Abstract: Long-term underground coal mining activities cause severe land subsidence and form large amount of subsidence ponds, and causing water pollution in different degrees. To study the distribution of heavy metals in subsidence pond and cause of formation, this paper chooses subsidence pond in Panji-1 coal mine in Huaihuan as research area and focused on eight kinds of heavy mental elements(Fe, Mn, Zn, Cu, Cr, Cd, Pb, Ni) impacting significantly environmental quality. We took water reflectance spectrum as covariate to estimate the spatial distribution characteristics of heavy metals in water bodies on the basis of collaborative kriging method. The results indicate that measured water spectra and heavy metal content have good relations. And collaborative kriging interpolation method which chooses water spectrum as covariate is much better than single variable ordinary kriging interpolation, the root mean square error of eight heavy metal elements between the predicted values and actual values decreased obviously, proving that the reflectance spectra of water as covariate is suitable for estimation of the spatial distribution of heavy metals in water. Furthermore, the comprehensive analysis results found that Cd, Pb, Cu, Ni in water-looged area mainly come from gangue dump in the northwest of the water body. And Cd, Cu and Pb concentrations exceeded local background value and had the highest pollution risk. Cr mainly comes from agricultural fertilizer, soil parent material and coal ash deposition of the surrounding road and coal gangue. Zn mainly comes from coal gangue, sewage, agricultural fertilizer and soil parent material. Because of its concentrations is low, it has a little effect on the water environment quality.