Study On The Fragmentation Of China's Landscape By Roads And Urban Areas

From an ecological perspective road construction in reality is a process of altering the habitat into "embankment and Row (right of way)", subsequently inducing formation of economic corridors and urbanization sprawel. In a landscape with roads scattered in it, the greatest risk to the persistence of animal populations and biodiversity is habitat loss and quality degradation, road kill, acoustic effect, deprivation of habitat, changes in runoff patterns, habitat changes due to drying or depositing process, migration barriers, alien plant invasion on roadside, landscape structural change, edge effects of forest roads will lead to a decline in habitat value, causing response of population size and behavior. And the roads superimposed with other natural and artificial factors, such as urban impervious surface, agricultural land, mountains and rivers will increase the disturbance and harm to the habitats. Therefore, the information quantization and relevant explore are the basic work needed to be carried out first in the field of "road ecology".In China, the significant landscape fragmentation factors are paved road with likely large traffic on it, these national, provincial and Township Roads, railways and urbanization areas will ultimately cause ecological fragmentation with the ecological flow disturbed due to their rapid expansion;. In order to provide constructive information in this regard, to make the planning sectors alert about the status of landscape fragmentation and be aware about the conservation in a timely manner, to study the present fragmentation at a national-scale is required. This study using mesh-size method launches the primary assessment of landscape fragmentation due to transport network in China. I Through assessment using Meff, we found a significant spatial pattern of landscape fragmentation caused by the major land transport systems and urban areas, less fragmentation in the vast western region, and most affected in southeastern China. The paved roads, railways and existing urbanized area are seriously affecting almost all the eastern provinces and counties in terms of ecological connectivity or integrity, especially around the major urban areas.ⅡBy provincial working unit in the assessment, the national-scale fragmentation trends from east to west shows 5 tiers, most severe in Shandong Province, to reduce west-oriented;ⅢChina's provinces, municipalities and autonomous region has been compared with the reported other parts in the world by using Meff value, a number of eastern provinces has become the most seriously fragmented landscape in the world, and some central and western provinces are relatively slight fragmented comparing with those less developed areas in Europe and California in United States.ⅣThrough the indicators which FRAGSTATS already has, we had California Central Valley (CCV) compared with a mature agricultural area (several counties in China's Jiangsu Province) in Yangtze River basin with a similar size (-40,000 km2) with CCV, California shown more smaller landscape patches than China's area, however it preserves several large area mountain patches, the area in Jiangsu Province is the lack of a large patch area except for some water bodies, most of the patch areas are within 20-1000 km2, medium-sized patches are dominant. Meanwhile we found California has a prominent feature of landscape division: certain kinds of patch area in all areas showed strong size distribution signature, including the area of 640 square acres (1 square mile) and their two partitions (such as 40 acres). In contrast, China has no obvious signature.Domestic landscape fragmentation studies on the different scales include:V Jiangsu, Zhejiang, Shanghai, Anhui, Jiangxi and Fujian study on fragmentation status quo along the FG1 to FG7 gradient;Ⅵcomparison by using other landscape index within and between typical provinces;Ⅶ, fragmentation status quo in the ecological zones in China;Ⅷfragmentation studies in both 8 endangered plant hotspots and IX on the whole China based on the mammals richness. The results show:in the 6 east China's provinces from Fg1 to Fg7 (newly adding fragmentation factors including rural road, cart road, water, mountains, farmland), when the rural road into the evaluation (Fg4), the fragmentation in the region were further increased, almost all the patches became less than the 1000Km2; after adding cart road (Fg5), fragmentation effects not obvious; after considering the water body (Fg6) the the county (cities or district) with Chaohu Lake, Taihu Lake, Poyang Lake, the Yangtze River flowing or existing in it has its fragmentation increased, with most counties' effective mesh size down to less than 500Km2, after considering the fields and mountains which higher than 2000m above sea level (Fg7), majority of counties have their effective mesh reduced to less than 200Km2. The effective meshes in Shanghai, Hangzhou, area around Ningbo, Anhui, Jiangsu and Jiangxi are less than 100Km2, fragmentation increasingly getting serious. We also chosed provinces (autonomous regions) such as Beijing, Shanghai and Tianjin; Jiangsu and Zhejiang; Hunan and Hubei; Xinjiang and Tibet in pairs or groups to carry out contrast by other landscape indices under Fg3 to show the supplementary landscape features, for example the mean patch size, the results show both in comparison between groups of 11 and only three municipalities group, MPS of Shanghai are the smallest. MPS largest in Tibet, and MPS Tibet＞MPS Xinjiang, MPS Beijing＞MPS Tianjin＞MPS Shanghai, at the same time, MPS Hunan, MPS Hubei. MPS zhejiang, MPS jiangsu MPS Beijing, are basically the same, slightly larger Jiangsu than Zhejiang, Hubei than Hunan. These results will help in comprehensive understanding of landscape fragmentation characteristics. An assessment on fragmentation based on the eco-geo-regions, threatened plant hotspots, and areas with high mammal richness in China shows the similiarity as the eastern provinces, some eco-geo regions have been listed among the most severe fragmentation landscapes in the world, while others are comparable to the least-developed areas in Europe and Ca. Threatened plant hotspots and areas with high mammal species diversity occurred in both highly and less fragmented areas, though future road development threatens already moderately divided landscapes.This result calls for the urgency and necessity of timely protection and mitigation.X In Gansu is a significant positive correlationship between the road length and several corridor features including patch number, section number, and number of nodes, also the number of townships; Pearson coefficient between patch number splited by NR and urban population is close to 8, then there is a correlation decline in turn with the road class hierarchy down. The determination coefficient (R2) between the distance gradient off subgrade and 8 landscape indicators all reaches 0.98. The above results indicate the road, human activities and the landscape pattern in Gansu have a close relationship between each other. Through network analysis we found in Gansu province along superposition of roads (Fg1-Fg5)αvalue reduced, showing the number of loop to reduce while the number of nodes increase, each county's a values tended to concentrated, low class cart road makes the loop rate slightly increased, indicating road network development Gansu is not perfect, linking roads are mostly low class or unclassified roads, such as dirt or gravel roads,βvalues show obvious dendritic structural features of road network in this province, together with Township Road, rural road, cart road, the unit node has 1.2-1.5 linking line on average. Theγvalues from Fg1-Fg5 tended to focus on the mean indicate the gap between the road network status and expectations achieved a shrinkage. Mean ofγdecrease along Fg1 to Fg4, indicating after joined the provincial, county and rural roads, the connected degrees is getting far off the ideal situation, also highlighting the dendritic network structure in this province and access from point to point is not convenient, largely rely on the main roads, cart road into, such network structure improve the situation to some extent.In terms of relationship between different class of roads, provincial density has a highest correlation with the Township Road (TR) (P-Coefficient: 0.93, P<0.0001), also there is a strong correlation between the densities of provincial road (PR) and TR, TR and rural road (RR); The correlations between national road (NR) and other 4 class road density are much lower than the 3 correlations between the provincial road and other 3 class of road density. TR related RR to a higher degree close to 0.8, but not with the cart road (CR) (R=0.63, P<0.0001); NR has a relatively highest correlation with the PR, reflecting a fact that the PR aligned based on the NR alignment. The importance of PR has been highlighted. Road network density and topography (elevation) associated weakly, and even showed weak negative correlation. As for the relations with economic dimension, the correlation between the road densities in Fg1-Fg5 scenario and forestry, animal husbandry, aquaculture and fisheries are very low. There is a weak correlation between GNP,2nd & tertiary industries and NR, PR density. G landscape feature divided by NR, PR, TR, railways and urban area in Gansu shows the division of andscape has its largest number of 60-120Km2 class (over 200 patches), followed by180-600 Km2 (160),30-60 Km2 (120),0.6-1 Km2 (100),10-20 Km2 (80),1-2 Km2 (85),5-10Km2 (85) and so on. Among which 6000-10000 Km2,10000-35000Km2 categories have the least number of patch, but largest area. Patch shape becomes more complex while the area increases. Through land use studies,we found that from 1980 to 2000 with a period of 20 years, within Gansu's 8Km wide corridor on both sides of NR, the rural residential land area has the largest increment (48.4525 Km2), followed by dry land increment (33.6825Km2), urban area (19.6326 Km2). Low & middle coverage grassland have the largest loss (38.396 and 30.95 Km2), high coverage grassland has relative less lose (about 11.457Km2). In addition, the irrigated field only has only 0.716 Km2 lost, sparse woodland, shrub land and forest lost respectively at amount of 5.022 km2,6.682 km2, and 3.611km2, "other woodland" incremental is about 6.1 km2, bare land, Gobi and marshes lost at amount of about 9 Km2, saline-alkali soil and desertified land increased by about 2 km2.Fragmentation study indicats, PR and railway enhance the fragmentation, RR has much less impact on the fragmentation pattern in Gansu, however the CR dividing the landscape further. In Gansu the fragmentation has expanding trends from high road density central area to the southeast, also enlarged in Hexi corridor. After urban area added into the factor combination, the impacts on the pattern are unobvious than CR. When farming land into the factor combination, the spacial pattern in the province did not change to a great extent, but the degree of fragmentation considerably deepened, the fragmentation area in Longnan and south Gannan area shows spreading outs. In consideration of highland fragmentating effect the degree of fragmentation in Qilian Mountainous area, Minxian mountainous area, Gannan and Longnan mountainous areas increased. The impacts from water bodies at 3 types of scale are likely unnoticeable.The present road network and urban sprawel have caused irreversible effects on the national landscape, especially on those ecological hotspots. The roads and urban area planned or under construction will have further impacts on the landscape integrity. For effective conservation of the threatened bio-diversity and landscape integrity, we advocate any regional and national planning shall reflect the existing land fragmentation pattern, all these efforts put inside are for realization of a sustainable and eco-friendly road network establishement and urban development.