Abstract: Against the backdrop of the deepening interplay between rapid urbanization and ecological environmental change, revealing the nonlinear effects of urbanization on land ecological functions and their spatial differentiation constitutes a fundamental basis for safeguarding regional ecological security and optimizing the spatial pattern of territorial development. Vegetation Net Primary Productivity (NPP) can effectively characterize the material production and carbon sequestration capacity of regional ecosystems, serving as a key indicator of changes in land ecological functions. To this end, this study takes Heilongjiang Province as the study area and constructs an integrated analytical framework comprising nonlinear threshold identification, zonal characterization, and pathway decomposition. By integrating the urbanization intensity index, locally weighted regression, and partial least squares path modeling, the study systematically elucidates the synergistic mechanisms through which urbanization and natural factors jointly influence NPP. The results showed that: 1) The growth of urbanization in Heilongjiang Province exhibited a spatial pattern of "concentration in the south and dispersion in the north": hotspot areas were concentrated in the southern core cities and their connecting corridors (represented by the main urban area of Harbin), while cold spot areas were scattered across parts of the northern and eastern regions. This differentiation was constrained by the dual constraints of the ecological redline management in the Greater Khingan Mountains and the agriculture-driven development of the Sanjiang Plain, indicating that urbanization did not proceed evenly but rather followed an evolutionary pattern dominated by agglomerative expansion of core cities, with relatively slow and dispersed development in peripheral areas; 2) A significant nonlinear threshold effect existed between the composite Urbanization Intensity Index (UII) and NPP, with the breakpoint located at UII = 0.66. In the low-urbanization stage (UII = 0.66), a trend of NPP recovery emerged, and this turning point showed temporal synchrony with the entry of urban development into a new stage; 3) The zonal mechanisms by which urbanization influenced NPP differed significantly: β represents the standardized path coefficient, with its sign indicating the direction of the effect and its absolute value reflecting the relative strength of the effect. In regions with a low urbanization growth rate and high NPP, urbanization exerts a significant direct inhibitory effect on NPP (β = −0.428); natural factors mainly play a limited buffering role that is insufficient to offset the negative impacts of direct urbanization disturbance, yielding a total effect of −0.263. In regions with a high urbanization growth rate and low NPP, the direct effect of urbanization on NPP is weak (β = −0.028), but urbanization weakens the positive supporting role of natural factors, thereby superimposing and amplifying direct and indirect negative effects, with a total effect reaching −0.381. In regions with a high urbanization growth rate and high fluctuation of natural factors, urbanization directly inhibits NPP (β = −0.321), exerts a positive effect on natural factors (β = 0.410), and natural factors positively drive NPP (β = 0.359), thereby mitigating the direct inhibitory effect of urbanization to some extent and reducing the total effect to −0.174. The results indicate that the impact of urbanization on regional vegetation carbon sequestration capacity exhibits pronounced threshold characteristics and zonal differentiation, with its pathways modulated by urbanization intensity and regional natural background conditions.