Water-use efficiency (WUE) is a key indicator of the interactions between the water and carbon cycles of terrestrial ecosystems. Understanding the spatiotemporal characteristics of WUE and the climatic factors that affect it could guide sustainable management of water resources and ecosystem services in areas lacking water. The Haihe River basin is located in a region sensitive to climate change and human activities, and water shortages have constrained sustainable development of regional agricultural and ecological environments. Using 2000-2014 Moderate Resolution Imaging Spectroradiometer (MODIS) and meteorology data, we investigated the spatiotemporal changes in gross primary productivity (GPP), evapotranspiration (ET), and WUE in the Haihe River basin using linear trend analysis, correlation analysis, Mann-Kendall (M-K) tests, and other statistical methods, revealing four key results. (1) The change in annual GPP trended upward non-significantly (R²=0.1784, P > 0.1), and ET trended downward non-significantly (R²=0.0269, P > 0.1). Annual WUE significantly increased over 2000-2014 at a rate of 0.0185 gC/kg H₂O a^-1 (R²=0.6299, P < 0.01). (2) The spatial patterns of average annual WUE and GPP in the Haihe River basin's vegetated areas varied widely, and the annual average WUE and GPP values decreased from the southeast to the northwest. High WUE and GGP values were mainly found in the North China Plain agricultural ecoregion and the Tianjin-Beijing-Tangshan urban and suburban agricultural ecoregion. The lowest WUE and GGP values were mainly found in the East Central Inner Mongolian Plateau typical steppe ecoregion, Loess Plateau agricultural and steppe ecoregion, and the deciduous forest ecoregion northwest of the Yanshan-Taihang Mountains. The WUE of the Haihe River basin increased, with WUE values of 91.11% and 60.17% of the Haihe River basin's vegetated areas increasing and increasing significantly (P < 0.05), respectively. That of the Loess Plateau agricultural and steppe ecoregion increased the most significantly. (3) The WUE means and trends among the different land use types also varied significantly. The mean WUE of croplands (CRO) was highest, and the WUE of grasslands (GRA) was lowest. The WUE of CRO, woody savannas (WSA), and GRA each increased significantly (P < 0.05). (4) Annual WUE was mainly affected by precipitation, and 41.44%, 33.23%, and 16.01% of WUE estimates in the Haihe River basin's vegetated areas were dominated by inter-annual variation in precipitation, drought, and temperature, respectively. The vegetated areas in which WUE was primarily shaped by precipitation were mainly found in the northern portion of the Yanshan-Taihang Mountains deciduous forest ecoregion, the North China Plain agricultural ecoregion, and the Tianjin-Beijing-Tangshan urban and suburban agricultural ecoregion. The vegetated areas in which WUE was primarily shaped by drought were mainly found in the northeast of the North China Plain agricultural ecoregion and the southern portion of the Yanshan-Taihang Mountains deciduous forest ecoregion. The vegetated areas in which WUE was primarily shaped by temperature were mainly found in the western portion of the Yanshan-Taihang Mountains deciduous forest ecoregion and the East Central Inner Mongolian Plateau typical steppe ecoregion.