Probabilistic Projections Research Articles

Probabilistic forecasting of maximum human lifespan by 2100 using Bayesian population projections.

We consider the problem of quantifying the human lifespan using a statistical approach that probabilistically forecasts the maximum reported age at death (MRAD) through 2100. We seek to quantify the probability that any person attains various extreme ages, such as those above 120, by the year 2100. We use the exponential survival model for supercentenarians (people over age 110) of Rootzén and Zholud (2017) but extend the forecasting window, quantify population uncertainty using Bayesian population projections, and incorporate the most recent data from the International Database on Longevity (IDL) to obtain unconditional estimates of the distribution of MRAD this century in a fully Bayesian analysis. We find that the exponential survival model for supercentenarians is consistent with the most recent IDL data and that projections of the population aged 110-114 through 2080 are sensible. We integrate over the posterior distributions of the exponential model parameter and uncertainty in the supercentenarian population projections to estimate an unconditional distribution of MRAD by 2100. Based on the Bayesian analysis, there is a greater than 99% probability that the current MRAD of 122 will be broken by 2100. We estimate the probabilities that a person lives to at least age 126,128, or 130 this century, as 89%, 44%, and 13%, respectively. We have updated the supercentenarian survival model of Rootzén and Zholud using the most recent IDL data, incorporated Bayesian population projections, and extended the forecasting window to create the first fully Bayesian and unconditional probabilistic projection of MRAD by 2100.

Visualising the Uncertainty Cascade in Multi-Ensemble Probabilistic Coastal Erosion Projections

Future projections of coastal erosion, which are one of the most demanded climate services in coastal areas, are mainly developed using top-down approaches. These approaches consist of undertaking a sequence of steps that include selecting emission or concentration scenarios and climate models, correcting models bias, applying downscaling methods, and implementing coastal erosion models. The information involved in this modelling chain cascades across steps, and so does related uncertainty, which accumulates in the results. Here, we develop long-term multi-ensemble probabilistic coastal erosion projections following the steps of the top-down approach, factorise, decompose and visualise the uncertainty cascade using real data and analyse the contribution of the uncertainty sources (knowledge-based and intrinsic) to the total uncertainty. We find a multi-modal response in long-term erosion estimates and demonstrate that not sampling internal climate variability’s uncertainty sufficiently could lead to a truncated outcomes range, affecting decision-making. Additionally, the noise arising from internal variability (rare outcomes) appears to be an important part of the full range of results, as it turns out that the most extreme shoreline retreat events occur for the simulated chronologies of climate forcing conditions. We conclude that, to capture the full uncertainty, all sources need to be properly sampled considering the climate-related forcing variables involved, the degree of anthropogenic impact and time horizon targeted.

Statistical emulation of a perturbed basal melt ensemble of an ice sheet model to better quantify Antarctic sea level rise uncertainties

Abstract. Antarctic ice shelves are vulnerable to warming ocean temperatures, and some have already begun thinning in response to increased basal melt rates. Sea level is therefore expected to rise due to Antarctic contributions, but uncertainties in its amount and timing remain largely unquantified. In particular, there is substantial uncertainty in future basal melt rates arising from multi-model differences in thermal forcing and how melt rates depend on that thermal forcing. To facilitate uncertainty quantification in sea level rise projections, we build, validate, and demonstrate projections from a computationally efficient statistical emulator of a high-resolution (4 km) Antarctic ice sheet model, the Community Ice Sheet Model version 2.1. The emulator is trained to a large (500-member) ensemble of 200-year-long 4 km resolution transient ice sheet simulations, whereby regional basal melt rates are perturbed by idealized (yet physically informed) trajectories. The main advantage of our emulation approach is that by sampling a wide range of possible basal melt trajectories, the emulator can be used to (1) produce probabilistic sea level rise projections over much larger Monte Carlo ensembles than are possible by direct numerical simulation alone, thereby providing better statistical characterization of uncertainties, and (2) predict the simulated ice sheet response under differing assumptions about basal melt characteristics as new oceanographic studies are published, without having to run additional numerical ice sheet simulations. As a proof of concept, we propagate uncertainties about future basal melt rate trajectories, derived from regional ocean models, to generate probabilistic sea level rise estimates for 100 and 200 years into the future.

Reduced Complexity Model Intercomparison Project Phase 2: Synthesizing Earth System Knowledge for Probabilistic Climate Projections.

Over the last decades, climate science has evolved rapidly across multiple expert domains. Our best tools to capture state‐of‐the‐art knowledge in an internally self‐consistent modeling framework are the increasingly complex fully coupled Earth System Models (ESMs). However, computational limitations and the structural rigidity of ESMs mean that the full range of uncertainties across multiple domains are difficult to capture with ESMs alone. The tools of choice are instead more computationally efficient reduced complexity models (RCMs), which are structurally flexible and can span the response dynamics across a range of domain‐specific models and ESM experiments. Here we present Phase 2 of the Reduced Complexity Model Intercomparison Project (RCMIP Phase 2), the first comprehensive intercomparison of RCMs that are probabilistically calibrated with key benchmark ranges from specialized research communities. Unsurprisingly, but crucially, we find that models which have been constrained to reflect the key benchmarks better reflect the key benchmarks. Under the low‐emissions SSP1‐1.9 scenario, across the RCMs, median peak warming projections range from 1.3 to 1.7°C (relative to 1850–1900, using an observationally based historical warming estimate of 0.8°C between 1850–1900 and 1995–2014). Further developing methodologies to constrain these projection uncertainties seems paramount given the international community's goal to contain warming to below 1.5°C above preindustrial in the long‐term. Our findings suggest that users of RCMs should carefully evaluate their RCM, specifically its skill against key benchmarks and consider the need to include projections benchmarks either from ESM results or other assessments to reduce divergence in future projections.

Probabilistic projections of future warming and climate sensitivity trajectories

Abstract Projections of future global mean surface warming for a given forcing scenario remain uncertain, largely due to uncertainty in the climate sensitivity. The ensemble of Earth system models from the Climate Model Intercomparison Project phase 6 (CMIP6) represents the dominant tools for projecting future global warming. However, the distribution of climate sensitivities within the CMIP6 ensemble is not representative of recent independent probabilistic estimates, and the ensemble contains significant variation in simulated historic surface warming outside agreement with observational datasets. Here, a Bayesian approach is used to infer joint probabilistic projections of future surface warming and climate sensitivity for shared socio-economic pathway (SSP) scenarios. The projections use an efficient climate model ensemble filtered and weighted to encapsulate observational uncertainty in historic warming and ocean heat content anomalies. The probabilistic projection of climate sensitivity produces a best estimate of 2.9°C, and 5th–95th percentile range of 1.5–4.6°C, in line with previous estimates using multiple lines of evidence. The joint projection of surface warming over the period 2030–40 has a 50% or greater probability of exceeding 1.5°C above preindustrial for all SSPs considered: 119, 126, 245, 370 and 585. Average warming by the period 2050–60 has a >50% chance of exceeding 2°C for SSPs 245, 370 and 585. These results imply that global warming is no longer likely to remain under 1.5°C, even with drastic and immediate mitigation, and highlight the importance of urgent action to avoid exceeding 2°C warming.

Camera-LIDAR Integration: Probabilistic Sensor Fusion for Semantic Mapping

An automated vehicle operating in an urban environment must be able to perceive and recognise objects and obstacles in a three-dimensional world for navigation and path planning. In order to plan and execute accurate and sophisticated driving maneuvers, a high-level contextual understanding of the surroundings is essential. Due to the recent progress in image processing, it is now possible to obtain high definition semantic information in 2D from monocular cameras, though cameras cannot reliably provide the high accuracy 3D information provided by lasers. The fusion of these two sensor modalities can overcome the shortcomings of each individual sensor, though there are a number of important challenges that need to be addressed in a probabilistic manner. In this paper we address the common, yet challenging, LIDAR/camera/semantic fusion problems which are seldom approached in a wholly probabilistic manner. Our approach is capable of using a multi-sensor platform to build a three-dimensional semantic voxelized map that considers the uncertainty of all of the processes involved. We present a probabilistic pipeline that incorporates uncertainty from the sensor readings (cameras, LIDAR, IMU and wheel encoders), compensation for the motion of the vehicle, and heuristic label probabilities for the semantic images depicted in <xref ref-type="fig" rid="fig1" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Fig. 1</xref> . We also present a novel and efficient viewpoint validation algorithm to check for occlusions within the camera frame. A probabilistic projection is performed from the camera images to the LIDAR point cloud. Each labelled LIDAR scan then feeds into an octree map-building algorithm that updates the class probabilities of the map voxels every time a new observation is available. We validate our approach using a set of qualitative and quantitative experiments using the USyd Campus Dataset. <xref ref-type="fn" rid="fn1" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">¹</xref> These tests demonstrate the usefulness of a probabilistic sensor fusion approach by evaluating the performance of the perception system in a typical autonomous vehicle application.

Evaluating the Economic Cost of Coastal Flooding

Sea level rise will cause spatial shifts in economic activity over the next 200 years. Using a spatially disaggregated, dynamic model of the world economy, this paper estimates the consequences of probabilistic projections of local sea level changes. Under an intermediate scenario of greenhouse gas emissions, permanent flooding is projected to reduce global real GDP by 0.19 percent in present value terms. By the year 2200, a projected 1.46 percent of the population will be displaced. Losses in coastal localities are much larger. When ignoring the dynamic response of investment and migration, the loss in real GDP in 2200 increases from 0.11 percent to 4.5 percent. (JEL E23, F01, Q54, Q56)

Probabilistic population projections for provincial levels in Indonesia

This research aims to produce a probabilistic population projection for the provincial level in Indonesia. The base population uses the base population used by Bappenas, BPS, and UNFPA Indonesia in compiling the 2015-2045 Indonesian official population projection. One thousand trajectories of future E0 values are projected using JPP with Asian countries as the prior distribution, while 1,000 trajectories of future TFR values are projected using BHM-based projections with Asian countries as the prior distribution. Meanwhile, the number and pattern of net internal migration are assumed to be constant during the projection period. The calculation process uses the bayesPop package in R software. The projection results show that the total population of the official projection is located at the predicted interval of the Bayesian population projection, except for the Kepulauan Riau.

Event‐Based Storylines to Address Climate Risk

AbstractThe climate science community is challenged to adopt an actionable risk perspective, which is difficult to align with the traditional focus on model‐based probabilistic climate change projections. Event‐based storylines can provide a way out of this conundrum by putting emphasis on plausibility rather than probability. This links directly to common practices in disaster risk management using “stress‐testing” for emergency preparedness based on events that are conditional on specific and plausible assumptions. Event‐based storylines allow for conditional explanations, without full attribution of every causal factor, which is crucial when some aspects of the latter are complex and highly uncertain.

Assessing uncertainty for decision‐making in climate adaptation and risk mitigation

AbstractFuture water availability or crop yield studies, tied to statistics of river flow, precipitation, temperature or evaporation over medium to long‐term horizons, are becoming frequent in climate impact and risk analysis. During the last two decades, access to multi‐system integration of climate models has given rise to the concept of using model ensembles to issue probabilistic climatological projections. These probabilistic projections have not yet been exploited to the full extent in decision support, and are still used to mainly quantify uncertainty bands only for selected climate variables and indicators. One of the reasons of this limited use is the fact that the multi‐system ensemble dispersion is sub‐optimal and does not provide an accurate and reliable representation of the predictive probability density, which is essential for rational decision support under uncertain conditions. The aims of this paper are twofold. First, it seeks to highlight the potential benefits of using climate projections in conjunction with Bayesian paradigms towards educated decision‐making. Second, it discusses how to appropriately formulate probabilistic forecasts by coherently integrating information contained in climate projection ensembles with observations to improve the estimation of the probability density function of future climate states. The results show that the proposed Bayesian approach yields unbiased and sharper predictive distributions for temperature with respect to using the unprocessed ensemble distribution. It also yields improved predictive densities with respect to the Reliability Ensemble Averaging (REA) method.

The Effects of Age Structure on Economic Growth: An Application of Probabilistic Forecasting in India

During recent years there has been an increasing awareness of the explanatory power of demographic variables in economic growth regressions. We estimate a new model of the effects of age structure change on economic growth. We use the new model and recent probabilistic demographic projections for India to derive the uncertainty of predicted economic growth rates caused by the uncertainty in demographic developments.

Probabilistic Projections of Multidimensional Flood Risks at a Convection‐Permitting Scale

AbstractUnderstanding future river flood risk is a prerequisite for developing climate change adaptation strategies and enhancing disaster resilience. Previous flood risk assessments can barely take into account future changes of fine‐scale hydroclimatic characteristics and hardly quantify multivariate interactions among flood variables, thereby resulting in an unreliable assessment of flood risk. In this study, for the first time, we develop probabilistic projections of multidimensional river flood risks at a convection‐permitting scale through the Weather Research and Forecasting (WRF) climate simulations with 4‐km horizontal grid spacing. Vine copula has been widely used to assess the multidimensional dependence structure of hydroclimate variables, but the commonly used frequentist approach may fail to identify the correct vine model and to obtain the uncertainty interval. Thus, a Bayesian vine copula approach is proposed to explicitly address the multidimensional dependence of flood characteristics (i.e., flood peak, volume, and duration) and underlying uncertainties. The proposed approach enables a robust assessment of return periods of future floods for Guadalupe and Mission river basins located in South Texas of the United States. Our findings reveal that the South Texas region is projected to experience more flood events with longer duration and greater discharge volume. The flood peak, however, will not necessarily increase even though precipitation extremes are expected to become more frequent. The projected flood return periods over the Guadalupe river basin do not show an obvious increase while the Mission river basin is projected to face a dramatic increase in flood risk with exposed to 100‐year and even severer floods nearly every 2 years, on average, when considering the combined effects of flood peak, volume, and duration.

Probabilistic Application of an Integrated Catchment-Estuary-Coastal System Model to Assess the Evolution of Inlet-Interrupted Coasts Over the 21st Century

Inlet-interrupted sandy coasts are dynamic and complex coastal systems with continuously evolving geomorphological behaviors under the influences of both climate change and human activities. These coastal systems are of great importance to society (e.g., providing habitats, navigation, and recreational activities) and are affected by both oceanic and terrestrial processes. Therefore, the evolution of these inlet-interrupted coasts is better assessed by considering the entirety of the Catchment-Estuary-Coastal (CEC) systems, under plausible future scenarios for climate change and increasing pressures due to population growth and human activities. Such a holistic assessment of the long-term evolution of CEC systems can be achieved via reduced-complexity modeling techniques, which are also ably quantifying the uncertainties associated with the projections due to their lower simulation times. Here, we develop a novel probabilistic modeling framework to quantify the input-driven uncertainties associated with the evolution of CEC systems over the 21stcentury. In this new approach, probabilistic assessment of the evolution of inlet-interrupted coasts is achieved by (1) probabilistically computing the exchange sediment volume between the inlet-estuary system and its adjacent coast, and (2) distributing the computed sediment volumes along the inlet-interrupted coast. The model is applied at three case study sites: Alsea estuary (United States), Dyfi estuary (United Kingdom), and Kalutara inlet (Sri Lanka). Model results indicate that there are significant uncertainties in projected volume exchange at all the CEC systems (min-max range of 2.0 million cubic meters in 2100 for RCP 8.5), and the uncertainties in these projected volumes illustrate the need for probabilistic modeling approaches to evaluate the long-term evolution of CEC systems. A comparison of 50thpercentile probabilistic projections with deterministic estimates shows that the deterministic approach overestimates the sediment volume exchange in 2100 by 15–30% at Alsea and Kalutara estuary systems. Projections of coastline change obtained for the case study sites show that accounting for all key processes governing coastline change along inlet-interrupted coasts in computing coastline change results in projections that are between 20 and 134% greater than the projections that would be obtained if only the Bruun effect were taken into account, underlining the inaccuracies associated with using the Bruun rule at inlet-interrupted coasts.

Future losses of ecosystem services due to coastal erosion in Europe

Coastal erosion is a major issue facing Europe that will only worsen under future climate change and the resulting sea level rise. One effect of erosion is the loss of ecosystem services, which are provided by coastal areas, such as provisioning, regulating, habitat, and cultural services. These services can be quantified in monetary terms. Here, we present comprehensive estimates of future decline in coastal ecosystem services due to the erosion of sandy coastlines. We used datasets derived from remote sensing products: a pan-European land cover/use dataset (Corine Land Cover) and new global probabilistic coastal erosion projections constrained by artificial and topographical barriers to erosion. The results include historical changes (2000–2018) and projections under two emission scenarios (RCP4.5 and RCP8.5) for 2050 and 2100 together with uncertainty bounds. We estimate that in 2018, the coastal zone (excluding open sea) included 579,700 km2 of habitats generating 494 billion euros of services annually. The future sea-level rise could erode 1.0% [90% confidence interval 0.7–1.5%] of the 2018 area under RCP4.5, and 1.2% [0.7–2.2%] under RCP8.5. The decline in services would be even greater: 4.2% [3.0–6.1%] under RCP4.5, and 5.1% [3.3–8.5%] under RCP8.5. The highest absolute losses would be sustained by salt marshes, while relative losses would be highest in beaches, sands, and dunes. The most affected countries in relative economic terms would be Denmark, Albania, Greece, Estonia, and Finland, but countries such as Germany, the Netherlands, and France would be among those losing the largest share of their coastal ecosystem services. Regional analysis using NUTS 3 regions shows high diversity of the impacts, with many regions along the North Sea and eastern Mediterranean Sea that are heavily affected by coastal erosion-induced loss of ecosystem services. The study highlights the urgency of undertaking mitigation actions.

Climate change impacts on peak river flows: Combining national-scale hydrological modelling and probabilistic projections

Potential future increases in flooding due to climate change need to be taken into consideration when designing flood defences or planning new infrastructure or housing developments. Existing guidance on climate change allowances in Great Britain was based on research that developed a sensitivity-based approach to estimating the impacts of climate change on flood peaks, which was applied with catchment-based hydrological models. Here, the sensitivity-based approach is applied with a national-scale grid-based hydrological model, producing modelled flood response surfaces for every river cell on a 1 km grid. This provides a nationally consistent assessment of the sensitivity of flood peaks across Britain to climatic changes. The flood response surfaces are then combined with the most recent climate change projections, UK Climate Projections 2018 (UKCP18), to provide location-specific information on the potential range of impacts on floods across the country, for three flood return periods, three future time-slices and four emissions scenarios. An accompanying web-tool provides a convenient way to explore the large amount of data produced. Consideration is now being given to how to use the latest work to update guidance on climate change and flood peaks, including a workshop held to gather stakeholder views.

Impacts of climate change on hurricane flood hazards in Jamaica Bay, New York

Sea level rise (SLR) and tropical cyclone (TC) climatology change could impact future flood hazards in Jamaica Bay—an urbanized back-barrier bay in New York—yet their compound impacts are not well understood. This study estimates the compound effects of SLR and TC climatology change on flood hazards in Jamaica Bay from a historical period in the late twentieth century (1980–2000) to future periods in the mid- and late-twenty-first century (2030–2050 and 2080–2100, under RCP8.5 greenhouse gas concentration scenario). Flood return periods are estimated based on probabilistic projections of SLR and peak storm tides simulated by a hydrodynamic model for large numbers of synthetic TCs. We find a substantial increase in the future flood hazards, e.g., the historical 100-year flood level would become a 9- and 1-year flood level in the mid- and late-twenty-first century and the 500-year flood level would become a 143- and 4-year flood level. These increases are mainly induced by SLR. However, TC climatology change would considerably contribute to the future increase in low-probability, high-consequence flood levels (with a return period greater than 100 year), likely due to an increase in the probability of occurrence of slow-moving but intense TCs by the end of twenty-first century. We further conduct high-resolution coastal flood simulations for a series of SLR and TC scenarios. Due to the SLR projected with a 5% exceedance probability, 125- and 1300-year flood events in the late-twentieth century would become 74- and 515-year flood events, respectively, in the late-twenty-first century, and the spatial extent of flooding over coastal floodplains of Jamaica Bay would increase by nearly 10 and 4 times, respectively. In addition, SLR leads to larger surface waves induced by TCs in the bay, suggesting a potential increase in hazards associated with wave runup, erosion, and damage to coastal infrastructure.

CH2018 – National climate scenarios for Switzerland: How to construct consistent multi-model projections from ensembles of opportunity

The latest Swiss Climate Scenarios (CH2018), released in November 2018, consist of several datasets derived through various methods that provide robust and relevant information on climate change in Switzerland. The scenarios build upon the regional climate model projections for Europe produced through the internationally coordinated downscaling effort EURO-CORDEX. The simulations from EURO-CORDEX consist of simulations at two spatial horizontal resolutions, several global climate models, and three different emission scenarios. Even with this unique dataset of regional climate scenarios, a number of practical challenges regarding a consistent interpretation of the model ensemble arise. Here we present the methodological chain employed in CH2018 in order to generate a multi-model ensemble that is consistent across scenarios and is used as a basis for deriving the CH2018 products. The different steps involve a thorough evaluation of the full EURO-CORDEX model ensemble, the removal of doubtful and potentially erroneous simulations, a time-shift approach to account for an equal number of simulations for each emission scenario, and the multi-model combination of simulations with different spatial resolutions. Each component of this cascade of processing steps is associated with an uncertainty that eventually contributes to the overall scientific uncertainty of the derived scenario products. We present a comparison and an assessment of the uncertainties from these individual effects and relate them to probabilistic projections. It is shown that the CH2018 scenarios are generally supported by the results from other sources. Thus, the CH2018 scenarios currently provide the best available dataset of future climate change estimates in Switzerland.

Nonlinear Interactions and Some Other Aspects of Probabilistic Sea Level Projections

Probabilistic sea level projections are frequently used to characterise the uncertainty in future sea level rise. Here, it is investigated how different modelling assumptions and process estimates affect such projections using two process-based models that add up the sea level contributions from different processes such as thermosteric expansion and ice sheet melt. A method is applied to estimate the direct contributions from the different processes as well as that of nonlinear interactions between the processes to the projections. In general, the nonlinear interaction terms are found to be small compared to the direct contributions from the processes, and only a few interaction terms give significant contributions to the projections. Apart from the process estimates, probabilistic models often also incorporate some expert judgements that inflate the uncertainty compared with that derived from climate and ice-sheet models, and the effects of some such judgements are also evaluated and found to have a considerable influence on the projections. Lastly, sea level projections are most often given contingent on representative concentration pathways for atmospheric greenhouse gases. Here, we generalize this approach by also providing projections for a probabilistic baseline scenario.

A computationally efficient method for probabilistic local warming projections constrained by history matching and pattern scaling, demonstrated by WASP–LGRTC-1.0

Abstract. Climate projections are made using a hierarchy of models of different complexities and computational efficiencies. While the most complex climate models contain the most detailed representations of many physical processes within the climate system, both parameter space exploration and integrated assessment modelling require the increased computational efficiency of reduced-complexity models. This study presents a computationally efficient method for generating probabilistic projections of local warming across the globe, using a pattern-scaling approach derived from the Climate Model Intercomparison Project phase 5 (CMIP5) ensemble, that can be coupled to any efficient model ensemble simulation of global mean surface warming. While the method can project local warming for arbitrary future scenarios, using it for scenarios with peak global mean warming ≤2 ∘C is problematic due to the large uncertainties involved. First, global mean warming is projected using a 103-member ensemble of history-matched simulations with an example reduced complexity Earth system model: the Warming Acidification and Sea-level Projector (WASP). The ensemble projection of global mean warming from this WASP ensemble is then converted into local warming projections using a pattern-scaling analysis from the CMIP5 archive, considering both the mean and uncertainty of the local to global ratio of temperature change (LGRTC) spatial patterns from the CMIP5 ensemble for high-end and mitigated scenarios. The LGRTC spatial pattern is assessed for scenario dependence in the CMIP5 ensemble using RCP2.6, RCP4.5 and RCP8.5, and spatial domains are identified where the pattern scaling is useful across a variety of arbitrary scenarios. The computational efficiency of our WASP–LGRTC model approach makes it ideal for future incorporation into an integrated assessment model framework or efficient assessment of multiple scenarios. We utilise an emergent relationship between warming and future cumulative carbon emitted in our simulations to present an approximation tool making local warming projections from total future carbon emitted.

Could the Last Interglacial Constrain Projections of Future Antarctic Ice Mass Loss and Sea‐Level Rise?

Previous studies have interpreted Last Interglacial (LIG; ∼129–116 ka) sea‐level estimates in multiple different ways to calibrate projections of future Antarctic ice‐sheet (AIS) mass loss and associated sea‐level rise. This study systematically explores the extent to which LIG constraints could inform future Antarctic contributions to sea‐level rise. We develop a Gaussian process emulator of an ice‐sheet model to produce continuous probabilistic projections of Antarctic sea‐level contributions over the LIG and a future high‐emissions scenario. We use a Bayesian approach conditioning emulator projections on a set of LIG constraints to find associated likelihoods of model parameterizations. LIG estimates inform both the probability of past and future ice‐sheet instabilities and projections of future sea‐level rise through 2150. Although best‐available LIG estimates do not meaningfully constrain Antarctic mass loss projections or physical processes until 2060, they become increasingly informative over the next 130 years. Uncertainties of up to 50 cm remain in future projections even if LIG Antarctic mass loss is precisely known (±5 cm), indicating that there is a limit to how informative the LIG could be for ice‐sheet model future projections. The efficacy of LIG constraints on Antarctic mass loss also depends on assumptions about the Greenland ice sheet and LIG sea‐level chronology. However, improved field measurements and understanding of LIG sea levels still have potential to improve future sea‐level projections, highlighting the importance of continued observational efforts.