Almost all ecosystems are open systems, meaning that significant changes in one ecosystem can lead to unexpected changes in others. Deer overabundance has become a problem worldwide, leading to forest degradation that has impacts not only on terrestrial but also aquatic ecosystems. This talk presents studies on the indirect effects of deer-induced changes in terrestrial ecosystems on Japanese stream ecosystems, and inter-stream order variation of patterns and timings of these changes that tightly correspond with hydro-geological processes. The earlier sections present case studies in the Ashiu research forest of Kyoto University that investigated the processes of deer-induced ecosystem alteration based on the continuous monitoring of certain streams and their catchments. In the forest, until the 1990s, forest floors were covered by dense vegetation. However, deer-grazing damages on forest floor vegetation increased from the late 1990s, and most forest floors were denuded until 2006 in the forest. In first-order streams, changes in aquatic ecosystems induced by changes in their catchment are relatively immediate. Fine sediments are directly supplied by slopes that are denuded by deer grazing, and the deposition of these leads to changes in macroinvertebrate assemblages, such as the increase of burrowers. In second-order and third-order streams, although fine sedimentation is less obvious than in first-order streams, forest degradation resulted in the notable decrease of collector-filterer macroinvertebrates 8 years after the forest degradation. Further, in fourth-order streams, fine sediments are transferred from headwaters upstream by stream flow, and there is an approximately 10 years of time-lag from the forest degradation. The later sections present a more general pattern of deer-induced effects that occur at a regional scale using conventional snorkeling observation and the quantitative eDNA metabarcoding method for estimating fish densities. The results indicate changes in stream ecosystems observed in previous studies in the Ashiu forest have a non-linear continuous pattern; fine sediments on riverbeds increased in the 3–15 years following forest degradation, as seen in previous studies, while decreasing after 16 years. Population densities of sand-preferring fishes, as inferred from both eDNA and snorkeling, reasonably followed this pattern. These patterns may be caused by the depletion of fine sediments on mountain slopes. The regional survey suggests deer-induced changes in stream ecosystems generally occur throughout Japan. Meanwhile, the local time-series monitoring has contributed to understanding the background processes driving regional-scale ecosystem alteration. Therefore, it is emphasized that multiscale perspectives are important for landscape ecological studies, both phenomenologically and methodologically.