**1. Introduction**

The Yellow River is the second largest river in China and the mother river of the Chinese nation. Early human beings made full use of the water and soil resources of the Yellow River. Early Chinese civilization was nurtured and developed in this area. The main stream of the Yellow River is 5464 kilometers long, with a large drop and a large amount of sand. There has been flooding since ancient times, and the Yellow River has also changed its river course many times, which exerted a huge impact on the life of early humans in the basin. In this process, people used it in an appropriate way, thus forming a series of measures to control the Yellow River and social organization. Today, the Yellow River is an important ecological security barrier in northern China and an important area for population activities and economic development. The study of the early Yellow River Basin environment is of great significance to current ecological protection [1], heritage protection [2], and human–water relations [3,4].

**Citation:** Li, J.; Song, Y.; Zhang, W.; Zhu, J. Analysis of the Aggregation Characteristics of Early Settlements in the Zhengzhou Ancient Yellow River Distributary Area. *Water* **2022**, *14*, 2961. https://doi.org/ 10.3390/w14192961

Academic Editor: Juan José Durán

Received: 30 July 2022 Accepted: 17 September 2022 Published: 21 September 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Zhengzhou (Henan Province, China), is located in the middle and lower reaches of the Yellow River, which is the core area of the origin and development of early Chinese civilization. The Yellow River and its ancient distributary have been the main driving forces for the formation of rivers, lakes, and landforms in this region since the Late Pleistocene. Studies by relevant scholars suggest that about 40,000 to 10,000 years ago, the flood of the Yellow River formed two ancient distributaries of the Yellow River in the east and west of Guangwu Mountain in Zhengzhou. The west ancient Yellow River distributary descended from the northwestern part of today's Xingyang, passed through the "western suburbs" of Zhengzhou, rushing to Zhongmou and Weishi, and entering the Ying River distributary. About 10,000 BP, it disappeared with the overall rise of the western part of Zhengzhou and the continuous subsidence of the northeast. The east ancient Yellow River distributary went southeast along the Bian River and Ying River and entered the Huai River Basin. Until the end of the historical period (100 years ago), the flooding of the Yellow River could still go southeast along the distributary. The Yellow River distributary not only contributed to the formation of the Xingyang–Guangwu trough south of Guangwu Mountain, but also with the disappearance of the west ancient Yellow River distributary about 10,000 BP, three lakes and swamps developed along the east–west distributaries of Guangwu Mountain. The Xingyang–Guangwu trough lake and swamp in the west flourished between 7000 and 4000 BP and disappeared at the end of the Longshan period; the Eastern Putian lake and swamp continued to evolve between 9000 and 2000 BP. The Xingze lake in the northeast developed around 9000 years ago and disappeared 2100 years ago. At the same time, the ancient Yellow River distributary and the ancient lakes and swamps affected the development of early sites in Zhengzhou. In this area, the development of ancient culture is coherent, and the Paleolithic, Neolithic, and Xia–Shang Dynasties' early sites are abundant [5–7].

At present, from the perspective of research objects, many scholars have studied the Central Plains area [8,9] (Henan Province) where Zhengzhou is located, the surrounding Songshan area [10,11] (Zhengzhou, Luoyang, Xuchang, Pingdingshan, and its surrounding areas), the Zheng–luo area [12,13] (Zhengzhou City, Luoyang City, and surrounding areas), etc. They mainly focused on the relationship between early sites and the environment at a macro level.

From the perspective of study direction and methods, the majority of research focused on the development of site culture and its influencing factors. Research involved many fields such as history, archaeology, geology, water conservancy, and so on. In the fields of archaeology, the cultural characteristics of the early sites were analyzed qualitatively, and the site pattern was also analyzed to explore the ancient social organization structure and the relationships between people. Song, A. [14], Zhang, H. [15], Zhao, C. [16], and other scholars, on the basis of culture stage of archaeology, focused on the qualitative analysis of the temporal and spatial evolution of the site from the distribution, form, and layout of specific sites. Scholars from the field of Geography mostly analyzed the driving factors of site distribution based on the spatial location of sites. Yan, L. [17], Lu, P. [18], Bi, B. [19], and other scholars provided methods and tools for site groups, settlement hierarchies, and settlement center transfer patterns in the early research on the relationship between the spatial distribution of sites and the natural environment. At the same time, Yan, L. [20] further used the adjacent index analysis method, standard deviation ellipse, kernel density, and other GIS spatial statistical methods. Furthermore, the study adopted the spatial point model to determine that the degree of aggregation in the Songshan Mountains area sites gradually increased, and the temporal and special pattern shifted from dispersion to aggregation in different periods. At present, scholars [5,21] have qualitatively determined that the existence of the Yellow River distributary in the late Pleistocene in Zhengzhou contributed to the formation of the trough in Xingyang–Guangwu, and this area also became an important area for early human settlements. However, these studies were basically restricted to the spatial distribution patterns of sites at the macro level, and there is still a lack of more in-depth quantitative research on the aggregation characteristics of

early sites in the ancient Yellow River distributary from a multi-dimensional perspective at the intermediate scale. gregation characteristics of early sites in the ancient Yellow River distributary from a multi-dimensional perspective at the intermediate scale. This research focuses on the area of the Zhengzhou ancient Yellow River distributary,

late Pleistocene in Zhengzhou contributed to the formation of the trough in Xingyang– Guangwu, and this area also became an important area for early human settlements. However, these studies were basically restricted to the spatial distribution patterns of sites at the macro level, and there is still a lack of more in-depth quantitative research on the ag-

*Water* **2022**, *14*, 2961 3 of 15

This research focuses on the area of the Zhengzhou ancient Yellow River distributary, which involves the central urban area of Zhengzhou and the four counties of Xingyang, Xinmi, Xinzheng, and Zhongmou (Figure 1), and it uses the method of GIS quantitative analysis to identify the typical areas where early sites were gathered. From the mesoscopic level, the study analyzes the influence of the evolution of lakes and swamps formed after the disappearance of the ancient Yellow River distributary on the aggregation of sites in different periods and provides support for explaining the relationship between early settlement structure and the ancient Yellow River environment. At the same time, the study provides new ideas for archaeological research work in the dense early site area and the systematic protection and utilization of cultural heritage. which involves the central urban area of Zhengzhou and the four counties of Xingyang, Xinmi, Xinzheng, and Zhongmou (Figure 1), and it uses the method of GIS quantitative analysis to identify the typical areas where early sites were gathered. From the mesoscopic level, the study analyzes the influence of the evolution of lakes and swamps formed after the disappearance of the ancient Yellow River distributary on the aggregation of sites in different periods and provides support for explaining the relationship between early settlement structure and the ancient Yellow River environment. At the same time, the study provides new ideas for archaeological research work in the dense early site area and the systematic protection and utilization of cultural heritage.

**Figure 1.** Research scope and location map. **Figure 1.** Research scope and location map.

#### **2. Data and Methods 2. Data and Methods**

#### *2.1. Data*

*2.1. Data*  The data of sites in this study were obtained from the third National Survey of Cultural Relics, including a total of 502 sites of Paleolithic, Neolithic, Xia Dynasty, and Shang Dynasty; its attributes include each site's location, area, period, grade, etc. In addition, the *Atlas of Chinese Cultural Relics: Henan Volume* [22], and the existing sites' research results at The data of sites in this study were obtained from the third National Survey of Cultural Relics, including a total of 502 sites of Paleolithic, Neolithic, Xia Dynasty, and Shang Dynasty; its attributes include each site's location, area, period, grade, etc. In addition, the *Atlas of Chinese Cultural Relics: Henan Volume* [22], and the existing sites' research results at this stage were also references for the study [23].

this stage were also references for the study [23]. The research objects were the historical water environments of the Zhengzhou ancient Yellow River distributary, lakes, and swamps; the Xingyang–Guangwu trough and other historical water environments; and the spatial aggregation patterns of early settlements in Zhengzhou from the Neolithic period (Peiligang period–Yangshao period– Longshan period). The characteristic information of the ancient Yellow River distributary, lakes, and swamps, and ancient geomorphic environment data were obtained from the research of relevant scholars on the historical geography of Zhengzhou and changes of the Yellow River course, and the vectorization operation was carried out through a GIS platform (Figure 2). The research objects were the historical water environments of the Zhengzhou ancient Yellow River distributary, lakes, and swamps; the Xingyang–Guangwu trough and other historical water environments; and the spatial aggregation patterns of early settlements in Zhengzhou from the Neolithic period (Peiligang period–Yangshao period–Longshan period). The characteristic information of the ancient Yellow River distributary, lakes, and swamps, and ancient geomorphic environment data were obtained from the research of relevant scholars on the historical geography of Zhengzhou and changes of the Yellow River course, and the vectorization operation was carried out through a GIS platform (Figure 2).

**Figure 2.** Location map of sites and ancient lakes and swamps in different periods. **Figure 2.** Location map of sites and ancient lakes and swamps in different periods.

The Xingyang–Guangwu trough lake and swamp in the west of Zhengzhou [7]:


the Xingze lake was left. The Putian lake and swamp in the east of Zhengzhou and in the west of Xingze lake The Putian lake and swamp in the east of Zhengzhou and in the west of Xingze lake [24]:

[24]: The Eastern Putian lake and swamp continued to evolve between 9000 and 2000 BP. The Xingze lake in the northeast developed around 9000 years ago and disappeared 2100 years ago. According to related studies, the boundaries of different periods are uncertain, The Eastern Putian lake and swamp continued to evolve between 9000 and 2000 BP. The Xingze lake in the northeast developed around 9000 years ago and disappeared 2100 years ago. According to related studies, the boundaries of different periods are uncertain, and thus a rough boundary has been plotted.

#### and thus a rough boundary has been plotted. *2.2. Methods*

[27,28] is:

*2.2. Methods*  2.2.1. Analysis Method of Research on the Distribution of Settlement Density

2.2.1. Analysis Method of Research on the Distribution of Settlement Density In this study, kernel density analysis was used to analyze the spatial distribution and aggregation characteristics of settlement sites in different periods and as the basis for di-In this study, kernel density analysis was used to analyze the spatial distribution and aggregation characteristics of settlement sites in different periods and as the basis for dividing the settlement cluster.

viding the settlement cluster. Sites with close social organization and production relationship are usually close in space. By analyzing the distribution density of sites, it is helpful to observe the social organization relationship between sites with resources and their environments [25]. American scholar Drennan was the first to draw a topographic map of site distribution in GIS Sites with close social organization and production relationship are usually close in space. By analyzing the distribution density of sites, it is helpful to observe the social organization relationship between sites with resources and their environments [25]. American scholar Drennan was the first to draw a topographic map of site distribution in GIS for regional sites analysis [26]. Kernel density analysis is similar. The calculation formula [27,28] is:

$$f\_n\left(\mathbf{x}\right) = \frac{1}{nh} \sum\_{i=1}^{n} \mathcal{K}\left(\frac{\mathbf{x} - \mathbf{x}\_i}{h}\right) \tag{1}$$

In the formula, *f* is the kernel density; *k* () is called the kernel function; *h* is the search radius (broadband), where *h* > 0; n is the number of known points in the broadband range, that is, the number of research samples; and (*x*–*x<sup>i</sup>* ) represents the distance from the estimated point *x* to the sample point *xi*. The larger the value of *f*(*x*), the denser the distribution of points.

#### 2.2.2. Analysis Method of the Complexity of the Settlement Cluster Structure

In this study, the structural complexity of the settlement cluster was judged by dividing the cluster and analyzing the situation of different levels of sites within each cluster.

The settlement clusters were divided based on the results of the kernel density analysis using the GIS Contour tool. The "Contour" tool is often used to demarcate the boundaries of geographic feature clusters [29]. Therefore, this research used the contour tool to express the settlement density topographic map in the form of contour lines. In this way, the "kernel density peaks" surrounded by contour lines with different values could represent clusters of different scales, and combined with the actual topographic map, the settlement clusters could be divided.

The structural complexity analysis of settlement clusters was further combined with the research on settlement complexity in archaeology. Relevant scholars have pointed out that since the middle and late Yangshao period in Zhengzhou, there has been an obvious hierarchy within the settlement group [30]. At the same time, because prehistoric ancestors were affected by traffic conditions and productivity levels, their social activities usually took place in a certain range of sites. By analyzing the hierarchical complexity within the settlement group, it is possible to judge the advantages and disadvantages of the environment of the settlement group. In the absence of historical records, this research adopted spatial cluster division and used site clusters instead of settlement groups for complexity analysis.

#### 2.2.3. Analysis Method of the Evolution of Settlement Aggregation

Based on the typical settlement aggregation areas identified by the above kernel density analysis, it was further divided using Thiessen polygons. According to the coefficient of variation (CV) of polygons in different periods, the aggregation degree and spatial distance of settlement sites were compared, and the aggregation degree and continuity characteristics of early settlement clusters were judged.

Thiessen Polygonal Analysis can clearly show the spatial distribution pattern of a set of points on the plane, which is judged by the coefficient of variation (CV). The coefficient of variation (CV) refers to the ratio of the standard deviation of the area of each polygon to the mean. When the points are evenly distributed, the polygon areas are similar, and the CV value is relatively small, but when the points are clustered, the polygon area difference is large, and the CV value is large [31].

#### 2.2.4. Analysis Method of Settlement Cluster Structure

Studying the distance differences between settlements of different levels and surrounding lakes and swamps can help us identify the spatial structure of settlement clusters under specific geographical factors. This distance difference can be analyzed by the ArcGIS buffer analysis tool. Taking lakes and swamps as the main bodies, multi-ring buffer zones with different radii were established and then spatially connected with settlement sites. Finally, the number of sites in buffer zones with different radii was counted, through the comparison of the number of sites, to find the spatial pattern.

#### **3. Results 3. Results**

*3.1. Identification of Typical Aggregation Areas of Early Sites in the Yellow River Distributary in Zhengzhou 3.1. Identification of Typical Aggregation Areas of Early Sites in the Yellow River Distributary in Zhengzhou* 

3.1.1. Characteristics of Sites Density Distribution 3.1.1. Characteristics of Sites Density Distribution

According to the relevant research and experimental analysis of the search radius and based on the overall scale of the study area, 4 km was selected as the search radius, and the kernel density analysis of the site data was carried out and visualized (Figure 3). The overall features are as follows: According to the relevant research and experimental analysis of the search radius and based on the overall scale of the study area, 4 km was selected as the search radius, and the kernel density analysis of the site data was carried out and visualized (Figure 3). The overall features are as follows:


**Figure 3.** Overall kernel density analysis map of the study area. **Figure 3.** Overall kernel density analysis map of the study area.

To sum up, the sites were mainly located along the area where the west ancient Yellow River distributary once flowed, forming a ribbon-like aggregation feature composed of multiple centers. The aggregation centers were the most concentrated in the middle reaches of the Suo-xu-ku River within the Xingyang–Guangwu trough and in the upper and middle reaches of the Jialu River. To sum up, the sites were mainly located along the area where the west ancient Yellow River distributary once flowed, forming a ribbon-like aggregation feature composed of multiple centers. The aggregation centers were the most concentrated in the middle reaches of the Suo-xu-ku River within the Xingyang–Guangwu trough and in the upper and middle reaches of the Jialu River.

#### 3.1.2. Complexity Analysis of Settlement Clusters

First, the research used "topographic map" of settlement density to divide the settlement clusters in each period, then extracted the contour lines with a kernel density value greater than 0.2 (this value was selected to show the settlement distribution pattern) and divided the clusters according to the actual topography. Second, the research analyzed the complexity of the settlement cluster structure in each period. The degree of structural complexity of a settlement cluster with two or more central settlements is high, the degree

of structural complexity of a settlement cluster with only one central settlement is average, and the degree of structural complexity of a settlement cluster without a central settlement is low. The determination of the central settlement was based on the existing archaeological classification of the settlement levels in the Zhengzhou area [15,16] (Figure 4), which mainly presented the following characteristics: of structural complexity of a settlement cluster with only one central settlement is average, and the degree of structural complexity of a settlement cluster without a central settlement is low. The determination of the central settlement was based on the existing archaeological classification of the settlement levels in the Zhengzhou area [15,16] (Figure 4), which mainly presented the following characteristics:

First, the research used "topographic map" of settlement density to divide the settlement clusters in each period, then extracted the contour lines with a kernel density value greater than 0.2 (this value was selected to show the settlement distribution pattern) and divided the clusters according to the actual topography. Second, the research analyzed the complexity of the settlement cluster structure in each period. The degree of structural complexity of a settlement cluster with two or more central settlements is high, the degree

*Water* **2022**, *14*, 2961 7 of 15

3.1.2. Complexity Analysis of Settlement Clusters


**Figure 4.** Complexity analysis of settlement clusters in the study area. **Figure 4.** Complexity analysis of settlement clusters in the study area.

To sum up, the early settlement clusters in the Xingyang–Guangwu trough area had a large number of clusters and high complexity. To sum up, the early settlement clusters in the Xingyang–Guangwu trough area had a large number of clusters and high complexity.

#### 3.1.3. Determination of Typical Settlement Aggregation Areas 3.1.3. Determination of Typical Settlement Aggregation Areas

By analyzing the overall density distribution pattern of the settlements and comparing the complexity of settlement clusters, it was found that the aggregation center in the middle reaches of the Suoxuku River in the Xingyang–Guangwu trough had a high aggregation degree, and more complex settlement clusters were formed in the Xingyang– Guangwu trough. At the same time, because the Xingyang–Guangwu trough was the main flow area of the west ancient Yellow River distributary, this area was selected as a typical settlement aggregation area. The relationship between the morphology of settlements in the region and the ancient lakes and swamps was further analyzed. By analyzing the overall density distribution pattern of the settlements and comparing the complexity of settlement clusters, it was found that the aggregation center in the middle reaches of the Suoxuku River in the Xingyang–Guangwu trough had a high aggregation degree, and more complex settlement clusters were formed in the Xingyang–Guangwu trough. At the same time, because the Xingyang–Guangwu trough was the main flow area of the west ancient Yellow River distributary, this area was selected as a typical settlement aggregation area. The relationship between the morphology of settlements in the region and the ancient lakes and swamps was further analyzed.
