| Summary: | Boreal forest landscapes have undergone severe anthropogenic fragmentation and their enormous values concerning, e.g., ecosystem service, biodiversity and culture are hazarded. Maintenance of the remaining intact boreal forest landscapes and restoring the structural and functional connectivity among these remnants are essential for boreal forest conservation. In Sweden, such conservation tasks are highlighted in planning and implementing green infrastructure (GI). At the national level, the GI is established upon a network of high conservation value forests (HCVF). However, HCVF are insufficient to guarantee a functional GI and thereby cannot effectively halt boreal forest degradation. Forest restoration is urgently needed. This study explored a restoration approach based on connectivity forest (CF), i.e., forest areas with intermediate to high conservation likelihood prescribed by a new GIS-empowered artificial intelligence model. By step-wisely inserting the CF into the current GI, represented by the HCVF, this study assessed how the GI was reconfigured and strengthened over 1.3 million hectares of boreal landscapes in northern Sweden. First, this study demonstrated good restoration potential in all three subregions of the study area (Mountainous, Inland, Coastal), since the total area of CF within a subregion accounted for at least 11% of the corresponding subregional forest area. Second, by evaluating the GI-area increase and the GI-density variation, this study showed the inland and coastal subregions, much underrepresented in the current GI configuration, might have a higher sensitivity to the CF-insertions than the mountainous subregion. By adding the CF, the GI-area was increased by over 400% in both the inland and coastal subregions versus 60% in the mountainous subregions. The GI-density increase, achieved per unit CF area input, was higher in the inland and coastal than in the mountainous subregion. However, with the proposed insertion of CF, the GI-patches in these two subregions were still scarce, ...
|
|---|