High-intensity human activities have changed land use/land cover (LULC) patterns in the Huangshui River Basin (HRB), which has brought significant challenges to ecosystems sustainable development. Discerning ecosystem service dynamic characteristics and responses under different land use/cover change (LUCC) scenarios are necessary to increase the public willingness to pay for the ecosystem and guide the decision-making process. We examined LULC spatiotemporal dynamics in the HRB from 2000 to 2020 and coupled Markov-chain, multi-objective programming (MOP), and patch-generating land use simulation (PLUS) models to optimize and simulate LULC spatial pattern for five scenarios: natural development scenario (NDs), city expansion scenario (CEs), ecological protection scenario (EPs), economic development scenario (EDs), ecological and economic balance scenario (EEBs). Given regional differences, a spatially modified ecosystem service value (ESV) assessment model was proposed to evaluate ESV. Factors driving ESV spatial stratified heterogeneity were identified using geographic detectors. Ecosystem service value sensitivity in response to LUCC was discriminated against using the elasticity model. The study area was dominated by 56.86–60.40 % grassland and 33.11–36.27 % cropland. Grassland and cropland area decreased by 579.75 km2 and 423.87 km2 over the period 2000–2020, while the other areas such as forestland, water area, construction land, and barren land increased by 289.81 km2, 140.77 km2, 489.10 km2, and 83.96 km2, respectively. Land conversion mainly occurred among grassland, cropland, and construction land. Total ESV was 39,665 million yuan in 2020, an increase of 2.25 % compared to 2000. Total ESV of NDs, EPs, EDs, and EEBs increased by 0.34 %, 1.04 %, 2.01 %, and 7.78 %, respectively compared to 2020, while that of CEs decreased by 0.17 %. ESV sensitivity coefficient response to LUCC was 0.43 during 2010–2020, indicating that LULC conversion of 1 % would result in average changes of 0.43 % in ESV. Ecosystem services response to LUCC was not very marked in the HRB. Elevation was the dominant driver of LULC and ESV stratified spatial heterogeneity in the HRB. The effects of elevation on LUCC and ESV should receive more attention in management. Multi-objective optimization and multi-scenario analysis effectively guided land-use planning and decision-making involved in ecosystem uncertainty, complexity, and interaction. The EPs and EEBs may be more suitable for future HRB development. © 2022 The Author(s)