Keywords

1 Introduction

Numerous studies have showed that Landscape is not only the typical scale dependence, but geographical and historical interactions related to ecological system. Obviously, landscape diversity index of Land-use types for grain size and scale changes in response to different sensitivities. At present, land use landscape pattern From Xiamen City, Guangzhou, Shanghai City, Pingyin County in Jinan and Jinghe Watershed landscape had showed landscape diversity and sensitivity of landscape pattern, which had an important impact on landscape pattern of land-use types whether between the landscape pattern index, or between different scales of landscape index [1–4].

From the scale perspective, there is better foundation in Yucheng city, one of the network stations of CAS. With the development of urbanization, information and modernization, land-use landscape pattern and process evolution made rapid changes. Therefore, the case of Yucheng city has great promotion value for the more similar country-level scale in China. Further retrieval of previous literatures have shown that land use landscape diversity and sensitivity of Yucheng City was not retrieved according to the new land use classification standard, the study had few in related landscape analysis, the current research situation is not commensurate with its status for Yucheng city, in the meanwhile, it is not conducive to the landscape of the overall planning and the process of urbanization process in Yucheng City.

2 Materials and Methods

2.1 Study Area Situation

Yellow-Huaihe Rivers Plain is the largest plain in China and an important area of grain, cotton, oil. Yucheng city is the part of Yellow-Huaihe Rivers alluvial Plain and located in the northwest of Shandong province in China, between 116º22′11″–116º45′00″ E and 36º41′36″–37º12′13″ N. The total area of the city is 990 km2, the study area belongs to semi-moisture monsoon climate area and has on average 2639.7 h of sunshine per year. The total radiation of sun is 124.8 K/cm2. The average temperature per year is 13.1°C, the 200 frost-free days, over 10°C and over 15°C in accumulated temperature are 4441°C and 3898°C in the study area, which provides plenty of thermal conditions, the average rainfall per year is 666 mm. The study area is the part of Yellow-Huaihe Rivers alluvial Plain, the site physiognomy is comparatively complex, there are 7 kinds of landforms in all, including flood land plateau, high land, even land, low-lying land, shallow land, sector crack land and arenaceous river channel. On the basis of topography, landform, parent material and climate, there are two kinds of soil types, Fluvo-Aquic soil and Solonchak. Salt-affected lands are small distributed in the study area, dynamic changes of land use, especially temporal and spatial changes of salt-affected lands, has an important role to improve land quality and promote agricultural sustainable development in the study area [5].

This paper depends on ETM image combined with land-use other data. In order to improve the accuracy of RS image, we referred to the data, 1:50 000 topography map in scale and other spatial maps. The same scale (1:50 000) maps, such as groundwater salinity map, groundwater depth map, soil organic matter content map, soil texture and configuration map. In the study, the author carried out the new land-use classification system, involved in 12 landscapes, they are Arable-land, Grassland, Traffic and Transmission Land Use, Residential land, Public management and service land, Commercial service land, Garden plot land, Mine and storage land, Woodland, Water and water facility land, Special land and other land, respectively.

2.2 Scale Effect System and Landscape Significance

Different landscape index has different ecology significance, the selected landscape index which is applied to analyze the scale effect is illustrated in Table 1 [6].

Table 1. Ecological significance and range of Index
Full size table

3 Results and Analysis

3.1 Correlation Analysis of Landscape Index

The correlation degree analysis results of six selected index in Yucheng city were showed in Table 2, which disclosed features and changes tendency of land-use landscape spatial patterns.

Table 2. Ecological significance and range of Index
Full size table

Table 2 summarized the correlation results for the key variables. 6 ecology index (PD, LSI, PAFRAC, CONTAG, SHDI and SHEI) was significant at the 0.01 level, the correlation coefficient was 0.956, −0.776 and 0.845 between PD and LSI, PD and PAFRAC, PAFRAC and CONTAG, respectively. There is significant Correlation between LSI and PD, LSI and PAFRAC, LSI and CONTAG, PAFRAC and CONTAG, the correlation coefficient was 0.956, −0.911, 0.961 and −0.976 at the 0.01 level, especially, the significant coefficient was reflected between SHEI and SHDI, the more significant correlation coefficient was 0.999.

3.2 Sensitivity Analysis and Spatial Pattern of 12 Landscapes

The typical scale effect was selected and explored in the study, including 5 m, 10 m, 15 m, 20 m, 25 m, 30 m, 40 m, 50 m, 60 m, 70 m, 80 m, 90 m, 100 m, 110 m, 120 m, 150 m, 180 m and 210 m. The series results were clarified by PD, LSI, PAFRAC, CONTAG, SHDI and SHEI responding to the different scales, or 18 different scales, the results were clarified in Table 3.

Table 3. Different scale statistics of landscape index
Full size table

According to Table 3 results, the decreasing trend was presented from 5 m scale to 210 m scale for the index, LSI and CONTAG. The increasing change of PD index was illustrated in the case area from 5 m scale to 210 m scale, furthermore, there was an important point, reflected on 25 m scale, or the decreasing trend of PD index was presented from 25 m scale to 210 m scale. The increasing change of PAFRAC was taken place from 5 m scale to 180 m scale, but on the key 50 m scale, the abruptly decreasing change was represented from 180 m scale to 210 m scale. For SHDI and SHEI, obviously, the change rule was consistent, the stable status from 5 m scale to 50 m scale, but the fluctuation was formed from 50 m scale to 210 m scale. Undoubtedly, diversity, fragmentation and hierarchy of landscape located in Yucheng city were interpreted by the change of LSI, PAFRAC, SHAPE-MN and AI landscape index on 30 m scale based on RS image pixel. The detailed tendency of landscape index was shown from Figs. 1, 2, 3, 4, 5 and 6. Obviously, it is crucial for land-use landscape pattern and spatial process evolution in the case study [7–9].

Fig. 1.
figure 1

Sensitivity of PD on different scale

Full size image
Fig. 2.
figure 2

Sensitivity of LSI on different scale

Full size image
Fig. 3.
figure 3

Sensitivity of CONTAG on different scale

Full size image
Fig. 4.
figure 4

Sensitivity of PAFRAC on different scale

Full size image
Fig. 5.
figure 5

Sensitivity of SHEI on different scale

Full size image
Fig. 6.
figure 6

Sensitivity of SHDI on different scale

Full size image

At present, it is widely used that Landscape TM/ETM images were interpreted on country level, whose resolution is 30 m. So 30 m pixel scale was considered and used in the land-use types study in the case region, mainly including Arable-land, Grassland, Traffic and Transmission Land Use, Residential land, Public management and service land, Commercial service land, Garden plot land, Mine and storage land, Woodland, Water and water facility land, Special land and other land. Moreover, the change characteristics of LSI, PAFRAC, SHAPE-MN and AI was calculated by FRAGSTAT4 software for 12 landscapes, the results were referred from Table 4, Figs. 7, 8, 9 and 10.

Table 4. Index values of 12 landscapes
Full size table
Fig. 7.
figure 7

LSI change for 12 landscapes

Full size image
Fig. 8.
figure 8

PAFRAC change for 12 landscapes

Full size image
Fig. 9.
figure 9

SHAPE-MN for 12 landscapes

Full size image
Fig. 10.
figure 10

AI for 12 landscapes

Full size image

LSI was applied to analyze shape characteristics of landscape types, The author analyzed the Landscape Shape Index of 12 landscapes, Arable-land, Grassland, Traffic and Transmission Land Use, Residential land, Public management and service land, Commercial service land, Garden plot land, Mine and storage land, Woodland, Water and water facility land, Special land and other land. In terms of LSI, the drastically changed from the high value (70.814) to the small value (7.376). Water and water facility land had the high value is 71.814, which meant that Water and water facility land had the most complicate shape and was most influenced by various interventions, which showed the most complex diversity according to LSI ecological significance. Obviously, there are all kinds of Water and water facility land in the case region, including Rivers, water, lake water, water reservoir, ponds, coastal beach, inland beach, ditches, glaciers and permanent snow landform, which explained the complicated landscape types. From the shape matrix perspective, LSI value of Mine and storage land is less than 10, and indicate the landscape types is simple, Mine and storage land had the second most complicate shape, while Public management and service land LSI has the lowest value, which indicated that it had the simplest shapes and were influenced by human interventions [10].

Perimeter-area fractal dimension (PAFRAC) to some extent reflects the degree of human disturbance, and indicates the relationship between shape and area of landscapes consisting of patches, and the index at the landscape level is equal to the class level. For PAFRAC, Traffic and Transmission Land Use had the high value is 1.545, the value decreased step by step, PAFRAC is slight fluctuation centered on 1.2 of 11 landscapes, except 1.6 of Traffic and Transmission Land Use [10].

SHAPE-MN MN (Mean) equals the sum of the corresponding patch metric values, which is divided by the total number of patches. In terms of SHAPE_MN, Traffic and Transmission Land Use had the high value is 3.552, drastically change presented, the second high value is 1.675 of Public management and service land, the value decreased step by step for other landscapes, Arable-land, Residential land, Water and water facility land, other land, Grassland, Commercial service land, Woodland, Mine and storage land, Special land, Garden plot land.

Area index (AI) does not have a lot of interpretive value about evaluating landscape pattern, but it is important which defines the extent of the landscape. Moreover, many class and landscape metrics computations use total landscape area for. AI of Public management and service land, had the high value is 93.911, and AI of 12 landscapes to some extent presents the variability and gradient characteristics. Basically the kind of change can be divided into three levels in accordance with the area index value, the first gradient change is mainly involved in the three landscapes, Arable-land, Residential land, Public management and service land. The second, Grassland, Woodland, Water and water facility land, Commercial service land, Traffic and Transmission Land Use, other land. The third gradient change included the three landscapes, Garden plot land, Mine and storage land, Special land.

As the precious analysis, the landscape index (LSI, PAFRAC, SHAPE-MN, AI) results showed that they has a certain change and variability for each landscape, especially the landscape is in response to the scale, which indicated the scale effects. From the value of the index perspective, is not steep, or the ecological significance of landscape index indicated has the smooth change, AI has a certain hierarchical characteristics, PAFRAC has a small change, LSI and SHAPE-MN reflected the real landscape, in accordance with own characteristics, and there is no obvious stratification change.

4 Conclusions

In a word, the main contents and conclusions are as follows:

We also definitely understand and investigate the question how changing scale, such as grain size affects pattern analysis. The ecological significance of CONTAG, PD, LSI, PAFRAC, SHDI, SHEI indicated that had a certain scale effects of 18 scales in Yucheng city, though the different degree and different curve. CONTAG changed slightly, which indicated the stable landscape equilibrium. PD has increased from 5 m scale to 30 m scale, but decreased from 30 m scale to 210 m scale, so fragmentation of regional landscapes had presented fluctuation. In terms of LSI and PAFRAC increased from small scale to big scale. SHDI and SHEI changed slightly and showed the balanced landscape diversity, landscape types were evenly distributed, landscape fragmentation and heterogeneity changed slightly [11–13].

We may detect or identify characteristics scales and hierarchical levels to understand and predict ecological phenomena. Based on image by the change of LSI, PAFRAC, SHAPE-MN and AI on 30 m pixel scale, we also emphatically analyzed the LSI, PAFRAC, SHAPE-MN, AI of 12 landscapes. Further, we explored and explained the ecological significance of different landscapes in the case city.

Applying the principles of the landscape ecology, the paper analyzes the landscape diversity and sensitivity of 12 landscapes, points out that landscape sensitivity is the landscape systems response to disturbance at different spatial scales, and reveals that landscape spatial pattern and ecological processes of interaction couple of natural factor and Human disturbance.