1. Introduction
Severe weather storms and flooding are two of the most damaging and costliest natural hazards in the world (WMO 2021; Li et al. 2021). Severe storms have caused 577 232 deaths and USD 521 billion and floods have caused 58 700 deaths and USD 115 billion over the globe from 1970 to 2019 (WMO 2021). The warming atmosphere, in tandem with demographic and land-cover changes, will likely place more people at risk (Tellman et al. 2021), especially Indigenous people, who are often overlooked and marginalized.
Climate change has precipitated disparate risks, resulting in what has been termed “climate injustice,” where marginalized and socioeconomically disadvantaged populations disproportionately bear the risks (Smiley et al. 2022; Tate et al. 2021; Wing et al. 2022). For example, certain communities within the United States, including but not limited to Black communities, have been observed and will be likely to experience higher flood risks from climate change in comparison with national averages (Wing et al. 2022). Similar patterns were observed within some Hispanic communities in Texas during Hurricane Harvey (Smiley et al. 2022). Among various groups, Indigenous people often face pronounced challenges from climate change effects (Jantarasami et al. 2018). Despite their heightened vulnerability, they are often overlooked in federal disaster risk reduction policies, exacerbating their susceptibility to harm (Whyte 2017; Hadlos et al. 2022).
To help to identify vulnerable communities in the wake of climate change, researchers and stakeholders continue to develop and operate climate justice screening tools (Kuruppuarachchi et al. 2017). Since the justice40 initiative (viz., 40% of the overall benefits of federal investments contribute to disadvantaged communities) put forward by President Biden, a climate justice screening tool has been published by the Council on Environmental Quality (https://screeningtool.geoplatform.gov). However, this product is riddled with several issues, one of which is a bias toward urban areas (Mullen 2022). It integrates all natural hazards into impact-based assessment—expected agriculture loss, expected building loss rate, and expected population loss rate, while not specifying different kinds of natural hazards. In addition, its climate change assessment is based on historical observations, but future climate projection is not readily incorporated. Alternatively, climate projection has far-reaching suggestions for climate adaptation and mitigation, as some of the natural hazard (e.g., flooding) generating mechanisms are likely to change (Li et al. 2022a).
For those climate justice tools that feature projected information, global climate models (GCMs) are used to infer environmental risks in the future, even though they inevitably underestimate the magnitude of risk and do not provide fine-scale information (Schewe et al. 2019; Slingo et al. 2022). Such model results will subsequently bias the decision-making process and generate inadequate plans. With the rapid growth of available models and data, there is a need to update existing climate/environmental justice screening tools so that they are more suitable for future risk assessments.
Native Americans (or Indigenous people of North America), unlike other communities of color, have lived in certain areas of their homelands for hundreds or thousands of years. They have unique lived experiences, cultures, and spiritual practices that have evolved and adapted to climate and environmental change (Wildcat 2010). Global Indigenous wisdom formed from environmental changes, accumulated over generations, is particularly insightful when planning climate adaptation (Chief et al. 2016; Jantarasami et al. 2018). Indigenous people residing along the Mekong River predict the weather with their observations, such as the shadow of water, stroking the lemongrass leaves, and animal behaviors, and then decide on what to harvest (Pauli et al. 2021). Their nature-based solutions can provide deep insight into designing climate mitigation strategies. Such solutions include, but are not limited to, 1) leveraging local knowledge such as geography, soil quality, and groundwater levels to inform water resources management plans; 2) implementing ad hoc farming techniques to mitigate soil erosion and water runoff; 3) adopting sustainable livestock grazing practices to prevent overgrazing and land degradation; and 4) integrating traditional Indigenous calendars with state-of-the-art numerical models for weather prediction. However, in the United States, certain tribal nations such as the Otoe-Missouria Tribe in Oklahoma lack institutional and economic resources to properly assess climate risks for upcoming plans. Research institutes and the federal government play a critical role in helping tribal nations to prepare for future climate risks. The U.S. Environmental Protection Agency (EPA) was one of the first federal agencies to interact with tribal governments to protect their health and the environment. Also, when developing climate justice screening tools or climate risk assessment, it is crucial to engage Native Americans to encapsulate their Indigenous knowledge into the process.
Given the understudied and inequitable climate risk to Native Americans, our study aims to assess future climate extremes (focusing on pluvial events) and their impacts based both on climate and demographic changes. We apply a newly designed hazard–exposure–vulnerability–risk framework to investigate current risks and future changes through a coupled model chain: GCM–regional climate model (RCM)–hydrologic model–risk assessment model. This framework enables a physics-based and high-resolution climate assessment for tribal jurisdictions in Oklahoma. The objectives of this study are threefold: 1) projecting future climate extremes (i.e., heavy rainfall, flooding, and flash flooding) in Oklahoma, 2) evaluating the impact of climate and demographic changes on Native Americans through the hazard–exposure–vulnerability framework, and 3) developing risk maps for Native Americans in both contemporary and future climates. The study’s intent is to assist tribal nations in Oklahoma in assessing future risks, on which disaster risk reduction plans could be built. It is hoped that this study can further call attention to marginalized Native American communities and advocate for intersectoral partnerships that work with tribal nations and implement nature-based solutions stemming from Indigenous knowledge.
2. Study area, data, and methods
a. Study area
Our research uses the state of Oklahoma as a test case to investigate the climate risk for Native American residents. Oklahoma has the second largest American Indian and Alaska Native population percentage (16%) in the United States, next only to Alaska (22%). Figure 1 shows its topography (Fig. 1a), land use/land cover (Fig. 1b), and Native American jurisdiction in Oklahoma (Fig. 1c). The average elevation is 395 m, with a standard deviation of 234 m. It features a steep slope, descending from the highlands in the northwest to the plains in the southeast (Fig. 1a). This state is mostly covered by grassland (29.2%), followed by deciduous forest (21.1%), pasture (15.7%), cultivated crops (13.3%), as shown in Fig. 1b. There are 39 federally recognized tribal nations within the state, with their jurisdiction delineated in Fig. 1c. Among them, the Five Tribes, known historically as the “Five Civilized Tribes,” are the major tribal nations (sorted by population size), which include the Cherokee Nation, Choctaw Nation, Muscogee (Creek) Nation, Chickasaw Nation, and Seminole Nation. Three major rivers that run across this state are the Canadian River, Arkansas River, and Red River.
Maps of study area (a) topography (source: NASA Shuttle Radar Topography Mission), (b) land use/land cover (source: National Land Cover Database), and (c) federal recognized Native American jurisdiction (source: Bureau of Indian Affairs).
Citation: Weather, Climate, and Society 16, 1; 10.1175/WCAS-D-23-0005.1
Climate change shows how, over time, inequality and injustice continue to plague Native Americans in Oklahoma. Land theft and dispossession have contributed to many of the vulnerabilities of Native Americans for generations, and they have shaped the lands on which they have lived under duress (Justice and O’Brien 2022).
b. Data
1) Native American population and growth
2) Social vulnerability
The social vulnerability index (SVI) is a composite index from the Centers for Disease Control and Prevention (CDC), composed of four parts: 1) socioeconomic status (including “below poverty,” “unemployed,” “income,” and “no high school diploma”); 2) household composition and disability (including “aged 65 or older,” “aged 17 or younger,” “civilian with a disability,” and “single-parent households”); 3) minority status and language (including “minority” and “aged 5 or older who speaks English less than well”); and 4) housing type and transportation (including “multi-unit structures,” “mobile homes,” “crowding,” “no vehicle,” and “group quarters”) (CDC/Agency for Toxic Substances and Disease Registry 2022). This composite index is computed by 1) depositing and populating all socioeconomic indicators into a common database and 2) ranking them with respect to percentiles from 0 to 1 (Flanagan et al. 2011). (Detailed documentation is available online: https://www.atsdr.cdc.gov/placeandhealth/svi/documentation/SVI_documentation_2018.html; last accessed in September 2022.) It reflects the degree to which a community responds to an emergent disaster (including natural hazards and human diseases). It has been previously used in flood studies to assess community vulnerability (Alipour et al. 2020). Figure 2 depicts the state SVI distribution as a spatial map (Fig. 2a) and among different communities (Fig. 2b). A higher value of SVI indicates that the regions is more vulnerable. Highly vulnerable communities mainly reside in the northeastern part of the state, where most Native Americans reside (i.e., the Cherokee Nation). The SVI values among different communities in Oklahoma show that the Native American community is the most vulnerable group in Oklahoma, which is markedly above the state average.
SVI for the local community in Oklahoma: (a) map of the spatial distribution of SVI and (b) SVI grouped by ethnicity.
Citation: Weather, Climate, and Society 16, 1; 10.1175/WCAS-D-23-0005.1
c. Regional climate model + hydrologic simulation
To investigate climate risks, we use a dynamically downscaled GCMs to predict changes in the atmosphere. The dynamically downscaled climate data can better represent precipitation and related extreme events than GCMs (Slingo et al. 2022). The recent CONUS-I dataset (Liu et al. 2017), produced at hourly and 4-km resolution, has become one of the highest resolution and continuous continental climate simulations, which has empowered a wide range of regional climate studies, including extreme precipitation (Prein et al. 2017), snowmelt (Musselman et al. 2018), flash flood-producing storms (Dougherty and Rasmussen 2020; Li et al. 2022b), and flood frequency analysis (Yu et al. 2020). The contemporary climate from 2000 to 2013 was simulated to contrast future changes, denoted as CTL. It was run using the Weather and Research and Forecasting (WRF) Model. The ERA-interim reanalysis data were used to provide initial and boundary conditions (Liu et al. 2017). For the counterpart future climate, we use the “pseudo-global-warming” (PGW) data, as simulated along with the CONUS-I dataset. The essence of PGW is that some of the climate states were perturbed with an ensemble mean of an array of CMIP5 models (Schär et al. 1996). Due to the unavailability of CMIP6 models at the time of initiating the model downscaling and considering the extensive computational requirements, the CMIP5 models were used. These were applied in a high-emissions scenario, consistent with the approach taken for the population growth data. Those perturbed states include wind speed, surface temperature, sea surface temperature, humidity, geopotential, and so on. The represented future climate is from the end of this century (2071–2100). The advantage of such a method is that it bypasses the most uncertain dynamic changes in climate model simulation, while only allowing the assessment of the thermodynamic changes (Liu et al. 2017).
Although WRF simulation includes the land surface component, streamflow is not an available output, as the routing process is absent. In this study, we simulate overland streamflow using the Ensemble Framework for Flash Flood Forecast (EF5). This framework is capable of producing flash-flood-scale hydrologic prediction across a continent, which is currently operated by the National Weather Service for real-time flash-flood monitoring (Flamig et al. 2020; Gourley et al. 2017). The Coupled Routing and Excess Storage (CREST) model was used as the hydrologic model within the framework (Wang et al. 2011; Li et al. 2023). This model has been primarily in global flood monitoring systems (Wu et al. 2012). We output streamflow at 1-km spatial resolution and hourly frequency under two climates.
Figure 3 illustrates the coupled climate and hydrologic simulation as an integrated system to assess the impact of climate change on heavy rainfall and floods. We have applied the same framework in previous studies to quantify future climate extremes across the continental United States (Li et al. 2022a,b). It has shown that extreme event rainfall will become 24% heavier, and future floods are becoming 7.9% flashier—higher flood magnitude Qp and shorter flood peaking time Tp (Li et al. 2022b). Both heavy rainfall and floods are becoming more frequent, wider spread, yet less seasonal (Li et al. 2022a).
An illustration of coupled climate–hydrologic models and their produced hydrograph at the basin outlet. The hydrograph for the future climate is flashier, as dictated by a shorter peaking time Tp and a higher flood peak Qp.
Citation: Weather, Climate, and Society 16, 1; 10.1175/WCAS-D-23-0005.1
d. Quantifying hazards, exposures, and hazards
The climate extremes are analyzed in this study for Native Americans as hazards, which indicates the probability and intensity of climate extremes. We focus on three pluvial hazards: heavy rainfall, floods, and flash floods. For heavy rainfall, we first accessed the intensity-duration-frequency data from NOAA Atlas 14 (Miller et al. 1973). It is one of the most commonly used heavy rainfall threshold data sources in the United States (Li et al. 2022b; Wright et al. 2019). For this study, we chose a 2-yr and 12-h event rainfall value as a threshold to count the rainfall occurrences. To account for flood risks, we extracted streamflow from hydrologic simulation and calculated 2-yr streamflow as thresholds for overbanking (He and Wilkerson 2011; Li et al. 2022b). Flood occurrences are computed with hourly streamflow exceeding thresholds. The flash-flood risk is quantified with the “flashiness” index, as shown in Fig. 3. It is a term that takes into account both the flood magnitude Qp and flood peaking time Tp to illustrate how steep the hydrograph is (the steeper, the flashier). It is normalized by the catchment drainage area (FAC) to reflect that small river reaches are more susceptible to flash floods (Saharia et al. 2017). All three indicators are mapped to land-tract levels to be consistent with socioeconomic data using the average operator.
3. Results
a. Population growth for Native Americans
Figure 4 shows the population distribution for Native Americans (Fig. 4a) and its future changes (Fig. 4b) on land-tract levels. In 2013, the total Native American population in Oklahoma was estimated to be 288 801. This number is projected to become 603 034 by the end of this century, an average increase of 108.8%. The total population of Oklahoma was close to 4 million, and that number is expected to increase by 194.8% in 2100. The total population increases at a rate faster than Native Americans in Oklahoma. Most of the Indigenous population reside in northeast Oklahoma in the Cherokee and Muscogee (Creek) Nations, with a cluster of population exceeding 1000. For these regions, there will be subtle decreases in population by the end of this century. The currently less-populated regions in the west will have a large increase in population in the future.
Maps of (a) population of Native Americans in 2013, (b) population growth of Native Americans by 2100, (c) total population in 2013, and (d) total population growth by 2100.
Citation: Weather, Climate, and Society 16, 1; 10.1175/WCAS-D-23-0005.1
b. Heavy rainfall, flood, and flash-flood hazards
The climate hazards shown in Fig. 5 depict the spatial distribution of hazardous regions in Oklahoma. First, we see a widespread increase in heavy rainfall, floods, and flash floods at the end of this century, with 81.4%, 137.3%, and 10.3% changes, respectively. Changes in heavy rainfall are almost uniform across the state (Fig. 5a). The increase in heavy rainfall is primarily due to the thermodynamic changes of future storms with higher environmental temperature, water vapor, and humidity, inferred from the Clausius–Clapeyron equation (i.e., saturation vapor pressure increases by 7% °C−1 of warming) (Allen and Ingram 2002). The recent IPCC AR6 report describes this region as a “very likely” increase in extreme precipitation (Caretta et al. 2022). Notably, this value calculated by RCM seems to be greater than some studies using the GCM model (Tabari 2020; Swain et al. 2020). We can attribute this to the representativeness of heavy precipitation in different climate models. The deep convection process in the convection-permitting RCM has been resolved instead of parameterized (within GCMs) with reduced errors and uncertainties, although with some caveats (Prein et al. 2015; Slingo et al. 2022). In short, it is justified that our model predicts more extreme values than coarse-resolution GCMs (Prein et al. 2013).
Maps of (top) hazards in current climate and (bottom) their percent changes for (a) heavy rainfall, (b) floods, and (c) flash floods.
Citation: Weather, Climate, and Society 16, 1; 10.1175/WCAS-D-23-0005.1
Flooding is a multifaceted disaster, encompassing atmospheric processes and land surface conditions (Li et al. 2022a; Merz et al. 2021). Floods over Oklahoma have a higher average increase than rainfall, due to conducive land surface conditions such as anomalously wet antecedent soil moisture and groundwater. Greater changes are present in small river reaches, relative to large rivers, because they are susceptible to climate change. Flash floods, within several hours of the onset of heavy rainfall, show a different pattern. Regions with above-normal increases in flashiness are located along the Canadian River, where we expect to experience higher flood peaks and shorter flood peaking times. It poses threats to Oklahoma City, north of the Canadian River, bearing greater flash-flood risks.
c. Exposed Native Americans to climate extremes
The exposed Native American population to the three types of climate extremes is partitioned into four components (to attribute exposure changes to demography and climate): 1) population in 2013 and climate in 2013 (baseline), 2) population in 2100 and climate in 2013 (demographic change), 3) population in 2013 and climate in 2100 (climate change), and 4) population in 2100 and climate in 2100 (combined). As shown in Fig. 6, the future exposed Native American population (combined), unsurprisingly, is greater than the current condition (baseline), with values varying with hazard types. For heavy rainfall (floods/flash floods), the future exposed population is about 3.8 (4.9/2.4) times of currently exposed population. While attributed to climate change and demographic changes, exposed populations to different types of extremes are clear cut. Of three climate extremes, heavy rainfall and flash floods show that demographic changes (24.0% and 34.8%) outweigh climate changes (21.3% and 15.0%), meaning that population growth contributes more to increased exposure than climate change. In particular, exposures due to flash floods are primarily driven by population growth, as the percent changes in flash floods only average 10.3% (Fig. 5c). On the contrary, floods show that exposed populations due to climate changes outweigh demographic changes. Exposures due to climate change (23.9%) are 2.5 times higher than baseline (9.65%). With that being said, if we keep either current population or climate, future risks of heavy rains and floods raised by climate change will be halved. If population stays unchanged, risk of flash floods remains at current level; however, the risk of flash floods is almost tripled by climate change.
Contributions of climate and demographic changes to climate extremes: (a) heavy rainfall, (b) floods, and (c) flash floods. “Pop” and “Clim” are short for population and climate, respectively; for example, “Pop2013” and “Clim2013” refer to the Native American population in 2013 and climate extremes in 2013 in Oklahoma.
Citation: Weather, Climate, and Society 16, 1; 10.1175/WCAS-D-23-0005.1
d. Risk map
As introduced in section 2d, we formulate the risk factor as a product of hazards, exposures, and vulnerabilities [Eq. (2)]. To produce consistent risk maps for three climate extremes, we normalize the risk factor into the range of 0–1. This is done by calculating the ranks of risk factors and taking their percentages. Figure 7 shows risk maps for these climate extremes and their associated percent changes in a future scenario. In a contemporary climate, three maps show similar hotspot regions: northeast, southeast, and several land tracts in the southwest. The risks are becoming greater in the late century, though to different degrees. Flood risk will increase by 632.6% in the future; and heavy rainfall risks will increase by 501.1%. There is a moderate increase in flash-flood risk (296.4%), in comparison with the other two, but it still has profound socioeconomic impacts on Indigenous communities, especially for those who reside along rivers, that threaten their peoplehood and identity (Jarratt-Snider and Nielsen 2020; Kimmerer 2013). Regions with greater risk increases are found to be in northwestern Oklahoma, coinciding with a large increase in the Native American population in the future.
Maps of climate risks from (a) heavy rainfall, (b) floods, and (c) flash floods for two periods: (top) 2000–13 and (bottom) 2070–99 (their associated percent changes) under a high-emission scenario (RCP8.5).
Citation: Weather, Climate, and Society 16, 1; 10.1175/WCAS-D-23-0005.1
We further break down different tribal nations to inspect community risks, as shown in Fig. 8. Over one-half of the risk changes are from increases in population growth for heavy rainfall and flash floods. The Iowa Nation tops in all three measured risk changes (i.e., heavy rainfall, 2-yr floods, and flash floods), both of which are projected to become more than 10 times as high. It is mainly attributed to the population growth in the Iowa Nation. The Cheyenne–Arapaho Nation and Muscogee (Creek) Nation are ranked in second and third place, respectively.
Changes in climate risks, grouped by tribal nations and attributed to climate changes (red) or demographic changes (blue), for (a) heavy rainfall, (b) floods, and (c) flash floods. Here, CP is Citizen Potawatomi, KCA is Kiowa–Comanche–Apache, FSA is Fort Sill Apache, and CWD is Caddo–Wichita –Delaware.
Citation: Weather, Climate, and Society 16, 1; 10.1175/WCAS-D-23-0005.1
e. Comparing climate risks of Native Americans with those of all ethnic groups
The risks of heavy rainfall, 2-yr flooding, and flash flooding to Native Americans are compared with the risks to all ethnic groups in Oklahoma in Fig. 9. The median increases in rainfall risks for Native Americans (501.1%) are 68.0% higher than the general population over the state (298.2%). The increases in 2-yr flooding and flash flooding risks for Native Americans are 64.3% and 64.0% higher than the those of all ethnic groups, respectively. The major spatial differences are found to be in the panhandle area and western border of Oklahoma. The likely reason for this observed increase in risks is driven by the projected growth in Native American population, as compared with other ethnic groups (Fig. 4). Put differently, climate extremes are likely to exacerbate social injustice issues for Native Americans in Oklahoma.
Maps of changes in (a) rainfall, (b) 2-yr flood, and (c) flash-flood risks for (top) Native Americans and (bottom) all ethnic groups.
Citation: Weather, Climate, and Society 16, 1; 10.1175/WCAS-D-23-0005.1
4. Discussion
In this study, we used the dynamically downscaled climate simulation data and coupled it with a high-resolution hydrologic model to investigate changes in future climate extreme risks (i.e., heavy rainfall, floods, flash floods). Since climate models inevitably bear large uncertainties due to internal variabilities, model initialization, and numerical solutions, the projected rainfall changes and its propagated uncertainties to surface hydrology are likely to be affected. One of the limitations of the pseudo-global-warming scheme to represent future climate is that it only allows thermodynamic change, which becomes the main driver. Even so, Liu et al. (2017) stated that the dynamical changes from GCMs are weak relative to thermodynamic changes. Cheng et al. (2018) corroborated that thermodynamic changes outweigh dynamic changes in Oklahoma for convective rains. But the advantage of the pseudo-global-warming is twofold: 1) it endorses event-to-event comparison, because past events are represented in the future simulation; 2) it eliminates the most uncertain changes in climate simulation–dynamic changes. From our results, we showed that heavy rainfall is projected to increase by 81.4% in the future climate in Oklahoma. This value is subject to chosen thresholds since there are different ways to define heavy rainfall (Li et al. 2022b). For floods, hydrologic simulation also has errors originating from an inaccurate representation of antecedent land surface conditions, water cycle, and routing methods. We have previously validated our results across the continental United States by comparing them with data from stream gauges (Li et al. 2022b). Still, some river reaches, especially those by human interference, are challenging to simulate. To incorporate uncertainty quantification, future work can display the uncertainty range by embracing multimodel ensembles.
Another source of uncertainty in risk assessment is demographic changes. We project the Native American population in an aggressive manner, to match the “business as usual” scenario. Errors are larger for peoples with less population when a hindcast was performed. The simple redistribution method [see Eq. (1)] generally assumes that Native Americans, Asian Americans, and other minor ethical groups will grow at a similar rate in Oklahoma. We validated such method to infer 2020 Native American population with data in 2013. We found the error to be 2.7%, within the 5% range. Additionally, the projected data cannot account for dynamic population migration. But from a qualitative standpoint, some studies deliver similar results in comparing climate and demographic changes. For instance, Swain et al. (2020) showed the increase in flood exposure is more attributed to population growth than climate change. Historical observations show that the exposed population and assets have increased more rapidly than the overall population (Kundzewicz et al. 2014; Tellman et al. 2021). In addition, in this study, we made an assumption that social vulnerability will remain at the same level in the future, which can vary based on policies, governance, and other dynamics. Discussing how those factors lead to different socioeconomic pathways and hence the change in social vulnerability is beyond the scope of this study.
Last, there are unavoidable threshold setting and aggregation issues that could affect the results of this study. We chose the 2-yr streamflow and 12-h rainfall rates as the thresholds to determine flood and heavy rainfall events, based on the common metrics used in literature (Sampson et al. 2015; Burn et al. 2011; Li et al. 2022a). The reason for a relatively low return periods is that our downscaled climate data and simulated hydrologic data have 10-yr duration. The NOAA Atlas 14 has existing heavy rainfall thresholds for 2-yr and 12-h rainfall events. The aggregating method (i.e., averaging from grid cells to land tracts) is also likely to impose changes in our results (de Moel et al. 2015). Since we compare the relative changes between current and future climates, the effects by different thresholds and aggregating methods are expected to be marginal.
We acknowledge that this study is only a first step to address climate injustice issues for Native Americans. More actionable plans integrating Indigenous knowledge and scientific knowledge with nature-based solutions are needed. This article does not yet propose such solutions and intricate details, because each tribal nation is distinct even if they share similar experiences that connect them as “Native American.” Notably, we have established two-way collaboration and communication with Otoe-Missouria Tribe to work on this project, and part of the results from this study has been presented to tribal communities to support Otoe-Missouria efforts to adapt and mitigate the climate change impacts. We have hosted several meetings to learn oral histories and Indigenous traditional knowledge of lands and experiences with flooding and related dynamics for the Native Nation’s needs. Future work will focus on proposing climate mitigation and adaptation plans for Otoe-Missouria Tribe.
5. Conclusions
This study projects three types of future climate risks to Native Americans in Oklahoma through the assessment of climate and demographic changes. We highlight that Native Americans are facing a looming climate crisis on top of structural injustice issues. Some conclusions are drawn below:
-
Native Americans are the most vulnerable communities in Oklahoma among five major communities (vulnerability in a decreasing order: Native American > Hispanic > African > Asian > White).
-
Heavy rainfall and 2-yr flood risks are projected to be much greater in the future (increase by 501.1% and 632.6%) for Native Americans, driven by climate and demographic changes. Flash-flood risk has a moderate increase (296.4%).
-
Native Americans bear 68.0%, 64.3%, and 64.0% higher risks in heavy rainfall, 2-yr flooding, and flash flooding than general population in Oklahoma.
-
In comparing climate and demographic changes, it is seen that population growth generally leads to greater climate hazard risks than does climate change.
-
Tribal nations such as the Iowa Nation are projected to have 10 times more population, resulting in great exposures to climate extremes.
Facing increased climate risks in the future, Native Americans deserve greater attention to address climate injustice issues as a whole with the acknowledgment of their distinct relationships to their homelands as sovereign peoples. In a future study, we hope to conduct a comprehensive flood risk analysis for Native Americans in their jurisdiction, which engages Indigenous community partnerships to seek solutions to the interconnected impacts of vulnerability to flooding and climate injustice on Indigenous people in Oklahoma. It is also critical to understand Indigenous communities in urban and suburban settings outside of tribal nation territories.
Acknowledgments.
We acknowledge the Otoe-Missouria Tribe community for their knowledge sharing and University of Oklahoma Institute for Community and Society Transformation seed funding to support this study. We thank the National Center for Atmospheric Research (NCAR) for making the high-resolution climate simulation available to the public. We thank Editor Tanya Spero and three anonymous reviewers for their constructive comments and suggestions to improve the quality of this paper. We declare that there is no conflict of interest.
Data availability statement.
The CONUS-I climate simulation is accessed at NCAR Research Data Archive (https://rda.ucar.edu/datasets/ds612.0/). The future projection of county-level U.S. population by ethnicity is acquired online (https://doi.org/10.7927/dv72-s254). The EF5 model code is publicly available (http://ef5.ou.edu/).
REFERENCES
Alipour, A., A. Ahmadalipour, P. Abbaszadeh, and H. Moradkhani, 2020: Leveraging machine learning for predicting flash flood damage in the southeast US. Environ. Res. Lett., 15, 024011, https://doi.org/10.1088/1748-9326/ab6edd.
Allen, M. R., and W. J. Ingram, 2002: Constraints on future changes in climate and the hydrologic cycle. Nature, 419, 224–232, https://doi.org/10.1038/nature01092.
Burn, D. H., R. Mansour, K. Zhang, and P. H. Whitfield, 2011: Trends and variability in extreme rainfall events in British Columbia. Can. Water Resour. J., 36, 67–82, https://doi.org/10.4296/cwrj3601067.
Caretta, M. A., and Coauthors, 2022: Water. Climate Change 2022: Impacts, Adaptation and Vulnerability, H.-O. Pörtner et al., Eds., Cambridge University Press, 551–712, https://doi.org/10.1017/9781009325844.006.
CDC/Agency for Toxic Substances and Disease Registry, 2022: CDC/ATSDR Social Vulnerability Index 2014 Database Oklahoma. Centers for Disease Control and Prevention/Agency for Toxic Substances and Disease Registry/Geospatial Research, Analysis, and Services Program, accessed 15 September 2022, https://www.atsdr.cdc.gov/placeandhealth/svi/data_documentation_download.html.
Cheng, L., M. Hoerling, L. Smith, and J. Eischeid, 2018: Diagnosing human-induced dynamic and thermodynamic drivers of extreme rainfall. J. Climate, 31, 1029–1051, https://doi.org/10.1175/JCLI-D-16-0919.1.
Chief, K., A. Meadow, and K. Whyte, 2016: Engaging southwestern tribes in sustainable water resources topics and management. Water, 8, 350, https://doi.org/10.3390/w8080350.
de Moel, H., B. Jongman, H. Kreibich, B. Merz, E. Penning-Rowsell, and P. J. Ward, 2015: Flood risk assessments at different spatial scales. Mitigation Adapt. Strategies Global Change, 20, 865–890, https://doi.org/10.1007/s11027-015-9654-z.
Dougherty, E., and K. L. Rasmussen, 2020: Changes in future flash flood–producing storms in the United States. J. Hydrometeor., 21, 2221–2236, https://doi.org/10.1175/JHM-D-20-0014.1.
Flamig, Z. L., H. Vergara, and J. J. Gourley, 2020: The Ensemble Framework for Flash Flood Forecasting (EF5) v1.2: Description and case study. Geosci. Model Dev., 13, 4943–4958, https://doi.org/10.5194/gmd-13-4943-2020.
Flanagan, B. E., E. W. Gregory, E. J. Hallisey, J. L. Heitgerd, and B. Lewis, 2011: A social vulnerability index for disaster management. J. Homeland Secur. Emerg. Manage., 8, 3, https://doi.org/10.2202/1547-7355.1792.
Gourley, J. J., and Coauthors, 2017: The FLASH project: Improving the tools for flash flood monitoring and prediction across the United States. Bull. Amer. Meteor. Soc., 98, 361–372, https://doi.org/10.1175/BAMS-D-15-00247.1.
Hadlos, A., A. Opdyke, and S. A. Hadigheh, 2022: Where does local and indigenous knowledge in disaster risk reduction go from here? A systematic literature review. Int. J. Disaster Risk Reduct., 79, 103160, https://doi.org/10.1016/j.ijdrr.2022.103160.
Hauer, M. E., 2019: Population projections for U.S. counties by age, sex, and race controlled to shared socioeconomic pathway. Sci. Data, 6, 190005, https://doi.org/10.1038/sdata.2019.5.
He, L., and G. V. Wilkerson, 2011: Improved bankfull channel geometry prediction using two-year return-period discharge. J. Amer. Water Resour. Assoc., 47, 1298–1316, https://doi.org/10.1111/j.1752-1688.2011.00567.x.
Jantarasami, L. C., and Coauthors, 2018: Tribes and indigenous peoples. Impacts, Risks, and Adaptation in the United States: Fourth National Climate Assessment, D. R. Reidmiller et al., Eds., Vol. II, U.S. Global Change Research Program, 572–603, https://doi.org/10.7930/NCA4.2018.CH15.
Jarratt-Snider, K., and M. O. Nielsen, Eds., 2020: Indigenous Environmental Justice. University of Arizona Press, 229 pp.
Justice, D. H., and J. M. O’Brien, Eds., 2022: Allotment Stories: Indigenous Land Relations under Settler Siege. University of Minnesota Press, 376 pp.
Kimmerer, R. W., 2013: Braiding Sweetgrass: Indigenous Wisdom, Scientific Knowledge, and the Teachings of Plants. Milkweed Editions, 390 pp.
Kron, W., 2005: Flood risk = Hazard• values• vulnerability. Water Int., 30, 58–68, https://doi.org/10.1080/02508060508691837.
Kundzewicz, Z. W., and Coauthors, 2014: Flood risk and climate change: Global and regional perspectives. Hydrol. Sci. J., 59 (1), 1–28, https://doi.org/10.1080/02626667.2013.857411.
Kuruppuarachchi, L. N., A. Kumar, and M. Franchetti, 2017: A comparison of major environmental justice screening and mapping tools. Environ. Manage. Sustainable Dev., 6, 59–71, https://doi.org/10.5296/emsd.v6i1.10914.
Li, Z., M. Chen, S. Gao, J. J. Gourley, T. Yang, X. Shen, R. Kolar, and Y. Hong, 2021: A multi-source 120-year U.S. flood database with a unified common format and public access. Earth Syst. Sci. Data, 13, 3755–3766, https://doi.org/10.5194/essd-2021-36.
Li, Z., S. Gao, M. Chen, J. J. Gourley, and Y. Hong, 2022a: Spatiotemporal characteristics of US floods: Current status and forecast under a future warmer climate. Earth’s Future, 10, e2022EF002700, https://doi.org/10.1029/2022EF002700.
Li, Z., S. Gao, M. Chen, J. J. Gourley, C. Liu, A. F. Prein, and Y. Hong, 2022b: The conterminous United States are projected to become more prone to flash floods in a high-end emissions scenario. Commun. Earth Environ., 3, 86, https://doi.org/10.1038/s43247-022-00409-6.
Li, Z., and Coauthors, 2023: A decadal review of the CREST model family: Developments, applications, and outlook. J. Hydrol. X, 20, 100159, https://doi.org/10.1016/j.hydroa.2023.100159.
Liu, C., and Coauthors, 2017: Continental-scale convection-permitting modeling of the current and future climate of North America. Climate Dyn., 49, 71–95, https://doi.org/10.1007/s00382-016-3327-9.
Merz, B., and Coauthors, 2021: Causes, impacts and patterns of disastrous river floods. Nat. Rev. Earth Environ., 2, 592–609, https://doi.org/10.1038/s43017-021-00195-3.
Miller, J. F., R. H. Frederick, and R. J. Tracey, 1973: Washington. Vol. IX, Precipitation-Frequency Atlas of the Western United States, NOAA Atlas 14, 35 pp., https://hdsc.nws.noaa.gov/pub/hdsc/data/papers/Atlas2_Volume9.pdf.
Mullen, H., 2022: Indigenous environmental justice and screening tools: Lessons learned from EJSCREEN and paths forward for the climate and economic justice screening tool. M.S. thesis, School for Environment and Sustainability, University of Michigan, 53 pp., https://doi.org/10.7302/4329.
Musselman, K. N., F. Lehner, K. Ikeda, M. P. Clark, A. F. Prein, C. Liu, M. Barlage, and R. Rasmussen, 2018: Projected increases and shifts in rain-on-snow flood risk over western North America. Nat. Climate Change, 8, 808–812, https://doi.org/10.1038/s41558-018-0236-4.
Pauli, N., M. Williams, S. Henningsen, K. Davies, C. Chhom, F. Ogtrop, S. Hak, B. Boruff, and A. Neef, 2021: “Listening to the sounds of the water”: Bringing together local knowledge and biophysical data to understand climate-related hazard dynamics. Int. J. Disaster Risk Sci., 12, 326–340, https://doi.org/10.1007/s13753-021-00336-8.
Prein, A. F., A. Gobiet, M. Suklitsch, H. Truhetz, N. K. Awan, K. Keuler, and G. Georgievski, 2013: Added value of convection permitting seasonal simulations. Climate Dyn., 41, 2655–2677, https://doi.org/10.1007/s00382-013-1744-6.
Prein, A. F., and Coauthors, 2015: A review on regional convection-permitting climate modeling: Demonstrations, prospects, and challenges. Rev. Geophys., 53, 323–361, https://doi.org/10.1002/2014RG000475.
Prein, A. F., C. Liu, K. Ikeda, S. B. Trier, R. M. Rasmussen, G. J. Holland, and M. P. Clark, 2017: Increased rainfall volume from future convective storms in the U.S. Nat. Climate Change, 7, 880–884, https://doi.org/10.1038/s41558-017-0007-7.
Riahi, K., and Coauthors, 2017: The shared socioeconomic pathways and their energy, land use, and greenhouse gas emissions implications: An overview. Global Environ. Change, 42, 153–168, https://doi.org/10.1016/j.gloenvcha.2016.05.009.
Saharia, M., P.-E. Kirstetter, H. Vergara, J. J. Gourley, Y. Hong, and M. Giroud, 2017: Mapping flash flood severity in the United States. J. Hydrometeor., 18, 397–411, https://doi.org/10.1175/JHM-D-16-0082.1.
Sampson, C. C., A. M. Smith, P. D. Bates, J. C. Neal, L. Alfieri, and J. E. Freer, 2015: A high-resolution global flood hazard model. Water Resour. Res., 51, 7358–7381, https://doi.org/10.1002/2015WR016954.
Schär, C., C. Frei, D. Lüthi, and H. C. Davies, 1996: Surrogate climate‐change scenarios for regional climate models. Geophys. Res. Lett., 23, 669–672, https://doi.org/10.1029/96GL00265.
Schewe, J., and Coauthors, 2019: State-of-the-art global models underestimate impacts from climate extremes. Nat. Commun., 10, 1005, https://doi.org/10.1038/s41467-019-08745-6.
Slingo, J., and Coauthors, 2022: Ambitious partnership needed for reliable climate prediction. Nat. Climate Change, 12, 499–503, https://doi.org/10.1038/s41558-022-01384-8.
Smiley, K. T., I. Noy, M. F. Wehner, D. Frame, C. C. Sampson, and O. E. J. Wing, 2022: Social inequalities in climate change-attributed impacts of Hurricane Harvey. Nat. Commun., 13, 3418, https://doi.org/10.1038/s41467-022-31056-2.
Swain, D. L., O. E. J. Wing, P. D. Bates, J. M. Done, K. A. Johnson, D. R. Cameron, 2020: Increased flood exposure due to climate change and population growth in the United States. Earth’s Future, 8, e2020EF001778, https://doi.org/10.1029/2020EF001778.
Tabari, H., 2020: Climate change impact on flood and extreme precipitation increases with water availability. Sci. Rep., 10, 13768, https://doi.org/10.1038/s41598-020-70816-2.
Tate, E., M. A. Rahman, C. T. Emrich, and C. C. Sampson, 2021: Flood exposure and social vulnerability in the United States. Nat. Hazards, 106, 435–457, https://doi.org/10.1007/s11069-020-04470-2.
Tellman, B., J. A. Sullivan, C. Kuhn, A. J. Kettner, C. S. Doyle, G. R. Brakenridge, T. A. Erickson, and D. A. Slayback, 2021: Satellite imaging reveals increased proportion of population exposed to floods. Nature, 596, 80–86, https://doi.org/10.1038/s41586-021-03695-w.
Wang, J., and Coauthors, 2011: The Coupled Routing And Excess Storage (CREST) distributed hydrological model. Hydrol. Sci. J., 56, 84–98, https://doi.org/10.1080/02626667.2010.543087.
Whyte, K. P., 2017: Is it colonial déjà vu? Indigenous peoples and climate injustice. Humanities for the Environment: Integrating Knowledge, Forging New Constellations of Practice, J. Adamson, M. Davis, and H. Huang, Eds., Routledge Earthscan, 88–105.
Wildcat, D. R., 2010: Red Alert! Saving the Planet with Indigenous Knowledge. Fulcrum Publishing, 128 pp.
Wing, O. E. J., and Coauthors, 2022: Inequitable patterns of US flood risk in the Anthropocene. Nat. Climate Change, 12, 156–162, https://doi.org/10.1038/s41558-021-01265-6.
WMO, 2021: WMO atlas of mortality and economic losses from weather, climate and water extremes (1970–2019). WMO 1267, 90 pp., https://library.wmo.int/records/item/57564-wmo-atlas-of-mortality-and-economic-losses-from-weather-climate-and-water-extremes-1970-2019#.YS4KedP7TX0.
Wright, D. B., C. D. Bosma, and T. Lopez-Cantu, 2019: U.S. hydrologic design standards insufficient due to large increases in frequency of rainfall extremes. Geophys. Res. Lett., 46, 8144–8153, https://doi.org/10.1029/2019GL083235.
Wu, H., R. F. Adler, Y. Hong, Y. Tian, and F. Policelli, 2012: Evaluation of global flood detection using satellite-based rainfall and a hydrologic model. J. Hydrometeor., 13, 1268–1284, https://doi.org/10.1175/JHM-D-11-087.1.
Yu, G., D. B. Wright, and Z. Li, 2020: The upper tail of precipitation in convection-permitting regional climate models and their utility in nonstationary rainfall and flood frequency analysis. Earth’s Future, 8, e2020EF001613, https://doi.org/10.1029/2020EF001613.