Projecting spatial interactions between global population and land use changes in the 21st century

The study is motivated by the UN Sustainable Development Goals, particularly SDG 11, which calls for inclusive, safe, resilient, and sustainable cities. Rapid urban land expansion driven by population growth creates air pollution, resource depletion, and strains on infrastructure, especially in developing countries, threatening progress toward SDG 11. While assessing urban land expansion relative to population growth is important, it often neglects explicit trade-offs between land consumption and ecological services. Monitoring limitations have also hindered exploration of spatial interactions between global population and land use changes. This research develops a multiple-source satellite-image-driven framework to analyze trade-offs between land expansion and population changes (TLEPC) from 2010 to 2100. It asks: What is the significance and purpose of a trade-off development strategy, and what are its impacts on ecological services and food production under different scenarios? What is the most sustainable path for global development under SSPs, and which regions have the greatest potential for improving land consumption? By situating projections within the Shared Socioeconomic Pathways (SSPs)—sustainability, middle of the road, regional rivalry, inequality, and fossil-fueled development—the study examines how population, economy, land use, energy, and emissions may evolve and how urban expansion under SSPs affects habitats and biodiversity. The purpose is to provide a data-driven, policy-relevant framework for balancing urban development with ecological preservation across countries.

The paper situates its work within SSP-based projections used to explore demographic, economic, land-use, energy, and emissions trajectories in the 21st century. Prior studies mapped global urban land under SSPs and assessed impacts of urban expansion on terrestrial biodiversity, finding urbanization generally consumes previously uncultivated land and can have different implications than farmland expansion. Urban growth affects fragile habitats (e.g., drylands), intersecting with SDG 15 (life on land). Research has also examined uneven greening of built-up areas across economic levels. Earlier LEPC analyses at high spatial resolution under SSPs exist, but optimal global-scale trade-off solutions have been underexplored. This study extends the literature by explicitly modeling trade-offs between land consumption and population dynamics using multi-objective optimization and by quantifying implications for ecological services and food production across scenarios.

Data: The study aggregates panel data for 153 countries (2020–2100) from the SSP database (population, urbanization rate, GDP). Country-level land use and population change data (2000–2100; 1/8° grid) are obtained from the CLUE_S-based dataset (doi:10.7910/DVN/85PJ1D). Built-up area (BUA) boundaries are from Natural Earth; additional urban population data from WorldPop. Satellite-based multi-source inputs and expert knowledge underpin analyses. Modelling framework: The authors develop a data-driven framework linking land expansion and population change (LEPC) at national and regional scales with 1/8° grid resolution at decadal steps from 2010 to 2100. They simulate five SSP scenarios (1–5: sustainability, middle-of-the-road, regional rivalry, inequality, fossil-fueled development) using the CLUE_S model to predict urbanization paths by country. Trade-off modeling and optimization: A multi-objective optimization seeks to (1) minimize global BUA land consumption and (2) maximize per capita built-up areas (PBUAs), solved via Monte Carlo simulation to approximate the Pareto frontier. Constraints incorporate population change rate (PCR), land change rate (LCR), and their ratio PCR/LCR. LCR is computed at the country scale by LCR = (BUA_t1 − BUA_t2)/(n − 1) × 100% (PCR and GCR analogously), with APGDP, APPOP, and APLGDP as per-capita differences over the period. Objective functions are: g1(x) = 1/(U_2100 − U_2020) to minimize urban land consumption; g2(x) = Σ_2020^2100 P_urbanland to maximize PBUA. Factor analysis (SPSS) identifies synergistic drivers among population, land, and GDP indicators. Monte Carlo sampling explores feasible trajectories under national and global constraints; Pareto-optimal solutions are selected based on turning points and distance to the frontier’s upper edge. Validation and analysis: The framework is applied globally and by income groups, with policy-informed validation. Additional analyses quantify impacts on carbon sequestration (forest and grassland sinks) and food production (cereal yield), using coefficients from prior literature and World Bank yields (2019–2021).

- Global PBUA disparities (2020): High-income average 317.99 ± 176.49 m²; upper-middle-income 231.61 ± 181.94 m²; lower-middle-income 116.88 ± 146.48 m²; low-income 189.81 ± 162.84 m². High-income urbanization rate 81.53 ± 10.79%; low-income 40.69 ± 13.08%. 60% of middle-income countries with high PBUA are in Europe; 71.42% of low-income countries with high PBUA are in Africa and Asia. - Projected coordination of population and land expansion: Ratios of population to land expansion under SSPs 1–5 are 3.8977 ± 1.4342, 2.3452 ± 0.4291, 4.0865 ± 0.7365, 3.3264 ± 1.3070, and 2.8133 ± 1.1307. PBUAs under SSP1 reach a low around 2040–2050; SSP2–SSP5 increase rapidly after 2040. Regulating the ratio of population increase to land expansion can reduce post-mid-century urban land growth. - Economic projections: Per-capita GDP of BUAs by 2100 (vs. 2010) may grow by ~520% (SSP1), 258% (SSP2), 202.67% (SSP3), 182.6% (SSP4), and 319.1% (SSP5). - Factor analysis of synergistic drivers (2010–2020): Three factors represent trade-offs among land, population, and GDP; variance explained by Factors 1–3: 90.329%, 9.664%, 0.00319%. Positive PCR–LCR–GCR relationships indicate land–population synergy; GCR/LCR, APGDP, APLGDP indicate land–GDP synergy; APPOP and negative PCR/LCR reflect population–GDP dynamics. - Optimization of trade-offs (TLEPC) reduces land consumption substantially: By 2100, land consumption reductions vs. baseline are 8.07% (−0.08681 million km²; SSP1), 28.2% (−0.61 million km²; SSP2), 13.76% (−0.187 million km²; SSP3), 23.92% (−0.48 million km²; SSP4), and 55.28% (−1.92 million km²; SSP5). Countries where urban expansion exceeds population growth have high potential to cut future land consumption. - Urban land growth rates (2010–2100): 61.7% (SSP1), 208.6% (SSP2), 104.2% (SSP3), 187.11% (SSP4), 317.9% (SSP5). Asia and Africa growth rates are 1.5–2.9× other regions under SSPs 1–5. - Regional/sectoral implications: EU founding states’ PBUAs by 2050 expected to be 1.72–2.01× BRICS PBUAs across SSPs. BRICS average PBUA under optimization remains within 152.82, 196.16, 164.92, 187.99, and 212.88 m² (SSPs 1–5) by 2050. Megacity analyses (2000–2018) show rapid urbanization in some Chinese cities (e.g., Shanghai, Beijing) and slower patterns in EU cities and New York. - Ecological services: Trade-off methods increase per-person carbon sequestration potential (per decade) via forests by 0.0735 ± 0.0397 TC (SSP1), 0.4247 ± 0.2725 TC (SSP2), 0.1297 ± 0.0781 TC (SSP3), 0.284 ± 0.168 TC (SSP4), 0.7418 ± 0.5380 TC (SSP5); for grasslands by 0.0022 ± 0.0015 TC, 0.0211 ± 0.0133 TC, 0.0051 ± 0.0038 TC, 0.0146 ± 0.0086 TC, and 0.0413 ± 0.0302 TC, respectively. Under SSP1, carbon sink can be reduced to improve settlements by up to 74.4% of African countries and 57.89% of Asian countries, while 87.5% of European and 32.14% of American countries could slightly increase sequestration with slight PBUA reductions. - Food security: Compared to baseline SSPs, the trade-off path increases cereal yield in 2030 by 19.798 MT (SSP1), 74.589 MT (SSP2), 26.532 MT (SSP3), 60.965 MT (SSP4), and 121.414 MT (SSP5), based on World Bank average yields (2019–2021). - Policy-relevant insight: SSP1 emerges as the most sustainable path with least land consumption under current policies; trade-off strategies significantly reduce land use while supporting ecological services and food production.

The findings demonstrate that explicitly managing trade-offs between population growth and urban land expansion can substantially reduce global land consumption while improving or maintaining ecological services and food production. The optimization shows large potential reductions in built-up area expansion across all SSPs, with SSP1 offering the most balanced pathway consistent with SDG 11. Regional analyses suggest that Asia and Africa will experience the highest urban land growth, necessitating targeted trade-off policies to avoid excessive land consumption and ecological degradation. For developed regions, minimizing land consumption while maintaining adequate PBUA is feasible; for rapidly urbanizing regions, Pareto-optimal strategies balance human settlement needs with conservation. The carbon sequestration analysis indicates that modest adjustments to PBUA can enhance forest and grassland carbon sinks in many countries, though some regions may prioritize settlement improvements at the expense of some sink capacity. Observed megacity patterns corroborate the model’s ability to capture heterogeneous urban growth dynamics. Overall, the trade-off framework addresses the research questions by quantifying how strategic coordination between land expansion and population under SSPs can support sustainable urban development and ecological protection, providing concrete, scenario-based policy levers for different national contexts.

This study introduces a data-driven, multi-objective optimization framework that links land expansion and population change under SSP scenarios to identify trade-off strategies (TLEPC) that minimize land consumption and maximize PBUA. Applying the framework globally at country scale shows substantial potential reductions in urban land use—especially under SSP1—while supporting carbon sequestration and improving projected cereal yields. The work contributes a practical tool for planning urban development that aligns with SDG 11, highlighting where and how countries can adjust development to achieve more sustainable outcomes. Future research should refine socio-economic and ecological drivers at finer spatial scales (cities, special economic parks, rural areas), leverage new data sources (mobile signals, points-of-interest), and develop decision-support software for spatio-temporal optimization of PBUAs tailored to local policy contexts.

The study acknowledges data limitations that constrained exploration of socio-economic and ecological factors, including limited monitoring data and unavailability of adequate, detailed inputs for all countries. Countries with insufficient resources often have lower urbanization rates, and optimal strategies for improving these rates warrant further investigation. The model relies on scenario assumptions (SSPs) and CLUE_S-based projections, which carry inherent uncertainties. Carbon sequestration estimates depend on coefficients from prior studies and may vary by local conditions. Future work aims to incorporate richer multi-source datasets (e.g., mobile signals, POIs), examine smaller regions, and enhance decision-support tools to better capture local dynamics and uncertainties.