A Study on the Changes of Landscape Pattern of Estuary Wetlands of the Minjiang River

The Mingjiang River is a large river that rises from the east of the Wuyi Mountains of Jianning county in Fujian province of the southwest China. The climate type of study region is subtropical maritime monsoon climate, characterized by warm and humid days and distinct seasons. The average annual wind speed is 3.4 m/s. The annual precipitation is abundant and annual amount of precipitation is 1 346 mm. The average annual temperature is 19.6 ℃. In general, the monthly average of temperature in July is the highest during a year, and the lowest occurred in January.The Minjiang River has a great flow and low silt content, being a typical mountain river. Affluents of the Minjiang River are densely distributed, among which are the Anrenxi River, the Meixi River, the Daidangxi River, the Daimuxi River and the Yingquanxi River. The soils in the estuary region of the Minjiang River are mainly paddy soils and laterites. There are two kinds of natural vegetations in the drainage area; they are tropical rain forest of South Asia and tropical forest of central Asia. Due to the diversity of natural conditions, vegetation types are complicated, including green foliage forest, bamboo grove, brush, thick patch of grass, Shashenglin of coast and so on. Because of the interaction between the rivers and sea, there is a great diversity of the wetland landscapes. There is the richest wetland resources in the estuary region of the Minjiang River of Fujian province, but at the same time, the environment of wetlands is very vulnerable due to disturbances of human activities as a result of the influence of social and economic development in recent years. The area of the wetlands is reduced, the function of wetlands degenerates, and the environment was endangered as a whole. Quantitative analyses to all kinds of landscape pattern and their dynamic changes of wetlands were made in the estuary of the Minjiang River, in which Remote Sensing and Geographical Information System were applied and Landsat TM images were used in 1986 and 2002. The indexes reflecting the spatial structure and the heterogeneity of landscape pattern were selected based on the ecology principle of the landscape pattern. All the analyses were performed using GIS and the landscape pattern analysis software,FRAGSTATS.Selected landscape indexes were class area (CA), percent of landscape (PLAND), patch richness (PR), Shannon~ , s diversity index(SHDI), Shannon~ , s evenness index(SHEI), mean patch fractal dimension(MPFD) , number of patches (NP),the mean patch size(MPS), patch size coefficient of variation (PSCV), patch density(PD), patch cohesion index(PCI) and contagion index(CONT). The results showed that (1) the overall landscape area reduced by 10 007.19 hm~ 2 ; NP value was reduced by 34 from 1986 to 2002; the SHDI value increased from 1.129 8 in 1986 to 1.184 3 in 2002; and the SHEI value increased from 0.580 6 to 0.608 6 too. To some extent, rivers and paddy fields performed the function of controlling landscape pattern better, the others performed worse relatively. And the eigenvalue of indexes was also changed. For example, PD value increased from 0.465 3 in 1986 to 0.493 0 in 2002 per 100 hm~ 2 , MPFD value was 1.083 0 and PSCV value was 546.161 0 in 1986, they increased to 1.090 0 and 586.404 9 respectively in 2002, in contrast, CONT value decreased form 68.894 9% to 67.745 5%.In conclusion ,due to the disturbance of human activities, the degree of patch fragmentation further increased ,the difference between the patch size was enlarged, the separate degree increased, the connectivity was weakened, the irregularity of patch shape and the bigger isolation between patches as a whole.(2) The indexes of all kinds of landscape pattern with the more complicated structure have been changed too. By analysis of selected indexes, it was found that there were obvious different changes on various landscape patterns over 16 years. The area of wetland landscape has reduced in some degree besides the breeding farm. In respect of land use and land cover, the most dominant landscapes were paddy fields and rivers, and their influences have been lessened slightly. The total area of paddy fields and rivers was 47 232.94 hm~ 2 in 1986, but, it dropped to 38 407.33 hm~ 2 in~ 2002,with its proportion dropped from 77.52% to 64.25%. As the patch fractal dimension analysis implicated, the patch density of the reservoir, the breeding farm, and the paddy field increased, the degree of patch fractal dimension was further strengthened, and other wetland landscape types were opposite. The results of analysis of patch shape indexes showed that the fractal dimension of reservoir and mud sand reduced and the patch shape turned into simple and regular. The patch cohesion index of the reservoir, the sand beach and the paddy field were decreasing. The patch separate degree of those three landscapes increased, the connectivity was weakened, but the connectivity of other landscapes was strengthened actually. It is obvious that the phenomenon of mutual transformation and reverse among landscape patterns was due to the interference of human activities and natural reasons. The spatial variation of the riparian landscape along the river was affected by a suite of natural and anthropogenic factors, such as temperature,precipitation,soil,and various human activities.Human population density was one of the most important factors.Rapid population growth and economic development have significantly influenced the spatial pattern of the landscape of rivers and the paddy fields.Quantifying the landscape pattern and its change should provide useful information for effectively monitoring,assessing,and managing these ecosystems.