Climate Change in the Republic of the Marshall Islands: Indicators and Considerations for Key Sectors is one in a series of PIRCA reports. Authors from Arizona State University, the East-West Center, the Majuro Weather Service Office, and the University of Hawaiʻi—along with 29 technical contributors from local government, NGOs, and research—collaboratively developed the RMI PIRCA report.

Key Messages

Climate Change in the Republic of the Marshall Islands lays out the changes the country is already experiencing, and what lies ahead. The key messages for decision-makers include:

• Sea level rise threatens infrastructure, food and water security, and important ecosystems and cultural sites. Frequent and extensive flooding, coastal erosion, and saltwater contamination of groundwater are expected as sea level rise accelerates, threatening the long-term habitability of the atoll nation.
• Ocean changes disrupt fisheries and cause coral loss. Coral reefs are key to the Marshall Islands’ fisheries and protection from coastal flooding. Fisheries changes and extensive coral loss are possible within the next few decades if current trends in rising ocean temperatures continue.
• Hotter days and nights and stronger storms affect human health. Temperatures have risen, and heat waves stress water supplies and exacerbate a range of pre-existing health issues. More intense tropical cyclones mean a greater potential for flooding and associated public health and safety risks.
• Collaborations and increased climate finance can bolster resilience. National government, international partners, non-governmental organizations, and local communities can work to expand adaptation strategies and access to climate finance, which is needed to meet the scale of challenges facing the RMI.

About Climate Change in the Republic of the Marshall Islands and the PIRCA

The collective efforts of the technical contributors and coordinating authors made the RMI PIRCA report possible. The report builds upon the US National Climate Assessment, offering a closer look at climate change impacts in the RMI and providing information for a wide range of sectors.   

The PIRCA is funded and supported by Arizona State University’s Global Institute of Sustainability and Innovation, the US National Oceanic and Atmospheric Administration’s CAP Program (through Pacific RISA), and the East-West Center’s Research Program.

Cover photo: An aerial view of Majuro shows that atolls are primarily covered with forest or agroforest, surrounded by shallow reef. Photo courtesy of USGS project, “‘Vegetative Guide & Dashboard’ relating atoll traditional agroforestry recommendations to predicted climate and sea level conditions in the Marshall Islands.”

SEA technology is also generalizable across geoscience domains, through a framework called an Intelligent Data Exploring Assistant (IDEA), which can be demonstrated by asking it to analyze atmospheric observations from Mars collected by NASA’s InSight Mission (Try it!). By combining LLM capabilities with robust domain-specific customizations, SEA and the IDEA example generate accurate analyses, visualizations, and insights through natural-language prompts. This study highlights the potential of IDEA frameworks to lower technical barriers, enhance educational opportunities, and transform geoscientific workflows while addressing the limitations and uncertainties of current LLM technology. PI Widlansky’s work also highlights how AI can enhance scientific research and communication, and helps us to envision how the creation of similar tools can support scientists in many fields.

SEAinfo page, with several YouTube video demonstrations and presentations by PI Widlansky
https://uhslc.soest.hawaii.edu/research/SEAinfo/
GitHub IDEA page
https://github.com/uhsealevelcenter/IDEA
IDEA manuscript with a plain language summary and abstract
https://uhslc.soest.hawaii.edu/research/SEAinfo/IDEA_manuscript_latest.pdf

Summary

Studies have demonstrated that climate change is likely to impact mountain ecosystems across the globe, most typically through changes in temperature and shifts in water and nutrient availability, which could lead to upward shifts in species ranges over time. It is unclear, however, the extent to which plant species are shifting their ranges upslope in Hawaiʻi, and whether the rate of movement is different between native and non-native plant species. Mauna Loa, a 13,100 ft. (3,992 m) active volcano on Hawaiʻi Island, experienced a mean annual temperature increase of 1.07 degrees C (1.93℉) between 1970 and 2010, with no detected change in annual precipitation across the gradient, and the authors found it an excellent place to test for species range shifts.

The researchers analyzed long-term vegetation monitoring data from 1970 and 2010 from transects on the southeastern slope of Mauna Loa, Hawaiʻi in order to explore potential shifts in the elevation of both native and non-native plants. To do this, the team re-surveyed 46 vegetation plots in 2010 using the same methodology as the original 1970 surveys, which included collecting presence/absence information and the percentage cover in each vegetation layer (i.e., tree, shrub, and herb layers). To compare shifts in native and non-native species they calculated mean elevation, elevation range, and the upper and lower elevation limits of 69 species that occurred in both the 1970 and 2010 data sets.

The authors found that over a 40-year period, the direction and magnitude of shifts and whether they were driven by changes in the lower, mean, or upper limits depended on the individual species. Strawberry guava shifted its mean elevation ~233 ft. (71 m) upslope due to changes in both its upper and lower limits; weeping-grass had a large increase in mean elevation driven by an increase in its upper limit but not lower limit; molasses grass shifted downward in mean elevation by ~244 ft. (74 m), and Asian sword fern increased in mean elevation with a change in its upper, but not lower elevational limits

Take Home Points

• Individual non-native plant species showed wide variability in their elevation shifts with some species raising their upper elevational limit significantly but not their lower limit (e.g., weeping-grass, Microlaena stipoides), and others increasing both their lower and upper limits (e.g., bamboo orchid, Arundina graminifolia). Some non-native species actually shifted downward over the 40-year period (e.g., red-top grass, Melinis repens).
• The mechanisms driving shifting distributions of non-native plants are largely unknown and could be due to a variety of factors, including: 1) non-native species are still spreading to fill their climate niche, 2) increased disturbances/greater spread (e.g., changing fire regime), 3) enhanced anthropogenic dispersal (building of roads and their use), or 4) changes in habitat suitability due to a changing climate.
• The authors found that native species were generally experiencing shrinking ranges, because the lower limits of their range had increased in elevation and the upper range stayed the same. The authors believe that this stationary upper limit is likely due to the trade-wind inversion (TWI) which abruptly limits rainfall at that higher elevation. In contrast, non-native species experienced a rise in lower and higher elevation limits – perhaps because many non-native species had not yet reached the TWI elevation.

Management Considerations

• Consider increased monitoring efforts that may help detect shifts in non-native plant distributions, especially at the lower and upper elevation limits of species.
• Consider assisting native plant species dispersal across their ranges as conditions change, especially if natural dispersal corridors are absent or degraded. Non-native plants may be able to respond more rapidly to changes in environmental conditions, putting native species at a disadvantage.
• Consider increasing protection efforts for native plants or collecting seeds for future propagation and reintroduction efforts at low elevations, as non-native tree cover is increasing more rapidly than native species cover in these areas.

Summary

Invasive plants that modify ecosystems can harm native biodiversity and degrade important ecosystem services. These species, which are a subset of non-native species, are also likely to be influenced by climate change which could exacerbate impacts. To assess the vulnerability of native ecosystems and federally designated critical habitat in Hawaiʻi to these harmful invaders, the authors used species distribution models to project the current (2013) and future (2100) distribution of 17 particularly detrimental invasive plants across the main Hawaiian Islands. The climate change scenario used in the analysis was the 2080-2100 SRES A1B, which projects a moderately warmer and wetter future. By combining models for multiple invasive species, they projected likely hotspots of non-native species richness and diversity. They used dynamically downscaled projections from the Hawaiian Regional Climate Model, and used three different methodologies (MAXENT, Random Forest, and Gradient Boosting Model), as well as seven bioclimatic and topographic variables, to model species distributions over geographic space. They found that most of the 17 species increased in area under climate change, with higher elevations facing greater invasion risk in 2100.

Results

The area available for occupation by the 17 selected invasive plant species increased by ~11% overall, and by ~12% in federally designated critical habitat in 2100. Invasibility, a metric that includes invasive species richness and diversity, is predicted to increase in Hawaiʻi’s upper elevation areas by 2100. While the majority of invasive species increased in area under climate change, a few species decreased in suitable area at lower elevations.

Management Considerations

• Of the ~8,000 to 10,000 plant species introduced to Hawaiʻi, only about 90 are considered extremely harmful due to their ability to degrade entire ecosystems1. If resources are limited, consider prioritizing the control and prevention of these particularly harmful invaders.
• The distribution of many of the most harmful plant invaders is expected to increase in both area and elevation with climate change. Consider revisiting management goals and objectives as conditions change.
• Consider increasing monitoring efforts in upper elevation native ecosystems for invasive plant species that may be shifting upslope as the climate warms, especially in areas with large concentrations of invasive plant species at lower elevations.
• The quality of current and future projections relies on location data, which is limited for many invasive species. Consider recording both presence and absence location information for invasive species and in areas of both high and low conservation value for use in future modeling efforts.

Take Home Points

• Under increasing temperatures, both native and invasive plant species in Hawaiʻi are expected to shift to upper elevations to find temperature equivalent zones.
• Control of invasive species within and at the boundaries of upper elevation ecosystems will be critical in the coming decades to maintain ecosystem health and integrity.
• Invasive plant species may lose suitable habitat at lower elevations with climate change, though many of these low elevation areas are of marginal value for conservation.
• Given that many critical habitat areas are in high-elevation ecosystems that are vulnerable to invasive species shifts due to climate change, new designations of critical habitat should consider potential climate change impacts.

Summary

Extreme Climatic Events (ECEs) are rare, high-impact events such as hurricanes, heat waves, and extended drought, and they are increasing in frequency and intensity across the Pacific. These extreme events are one of the most immediate threats caused by climate change, and can provide increased opportunities for invasive species to colonize and spread. Despite potentially severe consequences, however, ECEs are rarely considered in planning efforts for ecological restoration. The authors examined the impacts of ECEs on restoration projects and the degree to which they were resilient to ECEs, and they found overwhelmingly negative impacts on restoration efforts. Impacts varied across geographies, species, and within sites, highlighting the need for restoration practitioners to adopt a “portfolio approach” to increase resilience of projects to ECEs. By diversifying the sites, species, and genotypes used as well as the methods employed, managers can reduce the risk of an entire restoration project failing when an extreme event occurs.

Results

Hurricanes and severe storms were the most reported ECEs, impacting 76% of the projects examined via wind, floods, and/or waves. The severity of impacts varied substantially by project and ECEs were not uniform across the restoration site, across all restoration methods, or across species, life stages, or genotypes.  Types of impacts included mortality, community shifts, impacts to reproduction of target species, change in vegetation structure, and changes in species cover. ECEs had overwhelmingly negative impacts on restoration projects, however a few reported both positive and negative impacts or neutral/no-damage impacts. One study reported a positive impact.

Management Considerations

• Plan for ECEs by considering the potential impacts of extreme events most likely to impact your area, possibly exceeding historical events.
• To spread risk through the “portfolio approach”, consider having multiple restoration sites instead of one; place restoration sites across scales (e.g., across elevational or rainfall gradients, or across multiple years or seasons to increase the likelihood of favorable conditions), make use of topographic complexity, and use multiple species.  If possible, identify likely spatial and temporal refugia and incorporate these into project design to reduce the impacts of ECEs.
• Consider selecting a diversity of propagule sources to enhance genetic diversity and adaptive potential to climate change. For example, incorporate species with traits that are more tolerant of extreme conditions such as droughts, heatwaves, or high winds.
• Consider having a post-ECE response plan in place that includes a budget for monitoring and invasive species control. Data on impacts from ECEs can help inform future restoration designs and guide adaptive management.

Take Home Points

• ECEs can create major setbacks for restoration projects by destroying or damaging structures or sites, and by threatening restored species.
• To increase the resilience of restoration projects to ECEs, spread risk across time and space by using the portfolio approach.
• Adapting restoration projects to ECEs or post-ECE recovery may require the use of propagules whose genotypes are more tolerant or resistant to ECEs.
• To adapt to climate change, including to an increasing frequency and severity of ECEs, restoration projects will need to plan for greater uncertainty, secure increased funding for monitoring and adaptive management in response to ECEs, and anticipate setbacks and longer timeframes for success.

Summary

Knowing how major vegetation types (biomes) might shift in a landscape as the climate changes is important for conservation planning.  Investments in restoration or species recovery might best focus on areas that are less likely to change to a different biome, while sites likely to change character to a different biome might present opportunities for translocation.  To investigate the stability of Hawaiʻi’s biomes to projected changes in climate, Fortini & Jacobi first modeled the potential current distribution of four main native biomes (dry shrubland, dry forest, mesic forest & wet forest) across the Hawaiian Islands, and then projected future biome distributions using precipitation and temperature from three end of century climate scenarios downscaled to Hawaiʻi that correspond to average temperature increases of 1.7 degrees Celsius (1.8 degrees Fahrenheit), 2.5 °C (4.5 °F), and 3.3 °C (5.9 °F). Despite large differences in future climate projections, the authors found that 35% of the areas considered are projected to maintain their currently most compatible native biome.  However, some biomes with little expected change in overall extent are projected to experience large shifts in location.

Results

The area suitable to native mesic forest is projected to decrease substantially, while the area suitable to native dry shrubland is projected to expand.  Area suitable to native wet forests increase in a wet climate change scenario and decrease in the drier scenarios. The total area suitable for dry forest remains relatively stable, but shifts across the landscape due to changes from dry forests to dry shrubland that are offset by changes from mesic forest to dry forest. Under all the climate change scenarios considered, a large portion of areas that are currently most suitable to dry shrubland and wet forest are expected to remain suitable to those same biomes. The patterns generally held across all islands, however dry shrubland increases are larger for Oʻahu and Kauaʻi, mesic forest loss is smaller for Hawaiʻi Island, and areas suitable to dry forest on Maui are projected to decrease.

Management Implications

• High-confidence, native-dominated “stasis” areas were mostly wet forest.  These areas offer options for long-term conservation as they are expected to have the greatest resilience across a range of potential future climates.
• Managers of mesic forest and dry forest face the loss or shift of large areas of these forest types and may need to consider a portfolio of tools to conserve the high levels of biodiversity found within.
• A decrease in biome suitability at a location implies increased risk of mortality of native component species and opportunities for establishment of invasive species.
• A portion of the landscape (primarily coastal areas and lower elevation wet forests) is projected to be only marginally suitable for any native biome by the end of the century. These areas may be especially susceptible to biological invasions.

Take Home Points

• Despite a high level of uncertainty, climate change is expected to force large shifts in the major biomes in Hawaiʻi by the end of the century, and proactive approaches will be needed to conserve biodiversity, ecosystem services, and livelihoods associated with affected biomes.
• While this approach reveals potential large-scale shifts in biomes, factors such as substrate age, fog interception, and non-climactic factors may be important for determining vegetation patterns at finer spatial scales. Therefore, there are likely other climate change refugia or areas for “stasis” for species not identified by this study that will be important to find and protect.

Summary

Small mammals, such as mice and rats, are some of the most problematic invasive species impacting native island biodiversity. In the Pacific region, rats were found to predate seeds causing several plants to become rare, damage coconut palm fruit production, and reduce seabird, green sea turtle, and crab populations by eating their eggs and juveniles. Severe storms, such as hurricanes and typhoons, can change invasive mammal populations by altering habitat or by facilitating dispersal between islands. Eradicating invasive rodent populations on islands can protect threatened or endangered species, restore native flora and fauna, and build resilience to climate change. On the islet of Irooj in the Marshall Islands, rat eradication restored native seabird populations which increased nutrients through guano deposition, and created healthier, more productive coral reef ecosystems.

In this study, non-native and invasive small mammals, including black rats (Rattus rattus), house mice (Mus musculus), and mongoose (Urva auropunctata), were monitored in 2017 and 2018 to examine the impacts of Hurricane Irma and Hurricane Maria on relative abundance across the tropical islands of St. Croix, United States Virgin Islands (USVI) in the Caribbean. Both hurricanes hit the region in September 2017 just north (Irma) and south (Maria) of the islands and measured as category 3. Comparative data from before the hurricanes in 2017 and after were collected from three study sites: Sandy Point National Wildlife Refuge (Sandy Point), Green Cay National Wildlife Refuge (Green Cay; a smaller island), and Buck Island Reef National Monument (Buck Island). Data were gathered through annual and semi-annual snap-trapping surveys and tracking tunnels that recorded animal presence via inked footprints.

Results

For Sandy Point, there was a significant increase in the relative abundance of mice and on Buck Island, the existing mouse population was found to have doubled after the hurricanes. The relative abundance of rats did not change after the hurricanes on Buck Island, however mongoose relative abundance slightly decreased after the hurricanes. The significant increase in mouse populations at Sandy Point and Buck Island may have been due to the increase in grass cover which increased food availability, and possibly a decrease in predator populations following the hurricanes. No invasive small mammals were present on Green Cay before the hurricanes, but afterwards, rats were found to be present. This rat introduction is likely due to animals rafting across the ocean on debris or being forced to swim between islands after being displaced by storms.

Management Considerations

• Consider increasing monitoring efforts for invasive small mammals both before and after storms to allow for a more rapid response if new incursions are detected. The use of tracking tunnels can enable rapid confirmation of new invasive small mammals in remote locations.
• Consider increased biosecurity efforts related to invasive species prevention on islands after severe storms. The rat incursion on Green Cay may have been an accidental introduction by humans via a rat-infested boat landing on the island.
• Consider increasing both biosecurity and monitoring efforts on offshore islands that are close to main islands because they are likely more vulnerable to invasion than remote offshore islands. Green Cay, which was invaded by rats after the hurricanes is only ~1400 ft. from St. Croix Island, whereas Buck Island, which remained rat-free is ~9000 ft. from St. Croix.
• Some invasive small mammals may decline after intense storms (e.g., mongoose tended to decrease in this study), creating an opportunity for eradication.

Take Home Points

• New introductions of invasive small mammals can occur on islands due to a variety of factors including severe storms.
• Each mammal species had a different population-level response following the hurricanes, with predators (mongoose) showing the smallest response.
• Routine monitoring for the presence, abundance, and species composition of invasive small mammals is essential for establishing baselines and detecting new arrivals.
• Managers should plan for the potential of increased impacts from invasive small mammals after large storm events from new introductions, and re-establishment of species that have been previously eradicated.

Summary

Biological control agents can be used to manage some invasive plant populations, but climate change may threaten their continued effectiveness. This study investigated the effect of elevated CO2 levels on the performance of a plant invasive to the southeastern U.S., tropical soda apple (Solanum viarum), and the host-specific biocontrol beetle (Gratiana boliviana) used against it. Plants were grown and beetles were exposed to ambient, medium, and high levels of CO2, and metrics of performance were recorded. Under high levels of CO2, plants had greater biomass. In contrast, the biocontrol beetle had reduced survival, slower developmental time, smaller body size, and lower consumption of leaves by fifth instars. These results suggest decreased performance of G. boliviana in controlling a more aggressive tropical soda apple as CO2 levels in our atmosphere continue to rise.

Take Home Points

• The effectiveness of biocontrol agents are expected to change under future climate conditions.
• The invasive plant tropical soda apple (Solanum viarum) performed better and its biocontrol beetle (G. boliviana) performed worse under elevated CO2 levels.
• The effectiveness of this biocontrol agent on tropical soda apple will be diminished as CO2 levels continue to increase.

Management Implications

• The effects of climate change should be added as a consideration when evaluating future biocontrol agents.
• The timing, source populations, and frequency of biocontrol releases may need to be adjusted under future climate conditions.
• Managers and researchers should collaborate to identify high priority biocontrol systems to study.

Summary

Extreme climate events (ECEs) can influence different stages of invasion and create an ‘invasion window’, or an opportunity for invasive species to take advantage of resources made available after an ECE. ECEs often alter ecosystem structure or function and may cause abrupt mortality of resident species. This disturbance provides introduced species the opportunity to establish and take advantage of resources available post-disturbance. Other ECEs may cause stress to resident species and limit their ability to recover and compete for resources with invasive species. Invasive species often are more likely to succeed post ECE because they tend to have broader environmental tolerances than co-occurring native species, however responses will depend on each species’ tolerance to ECEs. ECEs, however, do not uniformly favor non-native species and many non-natives that benefit may have no substantial ecological impact, or impacts may be context dependent. Some ECEs may negatively affect established invaders, providing opportunities for restoration of native species.

Take Home Points

• ECEs can allow invasive species to transform ecosystems to new and persistent states.
• ECEs can increase the transportation of invasive species and reduce the current ecosystem’s ability to resist or be resilient to new invasions.
• ‘Resource pulses’ (e.g., a disturbance-induced sudden increase in light, space, or water) created by ECEs can allow invaders to establish and spread.

Management Considerations

Pre-ECE (Planning):

• Identify what types of ECEs will likely affect your region and evaluate your current management goals and objectives to see how they may be affected. Successful mitigation of the future impacts of ECEs will likely require planning and mobilization of resources.
• Include the potential responses of invasive species to expected ECEs in regional risk assessments and species watch lists.
• Rank the value and importance of cultural and natural resources that may be impacted by ECEs, so that rapid protection efforts can be targeted after an ECE.
• Identify management areas that are vulnerable to disturbance and work to increase the resilience of those areas prior to ECEs. Develop greater capacity for post-ECE restoration such as saving a supply of seeds for future restoration efforts.
• Consider how barriers or buffers between invaded and uninvaded areas could be bolstered or increased prior to ECEs.
• Develop and coordinate efforts across agencies, individuals, and landscapes for addressing ECEs. Cooperative agreements may assist with cross-boundary impacts.

Post ECE (Response):

• Increase early detection and rapid response (EDRR) efforts after an ECE to eradicate new invasions. ECEs that greatly reduce populations of invasive species may also increase the likelihood of eradication of those same species if managers can respond quickly.
• Prioritize management of the most detrimental invasive species first, such as those that are known to transform ecosystems or are highly ranked on risk assessments.
• Enlist volunteer and community groups to help monitor for invasive species after an ECE and respond to those that are found.

Summary

Mainstream management approaches often focus on the eradication of newly arriving species as a default, desirable management strategy, yet human communities may have different perspectives around management. Indigenous responses to the arrival of new species rarely appear in the conservation literature. Commonly used conservation definitions of ‘native’ and ‘alien’ do not capture the array of relationships between Indigenous peoples and plants and animals. Invasive species plans that do incorporate Indigenous perspectives largely focus on perceived threats to cultural practices and not on reciprocal relationships.

Management Considerations

• Catalog diverse perspectives on introduced species, including Indigenous perspectives, to inform management plans.
• Include Indigenous knowledge, expertise, and perspectives in decision-making.
• Assess the social, economic, and cultural impacts of eradication of invasive species when planning management actions.
• Consider the various uses and significance of culturally important non-native species when weighing the benefits and harms before committing to management action. Existing invasive species management practices that interfere with the adaptive capacity of Indigenous communities need to be reevaluated and addressed.

Take Home Points

• Meaningful and effective place-based conservation initiatives rest on knowledge of the species people use. To address social and environmental justice issues, learn what communities need and value, and their relationship to resources, including introduced species.
• Whether a species is seen as invasive can vary across space and time, and invasive status may be contested and dynamic. Community relationships with introduced species may also develop over time as populations of native species decline.
• Successful approaches to incorporating Indigenous relational frameworks to invasive species management will include partnering with communities and centering community needs, using multidimensional impact measurements (including all positive and negative effects), and using deliberative engagement with communities upstream of policy creation or decisions.
• Moving from stewardship to kinship-based approaches, which includes both human and non-human beings, is a critical step towards engaging Indigenous approaches to landscape curation.