Hostname: page-component-586b7cd67f-vdxz6 Total loading time: 0 Render date: 2024-11-22T05:53:29.012Z Has data issue: false hasContentIssue false

Buy them out before they are built: evaluating the proactive acquisition of vacant land in flood-prone areas

Published online by Cambridge University Press:  12 March 2021

Kayode Atoba*
Affiliation:
Institute for a Disaster Resilient Texas, Texas A&M University, Galveston Campus, College Station, TX, USA
Galen Newman
Affiliation:
Landscape Architecture & Urban Planning Department, Texas A&M University, College Station, TX, USA
Samuel Brody
Affiliation:
Institute for a Disaster Resilient Texas, Texas A&M University, Galveston Campus, College Station, TX, USA
Wesley Highfield
Affiliation:
Institute for a Disaster Resilient Texas, Texas A&M University, Galveston Campus, College Station, TX, USA
Youjung Kim
Affiliation:
Geography, Planning and Environment Department, Concordia University, Montréal, QC, Canada
Andrew Juan
Affiliation:
Civil and Environmental Engineering Department, Rice University, Houston, TX, USA
*
Author for correspondence: Dr Kayode Atoba, Email: [email protected]
Rights & Permissions [Opens in a new window]

Summary

Rising flood damages have prompted local communities to implement buyout and property acquisition programmes to eliminate repetitive losses for at-risk properties. However, buyouts are often costly to implement and are reactionary solutions to flooding. This study quantifies the benefits of acquiring vacant private properties in flood-prone areas rather than acquiring such properties after they are built up. Using a geodesign framework that integrates concepts and analytical approaches derived from geographical, spatial and statistical-based disciplines, we analyse vacant properties with high development potential that intersect current and future floodplain areas in Houston (TX, USA). We use geospatial proximity analysis to select candidate properties, land-use prediction modelling to estimate future development and sea-level rise and benefit–cost analysis to assess the economic viability of buyouts. The results indicate that cumulative avoided flood losses exceed the cost of vacant land acquisition by a factor of nearly two to one, and up to a factor of ten to one in selected areas. This study emphasizes the benefits of proactive property buyouts that focus on acquiring parcels before they are built up, while also avoiding the social and institutional problems associated with traditional buyout programmes.

Type
Research Paper
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright
© The Author(s), 2021. Published by Cambridge University Press on behalf of Foundation for Environmental Conservation

Introduction

Following the catastrophic impact of large storms, purchasing chronically damaged properties and restoring them to open spaces continues to gain popularity as a flood mitigation strategy in the USA (Zavar Reference Zavar2015). This is usually referred to as a buyout, and it traditionally involves local governments’ purchase of buildings in flood-prone areas that have been repeatedly flooded and restoring them to vacant open spaces. Over 40 000 such properties have been acquired across the USA since 1989, primarily through the Federal Emergency Management Agency (FEMA) programme funds (Mach et al. Reference Mach, Kraan, Hino, Siders, Johnston and Field2019). Floodplain-based property acquisition is expected to play an increasing role in local floodplain management (Winsemius et al. Reference Winsemius, Aerts, van Beek, Bierkens, Bouwman and Jongman2016, Wing et al. Reference Wing, Bates, Smith, Sampson, Johnson, Fargione and Morefield2018).

Buyouts often result in economic and ecological benefits, such as the prevention of flood losses and protection of environmental assets (Conrad et al. Reference Conrad, McNitt and Stout1998, Zavar Reference Zavar2015). However, most buyout programmes are initiated in a reactionary manner, which leads to several problems. For example, homeowners who relocate after a buyout may face socioeconomic and psychological distress (Binder et al. Reference Binder, Baker and Barile2015, Reference Binder, Greer and Zavar2020). Many buyout programmes in the USA are also flawed by limited policy learning and a lack of transparency in the selection process, potentially contributing to social injustice and reduced participation for socially vulnerable households (Greer & Binder Reference Greer and Binder2017, Siders Reference Siders2019). Checker-boarding of buyouts may also limit opportunities for creating open space clusters, resulting in small, isolated patches of open spaces in residential neighbourhoods (Blanco et al. Reference Blanco, Alberti, Olshansky, Chang, Wheeler and Randolph2009, Maly & Ishikawa Reference Maly and Ishikawa2013, Freudenberg et al. Reference Freudenberg, Calvin, Tolkoff and Brawley2016, Zavar & Hagelman III Reference Zavar and Hagelman2016). Additionally, many property owners are emotionally attached to their homes and reluctant to leave, even under dire circumstances; the perception of the government taking over private property often makes homeowners less willing to participate in voluntary buyout programmes (Maldonado et al. Reference Maldonado, Shearer, Bronen, Peterson, Lazrus, Maldonado, Colombi and Pandya2013, Marino Reference Marino2018). These limitations make purchasing properties with existing development more problematic compared to buying out vacant land or promoting policies that prevent development in high-risk locations.

This study builds on the idea of proactive land acquisition and conservation by examining private vacant lands at multiple spatial scales. Rather than focusing on the existing open space restoration approach (demolishing buildings and restoring parcels to open space), this study analyses an open space conservation approach (buyout of vacant land before buildings are erected). ‘Vacant land’ is a ubiquitous urban land use (Pagano & Bowman Reference Pagano and Bowman2000) and accounts for nearly 17% of land in each large US city (Newman et al. Reference Newman, Bowman, Jung Lee and Kim2016). When vacant lands in flood-prone areas remain vacant, they automatically record no direct economic loss from flooding and can even reduce flood losses to surrounding development by storing floodwaters (Brody et al. Reference Brody, Blessing, Sebastian and Bedient2014, Reference Brody, Highfield, Blessing, Makino and Shepard2017, Highfield et al. Reference Highfield, Brody and Shepard2018). Vacant open spaces provide other opportunities such as protecting wetlands and floodplains (Conrad et al. Reference Conrad, McNitt and Stout1998), restoring native vegetation (Harter Reference Harter2007), increasing property value due to opportunities for recreation (Geoghegan Reference Geoghegan2002, Crompton Reference Crompton2005), biodiversity conservation (Hausmann et al. Reference Hausmann, Slotow, Burns and Di Minin2016) and so on.

Previous studies have assessed the economic benefits of conserving open spaces. For example, Johnson et al. (Reference Johnson, Wing, Bates, Fargione, Kroeger and Larson2020) estimated the cost and benefits of conserving open spaces across different flood zones in the conterminous USA while also accounting for future development projections. Although a nationwide study such as this provides a broad picture of proactive open space protection at a large scale, the study does not capture the nuanced local conditions at smaller scales, thereby providing a comparative component for local decision-makers to consider when making decisions about open space management. These decisions become even more critical due to changes in floodplain delineations resulting from sea-level rise (SLR). Kousky and Walls (Reference Kousky and Walls2014), on the other hand, estimated the benefits of open space conservation at a smaller scale. While their study provides a local context to vacant open space acquisition, it does not account for future development projections, which may result in overestimating open space benefits. Additionally, the study does not address the potential impact of SLR, which may lead to underestimating the avoided losses from vacant land conservation.

To address these gaps, we quantified the avoided flood damages to vacant private properties given future development and SLR projections at a small spatial scale. We develop a methodological framework that identifies eligible vacant properties for buyouts through a multi-criterion spatial procedure. This study is aimed at meeting two objectives: (1) apply a geodesign framework (Steinitz Reference Steinitz2012) in selecting vacant flood-prone properties that intersect future development and SLR conditions; and (2) assess the economic implications of buying out vacant flood-prone properties now as opposed to buying them out post-development given future built environment and SLR conditions. This approach highlights the importance of accounting for local demographic and development conditions in the configuration of open spaces in a manner that will save local communities the additional cost of constructing public infrastructure in vulnerable locations (BenDor et al. Reference BenDor, Salvesen, Kamrath and Ganser2020). This also ensures that local officials evaluate open space conservation options without overstating or underestimating their economic benefits, while also accounting for future environmental conditions.

Methods and data analysis

The analytical basis for selecting and evaluating candidate vacant properties for buyouts is informed by the geodesign framework (Steinitz Reference Steinitz2012). This model systematically integrates concepts and analytical approaches derived from geographical, spatial and statistical-based disciplines in site selection. We apply this framework by integrating methods related to flood resilience, urban growth prediction modelling and planned real estate development. Newman et al. (Reference Newman, Malecha, Yu, Qiao, Horney and Lee2019) applied a similar approach in selecting a site for landscape performance and flood analysis. We begin our analysis by selecting eligible vacant properties that intersect flood risk and future development in Houston, Texas. We further highlight a change watershed that is prone to flooding, new development and SLR. Next, we create an impact model, which tests a case of vacant test sites in the selected watershed. The final decision model focuses on conducting a site analysis to determine whether the benefit–cost ratio (BCR) of buying out vacant land now is higher than that of buying out a set of hypothetical built properties on the vacant land. Table 1 highlights the geodesign process, while Fig. 1 shows the selection and analytical process of the multi-scale analysis in this research.

Table 1. Vacant parcel selection and analysis through the geodesign process (adapted from Steinitz Reference Steinitz2012, Newman et al. Reference Newman, Malecha, Yu, Qiao, Horney and Lee2019).

BCR = benefit–cost ratio.

Fig. 1. Selection framework for vacant properties in flood-risk areas. BCR = benefit–cost ratio.

Study area

The Houston-Galveston region is one of the most flood-impacted area in the USA, experiencing on average one hurricane every 15 years (Parisi & Lund Reference Parisi and Lund2008). Recent hurricanes in the region such as Ike (2008) and Harvey (2017) caused damages exceeding US$35 and US$125 billion, respectively (NOAA 2017). FEMA has spent c. US$555 million to acquire over 4300 flooded properties throughout the state, with c. US$205 million of this total used to acquire properties in the Houston area alone, while Harris County (which is covered mainly by the city of Houston) has paid out over US$342 million to acquire c. 2462 properties since 1985 (Atoba et al. Reference Atoba, Brody, Highfield, Shepard and Verdone2020). Rising population coupled with expanding development in the region will continue to exacerbate flood problems (see HGAC 2018 for recent population statistics). Land cover data from the US Geological Survey (USGS) shows that urban area in Harris County expanded from 1854.5 km2 (716.0 mi2) in 2001 to 2203.5 km2 (850.8 mi2) in 2011. Currently, within the city of Houston, there are over 800 000 residential properties and over 90 000 vacant parcels (c. 22 400 of which are in the 100-year floodplain). Figure 2 shows the locations of all of the vacant properties in the study area, indicating the vast amount of vacant land available for future development.

Fig. 2. Hydrological Unit Code 12 (HUC-12) watersheds and vacant properties in Houston, Texas.

Predicting future development and SLR floodplains

We predict future urban growth using the Land Transformation Model (LTM), a spatial prediction tool that uses an artificial neural network model to predict future land use (Pijanowski et al. Reference Pijanowski, Brown, Shellito and Manik2002). This study employs 12 driving factors and 2 land cover maps from 2001 to 2011 and compares the predicted 2011 land use with the real land cover map in 2011 to justify model validation. The drivers include proximity-based measures (water, floodplain, highways, roads, bus routes, parks, businesses, existing urban, hospitals and schools) and sociodemographic measures (population density and race) frequently employed in urban land change literature (Pijanowski et al. Reference Pijanowski, Tayyebi, Doucette, Pekin, Braun and Plourde2014, Jafari et al. Reference Jafari, Majedi, Monavari, Alesheikh and Kheirkhah Zarkesh2016, Ku Reference Ku2016, Losiri et al. Reference Losiri, Nagai, Ninsawat and Shrestha2016, Kim & Newman Reference Kim and Newman2020). By following the previous urban growth ratio of land area (99.45 km2, 9445 pixels in 100 × 100 m resolution) for the population increase (287 343 persons) between 2001 and 2011, the forecasted future urban area expands to 273.57 km2 (27 357 pixels) for the projected population increase of 832 294 persons between 2012 and 2040, according to the State Plan Population Projection Data (TWDB 2012).

The prediction was validated with four calibration measures: percent correct metric (PCM), κ coefficient, overall agreement (OA) and area under curve (AUC). Each measure supports the accuracy of the predictions (PCM: 52%, κ: 41%, OA: 82% and AUC: 71%) at an acceptable level. We delineate future flood risk by combining 100-year floodplain zones with 3 and 6 ft (0.9 and 1.8 m) SLR conditions utilizing a bathtub approach (Marcy et al. Reference Marcy, Brooks, Draganov, Hadley, Haynes and Herold2011), representing extreme cases targeting years 2050 and 2100 (see Kim and Newman Reference Kim and Newman2019 for the prediction model and SLR details). To identify eligible vacant properties in the most extreme SLR condition, we select properties that are within the 6 ft (1.8 m) SLR floodplains for further analysis.

Selecting vacant properties at watershed and site levels

To select vacant properties, we use appraised parcel data from ATTOM, a private data company focused on collecting property datasets from appraisal districts and cleaning them to better present accurate estimates of property value. We select occupancy codes ‘C1’ and ‘C3’, which are defined as ‘vacant lots/tracts’ and ‘vacant’, and cleaned the dataset by removing parcels that appear to have been mislabelled as vacant properties, such as those that intersect road segments and those with buildings. This procedure resulted in removing 2015 vacant parcels from the dataset. As is shown in Fig. 1, the selection criteria prioritized candidate vacant properties for buyouts if they intersected or were within both the existing and future floodplain and areas of predicted future urban development by the LTM between 2012 and 2040. Out of all of the 90 506 vacant parcels in the study area, 9495 fell within the selection criteria, with a mean parcel area of c. 9000 m2.

To select a watershed for impact analysis, we developed a ranking system of vacant land and flood risk for all USGS Hydrological Unit Code 12 (HUC-12) watersheds using a composite scoring method. Factors considered for the ranking include the cost of acquiring vacant land, the amount of FEMA flood claim pay-outs and areas of vacant land, wetlands, floodplains and predicted development in each watershed. The ranking revealed that the Clear Creek watershed has the highest intersection of vacancy and flood risk in Houston. This watershed is one of the most flood-impacted watersheds in the region (Brody et al. Reference Brody, Blessing, Sebastian and Bedient2012, Highfield et al. Reference Highfield, Brody and Blessing2014). The landscape is characterized by typical gulf coast environmental attributes, such as low topographical relief, wide floodplain boundaries and low soil permeability, all of which contribute to frequent flooding. Additionally, the watershed contains several large vacant lots that are attractive to development interests as they are suitable for conversion to residential development types.

Vacant sites impact analysis

The site-level impact analysis focuses on the south-western corner of the Clear Creek watershed (see Fig. 3(c)). This section was selected because it has over 93% of its vacant parcels within the SLR floodplain, it is located at the intersection of two major highways and it is adjacent to existing flood-prone residential development. It also falls within the Beltway region in the south-western part of the study area, which is projected to have the highest household population and number of jobs by the year 2045 (HGAC 2018). Because 92% of floodplain properties in the region are designated as single-family residential, we assumed that the development on these test sites would be single-family residential homes. We further assumed that development types that will occur on these vacant sites will be similar to those of surrounding neighbourhoods, so we duplicated the building characteristics of residential properties within 1.5 km of the vacant sites. This procedure resulted in creating 1103 residential homes on the test sites.

Fig. 3. (a) Top 10 Hydrological Unit Code 12 (HUC-12) watersheds (Clear Creek watershed highlighted); (b) Clear creek watershed showing candidate vacant properties; and (c) test site showing proposed residential properties on vacant land.

Estimating flood depth and property damages

We modelled inundation levels for different storm intensities from 2-year, 5-year, 10-year, 25-year, 50-year, 100-year and 500-year flood levels using a hydraulic model developed by the Hydrologic Engineering Center (HEC) of the US Army Corps of Engineers (USACE), HEC-RAS. This system was designed primarily to simulate hydraulic responses such as flow conveyance and water surface elevation within open-channelled water bodies, and it has been applied in previous flood risk and hazard studies in the area (Gori et al. Reference Gori, Blessing, Juan, Brody and Bedient2019, Juan et al. Reference Juan, Gori and Sebastian2020). We used the two-dimensional version (HEC-RAS 2D), which allows for the simultaneous simulation of riverine and pluvial flooding. We used the 2018 terrain obtained from the Houston-Galveston Area Council and the 2016 Land Use Land Cover (LULC) data obtained from the Multi-Resolution Land Characteristics Consortium. These datasets were used to represent the physical characteristics of the study area. For the flood impact analysis, 24-hour duration storms ranging from 2-year to 500-year return periods were simulated, reflecting the most recent rainfall statistics for the region published by National Oceanic and Atmospheric Administration (NOAA) Atlas 14 (NOAA 2018). We further used depressed areas in the test sites as locations for retention/detention ponds. These pond locations were determined by evaluating topographical conditions from the national hydrography dataset, inundation data from HEC-RAS 2D and aerial imagery. As is shown in Table 2, the first development option generated 1103 properties and assumed that the parcels would be maximized for development irrespective of ground elevation conditions. A second development type generated 918 properties and assumed that there would be no development in naturally depressed areas, which will serve as retention or detention ponds.

Table 2. Summary statistics of development with and without retention ponds.

BCR = benefit–cost ratio.

Next, we performed a flood impact analysis of the projected future development using the inundation values generated from the HEC-RAS 2D model and USACE depth-damage functions. We further estimated the average annual losses (AALs) for each of the proposed residential developments using the following formula:

$$AA{L_j} = \mathop \sum \limits_i Flood \ Los{s_{ij}} \times prob\left( i \right)$$
$$i = {\rm{2}}{\rm{.5, 10, 25, 50, 100 \,and\, 500}}$$

where i is the flood return period and j is the inundated residential property.

A similar approach for calculating AALs has been used in previous flood and buyout studies in the USA (see Tate et al. Reference Tate, Strong, Kraus and Xiong2016, Atoba et al. Reference Atoba, Brody, Highfield, Shepard and Verdone2020). To determine the present dollar value of the future damages, we apply a 2.8% discount rate through a 30-year period, a method used for discounting nominal flows often encountered in lease-purchase analysis such as a federally backed mortgage (OMB 2016). We calculate the BCR by dividing discounted AALs by the sum of the property acquisition cost, demolition cost and cost to maintain the lot for the next 30 years, similar to methods used in recent buyout studies (Tate et al. Reference Tate, Strong, Kraus and Xiong2016, Johnson et al. Reference Johnson, Wing, Bates, Fargione, Kroeger and Larson2020). We estimate demolition cost at US$7 per building square foot (Mobley et al. Reference Mobley, Atoba and Highfield2020) and US$565.76 per hectare as the yearly maintenance cost of vacant properties or patches of land after buyouts (BenDor et al. Reference BenDor, Salvesen, Kamrath and Ganser2020).

The benefit–cost analysis (BCA) was based on the following assumptions: (1) the costs and benefits are borne by local governments; (2) vacant properties would be bought out now, retained as open space and maintained by local governments; and (3) built properties represent a hypothetical residential development that would be bought out now all at once, retained as open space and maintained by local governments.

Results

Vacant property conservation

The selection process identified 9495 vacant parcels as candidate vacant buyouts in the city of Houston, totalling c. 85.4 km2 (8540 ha) (Fig. 3(a)). The mean parcel size is c. 9000 m2 (0.9 ha), while the average distances to parks and streams are c. 146 and 217 m, respectively. The Clear Creek watershed has 581 vacant candidate properties with a total area of over 17 km2, with almost 42% of the watershed predicted for future development. This watershed also has the largest amount of federal flood insurance claim payments under the National Flood Insurance Program (NFIP) in the county, with over US$780 million already spent by FEMA as pay-outs from previous flood events. These historically flooded neighbourhoods are in close proximity to flood-prone vacant properties, averaging a distance of c. 270 m for the watershed.

Avoided flood losses from sample vacant sites

Our analysis results in two buyout scenarios: (1) buyout of existing vacant properties; and (2) buyout of a hypothetical built residential development on the vacant site. The built buyout scenario generated large amounts of potential property losses from flooding if built on the vacant sites. The average property value of the proposed development is US$218 000–220 000. The combined AALs from flooding is US$1.3–7.5 million depending on the development type, with the mean AAL ranging from US$2810 to US$12 040. These results show that an annual loss of US$6.73–31.65 per square foot would occur for each residential building if the vacant sites are developed. Table 2 shows the results of the 100- and 500-year storms to provide some context to losses from the top two highest storm intensities. For the development without retention/detention ponds, inundation averages 0.96 and 1.38 m for the 100- and 500-year floods. Development with retention/detention ponds has lower inundation values; average inundation is c. 0.32 m for a 100-year flood and 0.46 m for a 500-year flood, while the maximum is 2.04 m for a 500-year flood.

Benefit–cost analysis of proposed development

The results of the BCA indicate that proactive buyouts of vacant properties are generally more favourable than buying out already built properties in flood-prone areas. For the development without retention/detention ponds, if the vacant properties are bought out now, there are savings of US$10.37 for every US$1 spent (Table 3). On the other hand, if the buyouts occur in the built parcel scenario, only US$0.61 is saved for every US$1 spent. For development with retention/detention ponds, the BCR is much less than US$1, implying that it is not beneficial to buy out the properties when they are already built. Conversely, there will be savings of c. US$1.86 for every US$1 spent if local governments prevent development on those parcels. In general, individual structures in development types where retention/detention ponds are used have lower mean BCR values than those without a retention/detention pond (Table 2). For both development types, only c. 10% of the properties in the proposed development are responsible for high flood losses, and they primarily influence the BCR values.

Table 3. Summary of flood losses and benefit–cost ratios (BCRs) for vacant and built buyout scenarios.

Discussion and conclusion

Our results show that if new development continues to follow existing spatial and development patterns, the future cost of buyouts would be significantly higher than preventing these developments from occurring in the first place. Based on our findings, there are two key benefits of vacant open space buyouts. First, methodologically, the geodesign process allows local officials to identify areas where flooding and future development are likely to have the highest impact. Our multi-spatial selection criteria at the watershed scale can help officials prioritize vacant land based on ecological and open space attributes to protect the most stressed watersheds in their community. Although our study focuses on only one major city and watershed in the USA, it highlights the importance of selecting properties at smaller scales that a nationwide study may not expose. This approach can complement a larger study, which makes it easier to draw conclusions and add a comparative component to the analyses. Our second major finding is that a proactive strategy of vacant buyouts significantly outweighs a reactionary, post-development approach. Notably, these benefits are c. 10 times greater than the cost of the buyouts and provide a large return on investment in a selected location. The benefits could even be higher if consideration is given to the cost of infrastructure to support residential neighbourhoods, as well as the monetary loss of abandoning or even demolishing such infrastructure post-buyouts.

Our results also suggest that, although there is a high BCR for vacant buyouts, a review of the individual BCRs shows that only c. 10% of these properties are responsible for the high BCRs. It is not surprising that only selected properties within a buyout area are cost-effective (see Kousky & Walls Reference Kousky and Walls2014, Tate et al. Reference Tate, Strong, Kraus and Xiong2016). This also reflects the existing scattershot pattern of buyouts rather than the contiguous acquisition of entire neighbourhoods (Freudenberg et al. Reference Freudenberg, Calvin, Tolkoff and Brawley2016). Buying out these properties when vacant highlights the potential of accruing additional benefits, such as the protection of ecological services, strategic clustering of open spaces, reduced maintenance cost and the avoided losses in constructing new infrastructure or maintaining/demolishing already-built infrastructure.

The results also suggest that avoided losses are lower when additional mitigation strategies such as retention and detention ponds are used in floodplain development. However, this result should be interpreted with caution. For example, our analysis shows that only c. 190 buildings are removed from the placement of retention/detention ponds, leaving the remaining 910 properties in vulnerable locations. Construction of compensatory storage in flood-prone areas may be insufficient for addressing flooding problems, as previous research shows that innovative mitigation measures in floodplains can still exacerbate flooding (Stevens et al. Reference Stevens, Song and Berke2010).

Although the proactive acquisition of vacant properties proposed in this study restricts development in selected floodplains, it is up to local officials to harness additional site-level analysis in their decision-making. This includes flood vulnerability studies, community visioning and neighbourhood-scaled design through citizen science to select the most optimal ways of conserving or developing vacant land (Hendricks et al. Reference Hendricks, Newman, Yu and Horney2018, Newman et al. Reference Newman, Shi, Yao, Li, Sansom and Kirsch2020). Additionally, local officials need to identify the best methodological approach for estimating future hydrology conditions. The bathtub approach in this study only estimates inundation areas under hydrostatic conditions, ignoring wave, erosion and other hydrologic impacts (Anderson et al. Reference Anderson, Fletcher, Barbee, Romine, Lemmo and Delevaux2018). Other models, such as those that account for groundwater inundation (Habel et al. Reference Habel, Fletcher, Rotzoll, El-Kadi and Oki2019), can be utilized in future research.

While this research highlights the economic benefits of preserving flood-prone vacant lands, it is important to emphasize that without the provision of incentives, landowners would sell the land to the highest bidder or succumb to market forces. Landowners are likely to hold onto their properties until they bow to the pressure of selling them to developers due to rising tax rates (Lee et al. Reference Lee, Newman and Park2018). It is also up to local governments to reconcile the loss of their tax base with the benefits accrued from avoided flood losses. However, previous research shows that vacant lots repurposed as open space can increase the property value of surrounding residential development (Crompton Reference Crompton2005, Kousky & Walls Reference Kousky and Walls2014), subsequently increasing the tax base for local communities. Local governments also have to determine how to reconcile the benefits of open space conservation with the opportunity cost of preventing developments in flood-prone areas, as well as the potential cost-sharing if they are to receive federal funding post-disaster. Moreover, there are several other options for open space conservation that can be employed by local governments without compromising tax benefits. These include policies such as the transfer of development rights, density bonuses, development clustering and so on (Brody et al. Reference Brody, Atoba, Sebastian, Highfield, Blessing, Mobley and Stearns2020), and these can be addressed in future research.

Our analysis also identifies vacant properties in low-flood-risk areas that can be prioritized for new development. These can serve as viable development alternatives and ease the negative fiscal impacts of developing in other municipalities (Fig. 1). Recent studies show that for existing and future buyout programmes, factors such as the distribution of buyouts, relocation within or outside the city and management practices of acquired properties will determine whether these programmes come at a profit or loss to local communities (Lamie et al. Reference Lamie, Campbell and Molnar2012, BenDor et al. Reference BenDor, Salvesen, Kamrath and Ganser2020). A vacant property buyout policy can complement existing built buyout programmes. The approach will serve the dual purpose of preventing floodplain development (thereby avoiding the need for future buyouts) and complementing the open spaces currently recovered from existing built buyout programmes. This proactive vacant buyout model combined with other planning tools and incentives can encourage landowners to support the efforts of local governments in open space conservation.

This paper evaluated the potential for cost savings when vacant lands are preserved as open spaces compared to developing and buying them post-development. The geodesign framework helps local decision-makers to proactively identify flood-prone vacant properties at multiple scales in their communities while incorporating land transformation datasets, high-resolution flood simulations and future SLR floodplain delineation. The impact analysis enables local officials to create a decision model based on a BCA to determine whether development should occur, the property should remain vacant or development should be moved to other suitable locations. By evaluating the economic implications of proactive open space acquisition, we hope to promote an open space conservation culture to decision-makers so that future development can be adjusted in scope and scale to best support the mitigation and adaptation of flood risk within these communities for better future benefits.

This study represents an important step in performing BCA for open space protection and flood risk reduction. Additional research and data analytics should be conducted to address some of the limitations in this study and further advance this area of enquiry. First, the costs and benefits of property acquisition are not only borne by local governments, but also by federal governments as well as individuals. A detailed statistical analysis can further reveal how these benefits are distributed across different governmental agencies where most buyout funding comes from. Second, because buyouts occur incrementally over time, future studies need to account for the dynamic nature of the built environment and property acquisition programmes. Third, although SLR floodplain delineations can identify properties with increased future flood risk, future studies should also assess how the inundation and damages of residential structures in these floodplains will be exacerbated by rising precipitation and flooding events due to climate change.

Supplementary material

To view supplementary material for this article, please visit https://doi.org/10.1017/S0376892921000059

Acknowledgements

We thank Nick Hernandez, Andrea Meier, Riley Boesiger and Amada Palecek for their expertise in developing the retention pond dataset.

Financial support

This study was supported by the US National Science Foundation (grant number 1545837).

Conflict of interest

None.

Ethical standards

None.

References

Anderson, T, Fletcher, C, Barbee, M, Romine, B, Lemmo, S, Delevaux, J (2018) Modeling multiple sea level rise stresses reveals up to twice the land at risk compared to strictly passive flooding methods. Scientific Reports 8: 114.CrossRefGoogle ScholarPubMed
Atoba, K, Brody, S, Highfield, W, Shepard, C, Verdone, L (2020) Strategic property buyouts to enhance flood resilience: a multi-criteria spatial approach for incorporating ecological values into the selection process. Environmental Hazards. Epub ahead of print, doi: 10.1080/17477891.2020.1771251.CrossRefGoogle Scholar
BenDor, TK, Salvesen, D, Kamrath, C, Ganser, B (2020) Floodplain buyouts and municipal finance. Natural Hazards Review 21: 04020020.CrossRefGoogle Scholar
Binder, SB, Baker, CK, Barile, JP (2015) Rebuild or relocate? Resilience and postdisaster decision-making after Hurricane Sandy. American Journal of Community Psychology 56: 180196.CrossRefGoogle ScholarPubMed
Binder, SB, Greer, A, Zavar, E (2020) Home buyouts: a tool for mitigation or recovery? Disaster Prevention and Management 29: 497510.CrossRefGoogle Scholar
Blanco, H, Alberti, M, Olshansky, R, Chang, S, Wheeler, SM, Randolph, J et al. (2009) Shaken, shrinking, hot, impoverished and informal: emerging research agendas in planning. Progress in Planning 72: 195250.CrossRefGoogle Scholar
Brody, S, Atoba, K, Sebastian, A, Highfield, W, Blessing, R, Mobley, W, Stearns, L (2020) A Comprehensive Framework for Coastal Flood Risk Reduction: Charting a Course Towards Resiliency. Washington, DC, USA: Island Press.Google Scholar
Brody, S, Blessing, R, Sebastian, A, Bedient, P (2012) Delineating the reality of flood risk and loss in Southeast Texas. Natural Hazards Review 14: 8997.CrossRefGoogle Scholar
Brody, S, Blessing, R, Sebastian, A, Bedient, P (2014) Examining the impact of land use/land cover characteristics on flood losses. Journal of Environmental Planning and Management 57: 12521265.CrossRefGoogle Scholar
Brody, SD, Highfield, WE, Blessing, R, Makino, T, Shepard, CC (2017) Evaluating the effects of open space configurations in reducing flood damage along the Gulf of Mexico coast. Landscape and Urban Planning 167: 225231.CrossRefGoogle Scholar
Conrad, DR, McNitt, B, Stout, M (1998) Higher Ground: A Report on Voluntary Property Buyouts in the Nation’s Floodplains: A Common Ground Solution Serving People at Risk, Taxpayers and the Environment. Reston, VA, USA: National Wildlife Federation.Google Scholar
Crompton, JL (2005) The impact of parks on property values: empirical evidence from the past two decades in the United States. Managing Leisure 10: 203218.CrossRefGoogle Scholar
Freudenberg, R, Calvin, E, Tolkoff, L, Brawley, D (2016) Buy-in for Buyouts: The Case for Managed Retreat from Flood Zones. Cambridge, MA, USA: Lincoln Institute of Land Policy.Google Scholar
Geoghegan, J (2002) The value of open spaces in residential land use. Land Use Policy 19: 9198.CrossRefGoogle Scholar
Gori, A, Blessing, R, Juan, A, Brody, S, Bedient, P (2019) Characterizing urbanization impacts on floodplain through integrated land use, hydrologic, and hydraulic modeling. Journal of Hydrology 568: 8295.CrossRefGoogle Scholar
Greer, A, Binder, SB (2017) A historical assessment of home buyout policy: are we learning or just failing? Housing Policy Debate 27: 372392.CrossRefGoogle Scholar
Habel, S, Fletcher, CH, Rotzoll, K, El-Kadi, AI, Oki, DS (2019) Comparison of a simple hydrostatic and a data-intensive 3D numerical modeling method of simulating sea-level rise induced groundwater inundation for Honolulu, Hawai’i, USA. Environmental Research Communications 1: 041005.CrossRefGoogle Scholar
Harter, JL (2007) Riparian restoration: an option for voluntary buyout lands in New Braunfels, TX. Texas State University-San Marcos [www document]. URL https://digital.library.txstate.edu/bitstream/handle/10877/3371/fulltext.pdf Google Scholar
Hausmann, A, Slotow, RO, Burns, JK, Di Minin, E (2016) The ecosystem service of sense of place: benefits for human well-being and biodiversity conservation. Environmental Conservation 43: 117127.CrossRefGoogle Scholar
Hendricks, MD, Newman, G, Yu, S, Horney, J (2018) Leveling the landscape: landscape performance as a green infrastructure evaluation tool for service-learning products. Landscape Journal 37: 1939.CrossRefGoogle ScholarPubMed
HGAC (2018) Houston Galveston Area Council. Annual Regional Growth Forecast [www document]. URL https://datalab.h-gac.com/rgf2018/ Google Scholar
Highfield, WE, Brody, SD, Blessing, R (2014) Measuring the impact of mitigation activities on flood loss reduction at the parcel level: the case of the clear creek watershed on the upper Texas coast. Natural Hazards 74: 687704.CrossRefGoogle Scholar
Highfield, WE, Brody, SD, Shepard, C (2018) The effects of estuarine wetlands on flood losses associated with storm surge. Ocean & Coastal Management 157: 5055.CrossRefGoogle Scholar
Jafari, M, Majedi, H, Monavari, SM, Alesheikh, AA, Kheirkhah Zarkesh, M (2016) Dynamic simulation of urban expansion based on cellular automata and logistic regression model: case study of the Hyrcanian region of Iran. Sustainability 8: 810.CrossRefGoogle Scholar
Johnson, KA, Wing, OEJ, Bates, PD, Fargione, J, Kroeger, T, Larson, WD et al. (2020) A benefit–cost analysis of floodplain land acquisition for US flood damage reduction. Nature Sustainability 3: 5662.CrossRefGoogle Scholar
Juan, A, Gori, A, Sebastian, A (2020) Comparing floodplain evolution in channelized and unchannelized urban watersheds in Houston, Texas. Journal of Flood Risk Management 13: e12604.CrossRefGoogle Scholar
Kim, Y, Newman, G (2019) Climate change preparedness: comparing future urban growth and flood risk in Amsterdam and Houston. Sustainability 11: 1048.CrossRefGoogle ScholarPubMed
Kim, Y, Newman, G (2020) Advancing scenario planning through integrating urban growth prediction with future flood risk models. Computers, Environment and Urban Systems 82: 101498.CrossRefGoogle ScholarPubMed
Kousky, C, Walls, M (2014) Floodplain conservation as a flood mitigation strategy: examining costs and benefits. Ecological Economics 104: 119128.CrossRefGoogle Scholar
Ku, C (2016) Incorporating spatial regression model into cellular automata for simulating land use change. Applied Geography 69: 19.CrossRefGoogle Scholar
Lamie, RD, Campbell, W, Molnar, W (2012) The fiscal-geographic nexus: an overview of fiscal impact assessment in local policy development. Applied Geography 32: 5460.CrossRefGoogle Scholar
Lee, J, Newman, G, Park, Y (2018) A comparison of vacancy dynamics between growing and shrinking cities using the land transformation model. Sustainability 10: 15131530.CrossRefGoogle ScholarPubMed
Losiri, C, Nagai, M, Ninsawat, S, Shrestha, R (2016) Modeling urban expansion in Bangkok metropolitan region using demographic–economic data through cellular automata-Markov chain and multi-layer perceptron-Markov chain models. Sustainability 8: 686.CrossRefGoogle Scholar
Mach, KJ, Kraan, CM, Hino, M, Siders, AR, Johnston, EM, Field, CB (2019) Managed retreat through voluntary buyouts of flood-prone properties. Science Advances 5: eaax8995.CrossRefGoogle ScholarPubMed
Maldonado, JK, Shearer, C, Bronen, R, Peterson, K, Lazrus, H (2013) The impact of climate change on tribal communities in the US: displacement, relocation, and human rights. In: Maldonado, JK, Colombi, B, Pandya, R (eds), Climate Change and Indigenous Peoples in the United States (pp. 93106). Cham, Switzerland: Springer.CrossRefGoogle Scholar
Maly, E, Ishikawa, E (2013) Land acquisition and buyouts as disaster mitigation after Hurricane Sandy in the United States. In: Proceedings of International Symposium on City Planning (Vol. 2013, pp. 1–18) [www document] URL http://www2.cpij.or.jp/com/iac/sympo/13/ISCP2013-8.pdf Google Scholar
Marcy, D, Brooks, W, Draganov, K, Hadley, B, Haynes, C, Herold, N et al. (2011) New mapping tool and techniques for visualizing sea level rise and coastal flooding impacts. Solutions to Coastal Disasters 2011: 474490.Google Scholar
Marino, E (2018) Adaptation privilege and voluntary buyouts: perspectives on ethnocentrism in sea level rise relocation and retreat policies in the US. Global Environmental Change 49: 1013.CrossRefGoogle Scholar
Mobley, W, Atoba, K, Highfield, W (2020) Uncertainty in flood mitigation practices: assessing the economic benefits of property acquisition and elevation in flood-prone communities. Sustainability 12: 2098.CrossRefGoogle Scholar
Newman, GD, Bowman, AO, Jung Lee, R, Kim, B (2016) A current inventory of vacant urban land in America. Journal of Urban Design 21: 302319.CrossRefGoogle Scholar
Newman, GD, Malecha, M, Yu, S, Qiao, Z, Horney, JA, Lee, J et al. (2019) Integrating a resilience scorecard and landscape performance tools into a geodesign process. Landscape Research 45: 6380.CrossRefGoogle ScholarPubMed
Newman, GD, Shi, T, Yao, Z, Li, D, Sansom, G, Kirsch, K et al. (2020) Citizen science-informed community master planning: land use and built environment changes to increase flood resilience and decrease contaminant exposure. International Journal of Environmental Research and Public Health 17: 486.CrossRefGoogle ScholarPubMed
NOAA (2017) National Center for Environmental Information Database [www document]. URL https://www.ncdc.noaa.gov/billions/events/US/1980-2017 Google Scholar
NOAA (2018) National Weather Service Hydrometeorological Design Studies Center Precipitation Frequency Data Server (PFDS) [www document]. URL https://hdsc.nws.noaa.gov/hdsc/pfds/ Google Scholar
OMB (2016) Office of Management and Budget. Discount rates for cost-effectiveness, lease purchase, and related analyses. In: Circular A-94 Appendix C. Washington, DC, USA: White House Office of Management and Budget.Google Scholar
Pagano, M, Bowman, A (2000) Vacant land in cities: an urban resource. Brookings Institution, Center on Urban and Metropolitan Policy [www document]. URL https://www.brookings.edu/wp-content/uploads/2016/06/paganofinal.pdf Google Scholar
Parisi, F, Lund, R (2008) Return periods of continental US hurricanes. Journal of Climate 21: 403410.CrossRefGoogle Scholar
Pijanowski, BC, Brown, DG, Shellito, BA, Manik, GA (2002) Using neural networks and GIS to forecast land use changes: a land transformation model. Computers, Environment and Urban Systems 26: 553575.CrossRefGoogle Scholar
Pijanowski, BC, Tayyebi, A, Doucette, J, Pekin, BK, Braun, D, Plourde, J (2014) A big data urban growth simulation at a national scale: configuring the GIS and neural network based land transformation model to run in a high performance computing (HPC) environment. Environmental Modelling & Software 51: 250268.CrossRefGoogle Scholar
Siders, AR (2019) Social justice implications of US managed retreat buyout programs. Climatic Change 152: 239257.CrossRefGoogle Scholar
Steinitz, C (2012) A Framework for Geodesign: Changing Geography by Design. Redlands, CA, USA: Esri Press.Google Scholar
Stevens, MR, Song, Y, Berke, PR (2010) New urbanist developments in flood-prone areas: safe development, or safe development paradox? Natural Hazards 53: 605629.CrossRefGoogle Scholar
Tate, E, Strong, A, Kraus, T, Xiong, H (2016) Flood recovery and property acquisition in Cedar Rapids, Iowa. Natural Hazards 80: 20552079.CrossRefGoogle Scholar
TWDB (2012) Population Projections Data [www document]. URL http://www.twdb.texas.gov/waterplanning/data/projections/2012/popproj.asp Google Scholar
Wing, OE, Bates, PD, Smith, AM, Sampson, CC, Johnson, KA, Fargione, J, Morefield, P (2018) Estimates of present and future flood risk in the conterminous United States. Environmental Research Letters 13: 034023.CrossRefGoogle Scholar
Winsemius, HC, Aerts, JCJH, van Beek, LPH, Bierkens, MFP, Bouwman, A, Jongman, B et al. (2016) Global drivers of future river flood risk. Nature Climate Change 6: 381385.CrossRefGoogle Scholar
Zavar, E (2015) Residential perspectives: the value of floodplain-buyout open space. Geographical Review 105: 7895.CrossRefGoogle Scholar
Zavar, E, Hagelman, RR III (2016) Land use change on US floodplain buyout sites, 1990–2000. Disaster Prevention and Management 25: 360374.CrossRefGoogle Scholar
Figure 0

Table 1. Vacant parcel selection and analysis through the geodesign process (adapted from Steinitz 2012, Newman et al. 2019).

Figure 1

Fig. 1. Selection framework for vacant properties in flood-risk areas. BCR = benefit–cost ratio.

Figure 2

Fig. 2. Hydrological Unit Code 12 (HUC-12) watersheds and vacant properties in Houston, Texas.

Figure 3

Fig. 3. (a) Top 10 Hydrological Unit Code 12 (HUC-12) watersheds (Clear Creek watershed highlighted); (b) Clear creek watershed showing candidate vacant properties; and (c) test site showing proposed residential properties on vacant land.

Figure 4

Table 2. Summary statistics of development with and without retention ponds.

Figure 5

Table 3. Summary of flood losses and benefit–cost ratios (BCRs) for vacant and built buyout scenarios.

Supplementary material: File

Atoba et al. supplementary material

Figures S1-S2 and Tables S1-S2

Download Atoba et al. supplementary material(File)
File 490.1 KB