Gradient Analysis Of Dry Valley Of Minjiang River Landscape Pattern, Based On Moving Window Method

Great attentions have been paid on landscape changes in recent years. As a basisof further researches on landscape functions and dynamics, contributes to analyzingspatial distribution characteristics of landscape components, landscape pattern analysisbecame an important topic of landscape ecology. In landscape pattern analysis, landscapemetrics has been used as a common tool to exhibit the spatial distribution of landscape.During the past two decades, many processes have been used to analyze changes oflandscape and the relationships with human influences and environmental factors. As ahotspot of quantitative methods of landscape analysis, gradient analysis, which isconducive to revealing the evolutionary laws of spatial landscape patterns has became asignificant means in landscape ecology. However, the appropriate spatial scale is the keypoint to calculate the landscape metrics, and scale issues represent one of the foremostfrontiers of landscape ecology. It is well known that the observed landscape pattern andits relationship with (landscape) process depend upon the scale. Great developments havebeen made on researches on pixel and spatial extent of landscape pattern evolutionquantitatively and qualitatively. Little has been known on the accuracy of the landscapeanalysis is related to both pixel and extent effect since is the lackage of systematic studyon the selection of optimization scale in landscape gradient pattern analysis. In this study,taking the dry valley of Minjiang River as study area and using ARCGIS9.3, ENVI4.8and FRAGSTATS3.3, based on the image data of Land Resource Satellite ETM+(2000,2011) and Land Resource Satellite TM(2006),we aim to analyze the variation oflandscape index with grain size. To obtain gradient patterns of the landscapes,4transectshave been set up along both mainstream and tributaries of Minjiang River. A series ofmetrics at the class level(PLAND, MPS, LPI, and PD) and the landscape level (NP, PD,LPI, DIVISION, SHDI, and SHEI) were chosen and calculated using standard andmoving window approach with different spatial scale, respectively. Then, anoptimization-scale selection method, which integrates: the grain effect analysis oflandscape index, the data loss assessment and the landscape index range effect curve analysis, was developed to obtain accuracy and efficient scale. By analyzing the graininflexions of the landscape index comprehensively, we found that30-90m and110-160mwere the appropriate grain ranges. Data loss assessment showed that50m was theappropriate grain extent. In addition, based on moving window analysis, landscape indexrange effect curve analysis suggested that250m was the most appropriate spatial extentfor landscape pattern analysis. At the landscape level,6metrics were calculated byMoving window method: shrub land, accounting for73.8%of the total landscape, wasthe matrix in dry valley of Minjiang River in the year2000-2011; the forestland andgrassland’s landscape heterogeneity decreased and relatively high the fragmentationdegree of construct and farmland; water area has no obvious change; landscape metricsin the four transects present different amplitude and evident gradient diversity aslandscape type change, the index change is larger in transition zone as compared to thesingle region; landscape pattern develops in the direction of heterogeneity, diversificationand homogenization; and topography, precipitation, temperature and human activitieswere the factors for the gradient changes of landscape. As a kind of beneficial attempt,this study more finely analyzed the landscape pattern in the study area. The result oflandscape pattern gradient analysis provided a novel way for discerning the landscapepattern change in the mountainous areas. However, the study have some deficiencies, weneed further research in the future work.