly, for the flood exposure calculation, we used flood area and not flood depth or damages. Still, this result demonstrates an efficient method to get a snapshot of flood distribution and can be deployed again in conjunction with flood depth for a more accurate flood exposure calculation. Too often, flooding disproportionately affects vulnerable populations with the fewest resources to recover (Schaffer-Smith et al., 2020; Tate et al., 2021). Therefore, it is important to research this trend and highlight vulnerable areas that received high amounts of flooding and could benefit from additional support and resources to recover after flooding.

Accurate methods for quantifying flood extent can provide insights into the affected area and damages (Rosser et al., 2017), and inform response (Akhtar et al., 2021). Satellite imagery and the RF algorithm are a reliable combination to create a flood extent but are not often applied in urban areas. Our methods combine GEE, satellite imagery and other freely available data to create a RF model. This model created a new flood extent of Hurricane Ida in southeastern Pennsylvania with 99% accuracy and can be applied to other urban areas. This flood extent can be used to validate flood models and view patterns of flooding at the tract level. We investigated the distribution of this event in FEMA flood zones and found that most flooding occurred in the 100-yr and Other zones. In our study area, more floods occurred in the Minimal Hazard zone than the 500-yr zone, affirming previous research that consistently found flooding outside FEMA’s flood zones. This research refined methods for creating an accurate flood extent in urban areas and created a new one for our study area. Flood extent maps can serve stakeholders such as land use managers (Sofia et al., 2017), emergency planners (Goffi et al., 2020), city planners and residents (Hosseiny et al., 2020). The maps can help inform recovery efforts, prioritize mitigation efforts (Hosseiny et al., 2020) and plan for future development (Sofia et al., 2017), making our cities more resilient to the increasing risk of floods.

6. Acknowledgements

We appreciate the thorough comments and suggestions from the anonymous reviewers and editor. We also appreciate the European Space Agency for Sentinel-2A data, Planet Labs for PlanetScope data and Google for the Google Earth Engine (GEE) platform.

9.1 Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

9.2 Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

9.3 Author Contributions

Rebecca Composto: Conceptualization, Investigation, Methodology, Formal Analysis, Visualization, Writing - original draft, Writing - review & editing. Mirela Tulbure: Supervision, Conceptualization, Methodology, Writing - review & editing. Varun Tiwari: Methodology (supporting), Formal Analysis (supporting GEE code), Writing - review & editing. Mollie D. Gaines: Methodology (supporting), Formal Analysis (supporting Python code), Writing - review & editing. Júlio Caineta: Methodology (supporting), Writing - Review & editing.

Akhtar, Z., Ofli, F., & Imran, M. (2021). Towards Using Remote Sensing and Social Media Data for Flood Mapping. ISCRAM 2021 Conference Proceedings–18th International Conference on Information Systems for Crisis Response and Management, 536–551.

Apel, H., Martínez Trepat, O., Hung, N. N., Chinh, D. T., Merz, B., & Dung, N. V. (2016). Combined fluvial and pluvial urban flood hazard analysis: Concept development and application to Can Tho city, Mekong Delta, Vietnam. Natural Hazards and Earth System Sciences, 16(4), 941–961. https://doi.org/10.5194/nhess-16-941-2016

Ayanu, Y. Z., Conrad, C., Nauss, T., Wegmann, M., & Koellner, T. (2012). Quantifying and Mapping Ecosystem Services Supplies and Demands: A Review of Remote Sensing Applications. Environmental Science & Technology, 46(16), 8529–8541. https://doi.org/10.1021/es300157u

Bangira, T., Alfieri, S. M., Menenti, M., Van Niekerk, A., & Vekerdy, Z. (2017). A Spectral Unmixing Method with Ensemble Estimation of Endmembers: Application to Flood Mapping in the Caprivi Floodplain. Remote Sensing, 9(10), Article 10. https://doi.org/10.3390/rs9101013

Beven II, J., Hagen, A., & Berg, R. (2022). National Hurricane Center Tropical Cyclone Report: Hurricane Ida. NOAA. https://www.nhc.noaa.gov/data/tcr/AL092021_Ida.pdf

Billings, S. B., Gallagher, E. A., & Ricketts, L. (2022). Let the rich be flooded: The distribution of financial aid and distress after hurricane harvey. Journal of Financial Economics, 146(2), 797–819. https://doi.org/10.1016/j.jfineco.2021.11.006

Blum, A. G., Ferraro, P. J., Archfield, S. A., & Ryberg, K. R. (2020). Causal Effect of Impervious Cover on Annual Flood Magnitude for the United States. Geophysical Research Letters, 47(5), e2019GL086480. https://doi.org/10.1029/2019GL086480

Boschetti, M., Nutini, F., Manfron, G., Brivio, P. A., & Nelson, A. (2014). Comparative Analysis of Normalised Difference Spectral Indices Derived from MODIS for Detecting Surface Water in Flooded Rice Cropping Systems. PLOS ONE, 9(2), e88741. https://doi.org/10.1371/journal.pone.0088741

Brandt, S. A., Lim, N. J., Colding, J., & Barthel, S. (2021). Mapping Flood Risk Uncertainty Zones in Support of Urban Resilience Planning. Urban Planning, 6(3), 258–271. https://doi.org/10.17645/up.v6i3.4073

Breiman, L. (2001). Random Forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324

Brelsford, C., Lobo, J., Hand, J., & Bettencourt, L. M. A. (2017). Heterogeneity and scale of sustainable development in cities. Proceedings of the National Academy of Sciences, 114(34), 8963–8968. https://doi.org/10.1073/pnas.1606033114

C. Yang, J. H. Everitt, & J. M. Bradford. (2007). Using Multispectral Imagery and Linear Spectral Unmixing Techniques for Estimating Crop Yield Variability. Transactions of the ASABE, 50(2), 6676–674. https://doi.org/10.13031/2013.22658

Calma, J. (2023, May 31). Scientists say they can’t rely on Twitter anymore. The Verge. https://www.theverge.com/2023/5/31/23739084/twitter-elon-musk-api-policy-chilling-academic-research

CDC. (2020). CDC/ATSDR Social Vulnerability Index [Database State]. https://www.atsdr.cdc.gov/placeandhealth/svi/data_documentation_download.html

Census. (2021). City and Town Population Totals: 2020-2021. Census.Gov. https://www.census.gov/data/tables/time-series/demo/popest/2020s-total-cities-and-towns.html

Clement, M. a., Kilsby, C. g., & Moore, P. (2018). Multi-temporal synthetic aperture radar flood mapping using change detection. Journal of Flood Risk Management, 11(2), 152–168. https://doi.org/10.1111/jfr3.12303

Collins, E. L., Sanchez, G. M., Terando, A., Stillwell, C. C., Mitasova, H., Sebastian, A., & Meentemeyer, R. K. (2022). Predicting flood damage probability across the conterminous United States. Environmental Research Letters, 17(3), 034006. https://doi.org/10.1088/1748-9326/ac4f0f

Cooley, S. W., Smith, L. C., Stepan, L., & Mascaro, J. (2017). Tracking Dynamic Northern Surface Water Changes with High-Frequency Planet CubeSat Imagery. Remote Sensing, 9(12), Article 12. https://doi.org/10.3390/rs9121306

Cooper, K., Rizzo, E., & Schmidt, S. (2022, September 1). One year after Hurricane Ida, Pa. Residents are still paying the price. WHYY. https://whyy.org/articles/pensylvannia-hurricane-ida-one-year-anniversary/

CRED. (2015). The Human Cost of Natural Disasters: A global perspective. Centre for Research on the Epidemiology of Disasters (CRED). http://repo.floodalliance.net/jspui/44111/1165

de Bruijn, J. A., de Moel, H., Jongman, B., de Ruiter, M. C., Wagemaker, J., & Aerts, J. C. J. H. (2019). A global database of historic and real-time flood events based on social media. Scientific Data, 6(1), Article 1. https://doi.org/10.1038/s41597-019-0326-9

FEMA. (2020, July 8). Flood Zones. https://www.fema.gov/glossary/flood-zones

FEMA. (2023, October 23). Laws and Regulations. https://www.fema.gov/flood-insurance/rules-legislation/laws

Feyisa, G. L., Meilby, H., Fensholt, R., & Proud, S. R. (2014). Automated Water Extraction Index: A new technique for surface water mapping using Landsat imagery. Remote Sensing of Environment, 140, 23–35. https://doi.org/10.1016/j.rse.2013.08.029

Fielding, J. L. (2018). Flood risk and inequalities between ethnic groups in the floodplains of England and Wales. Disasters, 42(1), 101–123. https://doi.org/10.1111/disa.12230

Gao, B. (1996). NDWI—A normalized difference water index for remote sensing of vegetation liquid water from space. Remote Sensing of Environment, 58(3), 257–266. https://doi.org/10.1016/S0034-4257(96)00067-3

Garbutt, K., Ellul, C., & Fujiyama, T. (2015). Mapping social vulnerability to flood hazard in Norfolk, England. Environmental Hazards, 14(2), 156–186. https://doi.org/10.1080/17477891.2015.1028018

Goffi, A., Stroppiana, D., Brivio, P. A., Bordogna, G., & Boschetti, M. (2020). Towards an automated approach to map flooded areas from Sentinel-2 MSI data and soft integration of water spectral features. International Journal of Applied Earth Observation and Geoinformation, 84, 101951. https://doi.org/10.1016/j.jag.2019.101951

Gómez-Palacios, D., Torres, M. A., & Reinoso, E. (2017). Flood mapping through principal component analysis of multitemporal satellite imagery considering the alteration of water spectral properties due to turbidity conditions. Geomatics, Natural Hazards and Risk, 8(2), 607–623. https://doi.org/10.1080/19475705.2016.1250115

Hall, T. M., & Kossin, J. P. (2019). Hurricane stalling along the North American coast and implications for rainfall. Npj Climate and Atmospheric Science, 2(1), Article 1. https://doi.org/10.1038/s41612-019-0074-8

Hermas, E., Gaber, A., & El Bastawesy, M. (2021). Application of remote sensing and GIS for assessing and proposing mitigation measures in flood-affected urban areas, Egypt. The Egyptian Journal of Remote Sensing and Space Science, 24(1), 119–130. https://doi.org/10.1016/j.ejrs.2020.03.002

Holland, G., & Bruyère, C. L. (2014). Recent intense hurricane response to global climate change. Climate Dynamics, 42(3), 617–627. https://doi.org/10.1007/s00382-013-1713-0

Hondula, K. L., DeVries, B., Jones, C. N., & Palmer, M. A. (2021). Effects of Using High Resolution Satellite‐Based Inundation Time Series to Estimate Methane Fluxes From Forested Wetlands. Geophysical Research Letters, 48(6). https://doi.org/10.1029/2021GL092556

Hosseiny, H., Crimmins, M., Smith, V. B., & Kremer, P. (2020). A Generalized Automated Framework for Urban Runoff Modeling and Its Application at a Citywide Landscape. Water, 12(2), 357. https://doi.org/10.3390/w12020357

Huete, A. (1997). A comparison of vegetation indices over a global set of TM images for EOS-MODIS. Remote Sensing of Environment, 59(3), 440–451. https://doi.org/10.1016/S0034-4257(96)00112-5

Ireland, G., Volpi, M., & Petropoulos, G. P. (2015). Examining the Capability of Supervised Machine Learning Classifiers in Extracting Flooded Areas from Landsat TM Imagery: A Case Study from a Mediterranean Flood. Remote Sensing, 7(3), Article 3. https://doi.org/10.3390/rs70303372

Jones, J. W. (2019). Improved Automated Detection of Subpixel-Scale Inundation—Revised Dynamic Surface Water Extent (DSWE) Partial Surface Water Tests. Remote Sensing, 11(4), Article 4. https://doi.org/10.3390/rs11040374

Kawasaki, A., Kawamura, G., & Zin, W. W. (2020). A local level relationship between floods and poverty: A case in Myanmar. International Journal of Disaster Risk Reduction, 42, 101348. https://doi.org/10.1016/j.ijdrr.2019.101348

Knighton, J., Hondula, K., Sharkus, C., Guzman, C., & Elliott, R. (2021). Flood risk behaviors of United States riverine metropolitan areas are driven by local hydrology and shaped by race. Proceedings of the National Academy of Sciences, 118(13), e2016839118. https://doi.org/10.1073/pnas.2016839118

Knutson, T. R., Sirutis, J. J., Zhao, M., Tuleya, R. E., Bender, M., Vecchi, G. A., Villarini, G., & Chavas, D. (2015). Global Projections of Intense Tropical Cyclone Activity for the Late Twenty-First Century from Dynamical Downscaling of CMIP5/RCP4.5 Scenarios. Journal of Climate, 28(18), 7203–7224. https://doi.org/10.1175/JCLI-D-15-0129.1

Kossin, J. P. (2018). A global slowdown of tropical-cyclone translation speed. Nature, 558(7708), Article 7708. https://doi.org/10.1038/s41586-018-0158-3

Kossin, J. P., Knapp, K. R., Vimont, D. J., Murnane, R. J., & Harper, B. A. (2007). A globally consistent reanalysis of hurricane variability and trends. Geophysical Research Letters, 34(4). https://doi.org/10.1029/2006GL028836

Kriegler, F. J., Malila, W. A., Nalepka, R. F., & Richardson, W. (1969). Preprocessing Transformations and Their Effects on Multispectral Recognition (p. 97). https://ui.adsabs.harvard.edu/abs/1969rse..conf...97K

Lin, N., Emanuel, K., Oppenheimer, M., & Vanmarcke, E. (2012). Physically based assessment of hurricane surge threat under climate change. Nature Climate Change, 2(6), Article 6. https://doi.org/10.1038/nclimate1389

Liu, L., Liu, Y., Wang, X., Yu, D., Liu, K., Huang, H., & Hu, G. (2015). Developing an effective 2-D urban flood inundation model for city emergency management based on cellular automata. Natural Hazards and Earth System Sciences, 15(3), 381–391. https://doi.org/10.5194/nhess-15-381-2015

Lorenz, M. O. (1905). Methods of Measuring the Concentration of Wealth. Publications of the American Statistical Association, 9(70), 209. https://doi.org/10.2307/2276207

Markhvida, M., Walsh, B., Hallegatte, S., & Baker, J. (2020). Quantification of disaster impacts through household well-being losses. Nature Sustainability, 3(7), Article 7. https://doi.org/10.1038/s41893-020-0508-7

Mason, D. C., Giustarini, L., Garcia-Pintado, J., & Cloke, H. L. (2014). Detection of flooded urban areas in high resolution Synthetic Aperture Radar images using double scattering. International Journal of Applied Earth Observation and Geoinformation, 28, 150–159. https://doi.org/10.1016/j.jag.2013.12.002

Maxwell, A. E., Warner, T. A., & Fang, F. (2018). Implementation of machine-learning classification in remote sensing: An applied review. International Journal of Remote Sensing, 39(9), 2784–2817. https://doi.org/10.1080/01431161.2018.1433343

McFeeters, S. K. (1996). The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features. International Journal of Remote Sensing, 17(7), 1425–1432. https://doi.org/10.1080/01431169608948714

Mtapuri, O., Dube, E., & Matunhu, J. (2018). Flooding and poverty: Two interrelated social problems impacting rural development in Tsholotsho district of Matabeleland North province in Zimbabwe. Jamba : Journal of Disaster Risk Studies, 10(1), 1–7. https://doi.org/10.4102/jamba.v10i1.455

NOAA. (2021, September). National Weather Service Forecast Office. https://www.weather.gov/wrh/climate?wfo=phi

Olofsson, P., Foody, G. M., Herold, M., Stehman, S. V., Woodcock, C. E., & Wulder, M. A. (2014). Good practices for estimating area and assessing accuracy of land change. Remote Sensing of Environment, 148, 42–57. https://doi.org/10.1016/j.rse.2014.02.015

Pekel, J.-F., Cottam, A., Gorelick, N., & Belward, A. S. (2016). High-resolution mapping of global surface water and its long-term changes. Nature, 540(7633), 418–422. https://doi.org/10.1038/nature20584

Perin, V., Tulbure, M. G., Gaines, M. D., Reba, M. L., & Yaeger, M. A. (2022). A multi-sensor satellite imagery approach to monitor on-farm reservoirs. Remote Sensing of Environment, 270, 112796. https://doi.org/10.1016/j.rse.2021.112796

Phan, T. N., Kuch, V., & Lehnert, L. W. (2020). Land Cover Classification using Google Earth Engine and Random Forest Classifier—The Role of Image Composition. Remote Sensing, 12(15), Article 15. https://doi.org/10.3390/rs12152411

Pickens, A. H., Hansen, M. C., Hancher, M., Stehman, S. V., Tyukavina, A., Potapov, P., Marroquin, B., & Sherani, Z. (2020). Mapping and sampling to characterize global inland water dynamics from 1999 to 2018 with full Landsat time-series. Remote Sensing of Environment, 243, 111792. https://doi.org/10.1016/j.rse.2020.111792

Pinos, J., & Quesada-Román, A. (2022). Flood Risk-Related Research Trends in Latin America and the Caribbean. Water, 14(1), Article 1. https://doi.org/10.3390/w14010010

Pulcinella, M., Meyer, K., & Cooper, K. (2021, September 1). Long recovery ahead as Ida’s remnants lead to historic flooding, tornadoes in Philly region. WHYY. https://whyy.org/articles/philly-says-to-shelter-in-place-as-schuylkill-river-expected-to-rise-to-major-flood-stage/

Rentschler, J., Salhab, M., & Jafino, B. A. (2022). Flood exposure and poverty in 188 countries. Nature Communications, 13(1), Article 1. https://doi.org/10.1038/s41467-022-30727-4

Rosser, J. F., Leibovici, D. G., & Jackson, M. J. (2017). Rapid flood inundation mapping using social media, remote sensing and topographic data. Natural Hazards, 87(1), 103–120. https://doi.org/10.1007/s11069-017-2755-0

Sanders, B. F., Schubert, J. E., Kahl, D. T., Mach, K. J., Brady, D., AghaKouchak, A., Forman, F., Matthew, R. A., Ulibarri, N., & Davis, S. J. (2022). Large and inequitable flood risks in Los Angeles, California. Nature Sustainability, 1–11. https://doi.org/10.1038/s41893-022-00977-7

Schaffer-Smith, D., Myint, S. W., Muenich, R. L., Tong, D., & DeMeester, J. E. (2020). Repeated Hurricanes Reveal Risks and Opportunities for Social-Ecological Resilience to Flooding and Water Quality Problems. Environmental Science & Technology, 54(12), 7194–7204. https://doi.org/10.1021/acs.est.9b07815

Schnebele, E., Cervone, G., & Waters, N. (2014). Road assessment after flood events using non-authoritative data. Natural Hazards and Earth System Sciences, 14(4), 1007–1015. https://doi.org/10.5194/nhess-14-1007-2014

Settle, J. J., & Drake, N. A. (1993). Linear mixing and the estimation of ground cover proportions. International Journal of Remote Sensing, 14(6), 1159–1177. https://doi.org/10.1080/01431169308904402

Shen, L., & Li, C. (2010). Water body extraction from Landsat ETM+ imagery using adaboost algorithm. 2010 18th International Conference on Geoinformatics, 1–4. https://doi.org/10.1109/GEOINFORMATICS.2010.5567762

Smith, A. B. (2023). U.S. Billion-dollar Weather and Climate Disasters, 1980—Present (NCEI Accession 0209268) [dataset]. NOAA National Centers for Environmental Information. https://doi.org/10.25921/STKW-7W73

Sofia, G., Roder, G., Dalla Fontana, G., & Tarolli, P. (2017). Flood dynamics in urbanised landscapes: 100 years of climate and humans’ interaction. Scientific Reports, 7(1), Article 1. https://doi.org/10.1038/srep40527

Sweet, W., Dusek, G. (Gregory P. ), Marcy, D. C., Greg (Gregory W.), C., & Marra, J. (2019). 2018 State of U.S. High Tide Flooding with a 2019 Outlook. https://doi.org/10.25923/RBV9-TH19

Tanim, A. H., McRae, C. B., Tavakol-Davani, H., & Goharian, E. (2022). Flood Detection in Urban Areas Using Satellite Imagery and Machine Learning. Water, 14(7), Article 7. https://doi.org/10.3390/w14071140

Tate, E., Rahman, M. A., Emrich, C. T., & Sampson, C. C. (2021). Flood exposure and social vulnerability in the United States. Natural Hazards, 106(1), 435–457. https://doi.org/10.1007/s11069-020-04470-2

Tiwari, V., Tulbure, M. G., Caineta, J., Gaines, M. D., Perin, V., Kamal, M., Krupnik, T. J., Aziz, M. A., & Islam, A. T. (2024). Automated in-season rice crop mapping using Sentinel time-series data and Google Earth Engine: A case study in climate-risk prone Bangladesh. Journal of Environmental Management, 351, 119615. https://doi.org/10.1016/j.jenvman.2023.119615

Trenberth, K. E., Cheng, L., Jacobs, P., Zhang, Y., & Fasullo, J. (2018). Hurricane Harvey Links to Ocean Heat Content and Climate Change Adaptation. Earth’s Future, 6(5), 730–744. https://doi.org/10.1029/2018EF000825

Tulbure, M. G., & Broich, M. (2013). Spatiotemporal dynamic of surface water bodies using Landsat time-series data from 1999 to 2011. ISPRS Journal of Photogrammetry and Remote Sensing, 79, 44–52. https://doi.org/10.1016/j.isprsjprs.2013.01.010

Tulbure, M. G., Broich, M., Ju, J., Masek, J. G., & Wearne, J. (2018). Quantifying surface water extent and flooding in a dynamic dryland river system using the Harmonized Landsat/Sentinel-2 Reflectance Product. 2018, H21E-08.

Tulbure, M. G., Broich, M., Perin, V., Gaines, M., Ju, J., Stehman, S. V., Pavelsky, T., Masek, J. G., Yin, S., Mai, J., & Betbeder-Matibet, L. (2022). Can we detect more ephemeral floods with higher density harmonized Landsat Sentinel 2 data compared to Landsat 8 alone? ISPRS Journal of Photogrammetry and Remote Sensing, 185, 232–246. https://doi.org/10.1016/j.isprsjprs.2022.01.021

Tulbure, M. G., Broich, M., Stehman, S. V., & Kommareddy, A. (2016). Surface water extent dynamics from three decades of seasonally continuous Landsat time series at subcontinental scale in a semi-arid region. Remote Sensing of Environment, 178, 142–157. https://doi.org/10.1016/j.rse.2016.02.034

U.S. Census Bureau. (2020). County-level Urban and Rural information for the 2020 Census [dataset]. https://www.census.gov/programs-surveys/geography/guidance/geo-areas/urban-rural.html

US Department of Commerce. (2022, October 26). Hurricane Ida Emergency Response Imagery. https://oceanservice.noaa.gov/news/aug21/ngs-storm-imagery-ida.html

US Department of Commerce, N. (2021). NWS Preliminary US Flood Fatality Statistics. NOAA’s National Weather Service. https://www.weather.gov/arx/usflood

U.S. Geological Survey. (2023). 3D Elevation Program 10-Meter Resolution Digital Elevation Model. [dataset]. https://www.usgs.gov/the-national-map-data-delivery

U.S. Government Accountability Office. (2021). FEMA Flood Maps: Better Planning and Analysis Needed to Address Current and Future Flood Hazards. https://www.gao.gov/assets/gao-22-104079.pdf

Van Oldenborgh, G. J., Van Der Wiel, K., Sebastian, A., Singh, R., Arrighi, J., Otto, F., Haustein, K., Li, S., Vecchi, G., & Cullen, H. (2017). Attribution of extreme rainfall from Hurricane Harvey, August 2017. Environmental Research Letters, 12(12), 124009. https://doi.org/10.1088/1748-9326/aa9ef2

Wing, O. E. J., Bates, P. D., Smith, A. M., Sampson, C. C., Johnson, K. A., Fargione, J., & Morefield, P. (2018). Estimates of present and future flood risk in the conterminous United States. Environmental Research Letters, 13(3), 034023. https://doi.org/10.1088/1748-9326/aaac65

Wing, O. E. J., Lehman, W., Bates, P. D., Sampson, C. C., Quinn, N., Smith, A. M., Neal, J. C., Porter, J. R., & Kousky, C. (2022). Inequitable patterns of US flood risk in the Anthropocene. Nature Climate Change, 12(2), Article 2. https://doi.org/10.1038/s41558-021-01265-6

Winsemius, H. C., Jongman, B., Veldkamp, T. I. E., Hallegatte, S., Bangalore, M., & Ward, P. J. (2018). Disaster risk, climate change, and poverty: Assessing the global exposure of poor people to floods and droughts. Environment and Development Economics, 23(3), 328–348. https://doi.org/10.1017/S1355770X17000444

Woznicki, S. A., Baynes, J., Panlasigui, S., Mehaffey, M., & Neale, A. (2019). Development of a spatially complete floodplain map of the conterminous United States using random forest. Science of The Total Environment, 647, 942–953. https://doi.org/10.1016/j.scitotenv.2018.07.353

Xu, H. (2006). Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery. International Journal of Remote Sensing, 27(14), 3025–3033. https://doi.org/10.1080/01431160600589179

Download PDF

Journal Publication

published 27 Jul, 2024

Read the published version in Natural Hazards →

Editorial decision: Major revisions
03 Jun, 2024
Reviewers agreed at journal
27 Feb, 2024
Reviewers invited by journal
19 Feb, 2024
Editor assigned by journal
23 Jan, 2024
First submitted to journal
22 Jan, 2024

You are reading this latest preprint version

Quantifying Urban Flood Extent Using Satellite Imagery and Random Forest: A Case Study in Southeastern Pennsylvania

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

1.1 Floods Cause Harm

1.2 Hurricanes risk increasing

1.3 Unequal distribution of risk

1.4 Hurricane Ida 2021

1.5 Flood extent from satellite imagery

1.6 Gaps and objectives

2. Methodology

2.1 Study site

2.2 Datasets

2.2.1 Satellite Imagery

2.2.2 Sentinel-2 Indices

2.2.3 Additional Inputs

2.2.4 Training and Validation Data

2.2.5 FEMA Flood Zones

2.2.6 Vulnerability and Demographic Datasets

2.3 Random Forest Model

2.4 Assessment of flood impact

3. Results

3.1 Flood Extent Generated with Random Forest Model

3.2 Comparison to FEMA Flood zones

3.3 Vulnerability and Poverty Distribution

4. Discussion

4.1 Flood Extent

4.2 FEMA Flood Zones

4.3 Flood Distribution

5. Conclusion

Declarations

6. Acknowledgements

References

Status:

Journal Publication

Version 1