3.6 References co-cited analysis in hydrological connectivity research area
The co-cited analysis of references can obtain the scientific development and intellectual structure of related topics. Clustering analysis can aggregate literatures in related fields and identify the clusters (Ouyang et al., 2018). As shown in Fig. 8, the larger cluster means containing more terms, and the silhouette values of the top ten clusters we obtained are all over 0.5, which indicates that the cluster we generated is reasonable (Kaufman and Rousseeuw, 1990). The TF-IDF (term frequency-inverse document frequency), a commonly used weighting technique for information retrieval and data mining, is used to label the clusters. It can be seen in Fig. 8 that “amphibians” is the largest (#0) cluster, which also indicates that amphibians and biodiversity as well as ecosystem are the most productive area about hydrological connectivity. The second (#1) is the “channel network”, which means that soil erosion, sediment dynamics and river-floodplain interaction influenced by climate change and human activities also attract many researchers’ attention. Moreover, the study of channel network also provides better suggestions for the management of hydrological connectivity. Followed is “transport” (#2), the research of which is mainly focuses on the transport and utilization of water resources, agriculture and Nitrogen source. The No. 4 - No.9 clusters are “temporary rivers”, “hillslope scale”, “hyporheic zone”, “downstream waters”, “giws (geographically isolated wetlands)” and “soil”, respectively.
==============================================
Please Place Figures 8 here
==============================================
The timeline visualization displays the research time span of each cluster, as well as recent research hotspots. It can be seen from Fig. 9 that “transport” (#2) is the earliest research topic relate to hydrological connectivity and almost lasted from 1990 to nearly 2007, followed by “amphibians” (#0) from 1992 to 2005. In addition, “channel network” (#1) might have highly cited or citation bursts or both due to much of them has large size nodes, and the primrose yellow color means they attracted the most attention on the hydrological connectivity research recently. Moreover, giws (#8 “geographically isolated wetlands”) has the same primrose yellow as channel network, which means it is also a new hot topic. Studies have often focused on direct connectivity of water systems, ignoring the connectivity brought by geographically isolated wetlands might be affected by climate responses, such as rainfall and evapotranspiration. As an emerging research direction, the “geographically isolated wetlands” has attracted a lot of attention, and since it started later than the channel network (as shown in Fig. 8), it might indicate that geographically isolated wetlands will be the next research hotspot in the research of hydrological connectivity.