Public Aesthetic and Recreational Preferences for Historical Irrigation Canals in Rural Areas: A Case Study in Li Canal-Gaoyou Irrigation District

Abstract

As an integral part of rural built heritage, the aesthetic and recreational values of Historical Irrigation Canals (HICs) have increasingly garnered attention in recent years, providing new perspectives for heritage conservation and sustainable rural development. However, there is a scarcity of research conducted from the micro-perspective of public preferences for HICs in rural areas. The existing literature primarily focuses on how the landscape characteristics of natural and urban rivers affect public preferences, with a lack of attention to built heritage characteristics and rural settings, which are uniquely inherent to HICs in rural areas. This study aims to explore public preferences for HICs in rural areas from the perspectives of landscape characteristics and social backgrounds. The Li Canal-Gaoyou Irrigation District, one of the World Heritage Irrigation Structures, was selected as the case study. Selected HIC photos were utilized for measuring landscape characteristics and conducting survey questionnaires. Relationships among HIC landscape characteristics, social backgrounds, and aesthetic and recreational preferences were revealed through one-way ANOVA, Spearman’s correlation, and stepwise regression analyses. Key findings include the following: (1) the public can derive aesthetic and recreational value from HICs; (2) canals with large water bodies, artificial banks, and artificial vegetation are considered more aesthetically pleasing; (3) larger canals, extensive water bodies, and artificial banks are deemed more suitable for recreation; (4) factors such as age, education level, income, rural living experience and knowledge of HICs influence public preferences; (5) costs influence public aesthetic and recreational preferences, with people more willing to pay for roads and hardened banks. Finally, several recommendations for the protection and management of HICs have been summarized, offering guidance for rural managers and planners.

1. Introduction

Historical Irrigation Canals (HICs) are considered a type of built heritage [1,2,3], and their sustainable utilization has garnered significant international attention in recent years. Since 2014, World Heritage Irrigation Structures (WHIS) have been selected to promote the protection of historical irrigation structures [4], which have shaped significant built environments in the history of human settlement, and most still remain functional. HICs are the primary backbone of historical irrigation structures [5], which are extensively distributed across rural areas through a hierarchical structure, thereby creating distinct local characteristics [6]. Given that historical irrigation districts occupy vast tracts of rural land, the protection and utilization of HICs have tremendous potential for rural revitalization. For instance, in China, over 400 historical irrigation districts are still in use [7], with 34 recognized as WHIS, covering more than 1.82 million hectares [8]. In 2021, the Chinese government explicitly highlighted the need to coordinate rural development with the conservation and utilization of historical irrigation structures [9]. However, during rapid urban and rural development, irrigation canals are highly susceptible to damage or alteration due to their fragile built structures and singular functional value, and are often overlooked in heritage preservation efforts [10]. At the same time, internationally, multiple cases aimed at protecting and revitalizing HICs are emerging [11,12,13,14,15]. Exploring a sustainable approach to heritage protection of HICs is an urgent need today.

In recent years, the additional benefits of HICs beyond irrigation have garnered attention [16,17,18,19,20], particularly their aesthetic and recreational potential [21,22]. These additional values can enhance the attractiveness of rural areas to non-local populations, diversify the industrial structure, and can further enhance support for heritage preservation and regional sustainable development [23,24,25], particularly for rural heritage that lacks sufficient human and financial resources for protection. Therefore, exploring public aesthetic and recreational preferences for HICs can provide effective support for sustainable heritage protection [26]. In order to maximize aesthetic and recreational services, special attention should be given to public preference [27]. Current research on irrigation canals primarily focuses on macro-perspectives, such as the impact of canals on regional morphology and structure [28,29,30]. Studies combining micro-perspectives are less common, lacking a bottom-up approach to explore public preferences regarding irrigation canals. Public preferences have been proven as valuable in land use decision-making processes by determining the value users assign to land and by formulating appropriate land policies based on users’ preferences [31,32]. Some studies have focused on public preferences for rural landscapes, primarily concerning road landscapes [33], ponds [34], farmlands [35,36], abandoned farmlands [37], rangelands [38], and rural structures like agricultural greenhouses [39] and biogas plants [40]. However, to the authors’ knowledge, no literature has yet explored the public aesthetic and recreational preferences of HICs.

Landscape characteristics are often used to represent elements and patterns of the built environment, thereby quantitatively supporting decision-making and planning design [41]. The existing literature predominantly suggests that public aesthetic and recreational preferences have been demonstrated to be influenced by landscape characteristics [42]. For example, natural elements such as vegetation and water positively affect aesthetic quality, while roads and buildings generally have a negative impact [43]. The presence of gravel bars negatively correlates with recreational preferences [44]. Beyond the landscape features themselves, urban–rural differences also affect public perceptions. For instance, semi-natural rivers are preferred in urban settings [45], whereas natural rivers are more favored in rural areas [46]. Highly artificial elements, such as hedges, might be more favored in rural settings [47]. However, these studies primarily focus on urban and natural rivers, whose landscape characteristics significantly differ from those of HICs in rural areas. Moreover, previous research also often overlooks the historical and local significance of these linear water bodies [48,49]. Yet, it has been found that the aesthetic and perceptual experiences of built heritage structures can be influenced or even diminished by the natural environment, thereby burying their heritage value [50]. Therefore, the conclusions drawn from existing studies may not be applicable to HICs in rural contexts, highlighting a particular need for focused attention and exploration. Furthermore, the cost factors of different intervention strategies influence both public aesthetic and recreational preferences [51,52], which should be considered in research.

This research could yield insights for the protection and sustainable development of HICs in rural areas. Based on this, the main research objectives explored in this paper are outlined as follows:

(1)

Public perceptions of the aesthetic values of HICs in rural areas, as well as preferences for different types of recreational activities.

(2)

The relationship between landscape characteristics and aesthetic preferences for HICs in rural areas.

(3)

The relationship between landscape characteristics and recreational preferences for HICs in rural areas.

(4)

The relationship between social background and the public’s aesthetic and recreational preferences for HICs in rural areas.

(5)

The relationship between cost of various intervention strategies and the public’s aesthetic and recreational preferences for HICs in rural areas.

2. Materials and Methods

2.1. Research Framework

The research framework is illustrated in Figure 1. Initially, 18 representative photographs of HICs were collected from the study area, and a survey questionnaire tailored to these photographs was developed. The questionnaire was distributed online, collecting respondents’ demographic information and their aesthetic and recreational perceptions of the photographs. Subsequently, the photographs were manually measured to identify landscape characteristics, with each photograph being interpreted into 13 factors. Finally, statistical methods were employed to analyze the correlations between different datasets, aiming to explore the objectives of the study. Descriptive analysis was utilized to explore the differences in public aesthetic and recreational preferences across respective sub-factors (Objective 1). The link between landscape characteristics and both aesthetic (Objective 2) and recreational preferences (Objective 3) was evaluated using Spearman’s correlation and stepwise regression analysis. The association between social background and both aesthetic and recreational preferences (Objective 4) was assessed using one-way ANOVA. Building on this foundation, we added a questionnaire survey to further explore the relationship between key landscape characteristics and public preferences from the perspective of cost factors (Objective 5).

2.2. Sample Sites

The study selected the Li Canal-Gaoyou Irrigation District in China as the sample site. Listed in WHIS in 2021 [53], it represents a significant example of historical irrigation structures within China’s water network plain area. This historical irrigation structure consists of Gaoyou Lake, Li Canal, and the Gaoyou Irrigation Area. The HICs, located within the Gaoyou Irrigation Area, are supplied with water from Gaoyou Lake and Li Canal. The structure is made up of a multi-tiered irrigation canal system (Figure 2), comprising six main canals, 127 branch canals, and 3250 tertiary canals, which continue to function for irrigation. Canals at different scales exhibit variations in width and surrounding environments, potentially offering diverse aesthetic and recreational possibilities.

2.3. Photograph Collection, Selection, and Preprocessing

In this study, full-color photographs were used as substitutes for real scenery, a method employed in aesthetic and recreational preferences studies [44,48,54]. All photographs were taken in March 2024 using the same camera, under clear weather conditions, between 9 a.m. and 4 p.m., at a consistent shooting height of 1.6 m. In the photographs, irrigation canals were centrally positioned, and the horizon line was uniformly placed to minimize perceptual biases caused by framing. The photographs were taken along the canals throughout the entire rural area, striving to cover the full range of the irrigation area as depicted in Figure 2. They feature various canal hierarchies and an array of landscape characteristics, including vegetation, farmland, roads, hydraulic facilities, and buildings. After excluding duplicate and low-quality photographs, a total of 394 photographs were captured.

Ten landscape architecture experts were invited to select representative photographs from the collection to be used in questionnaires and for measuring landscape characteristics, as shown in Figure 3. During the selection process, the experts adhered to three main principles and determined the final set of 18 photographs through discussions as follows: (1) To better explore the characteristics of the multi-tiered historical irrigation structure, the 18 photographs cover canals at various scales. In Figure 3, the photographs are arranged according to their “main–branch–tertiary” scale differences. (2) To include as many types of element combinations as possible, the selected photographs show items such as vegetation, trees, farmland, bare land, hydraulic facilities, roads, and buildings, while maintaining a variety of naturalness in the bank and vegetation. (3) To ensure that the selected photographs are representative, scenes that are too similar or have excessive interfering elements (such as numerous vehicles, waste, or messy environments) were avoided. After selecting the 18 photographs, elements such as trash, irrelevant damaged facilities (such as broken road signs and billboards), and temporary construction equipment (such as excavators in the background) were digitally removed using Photoshop CC 2018. This step was taken to prevent any influence on respondents’ perceptions and to ensure the reliability of the results.

2.4. Questionnaire Survey

The survey was divided into online and offline categories. The online surveys were distributed via the internet to people outside the Li Canal-Gaoyou Irrigation District and surrounding counties and cities, who constituted the main subjects of the study. The offline surveys were distributed locally by the research team using paper questionnaires or digital media. Online completed questionnaires also collected IP information, which was used along with additional questions to verify whether the respondents were local residents. At the beginning of the questionnaire, respondents were presented with a brief introduction to the Li Canal-Gaoyou Irrigation District, followed by a simultaneous display of all photographs to provide an overview. Subsequently, the 18 photographs appeared randomly across the questionnaire pages, and respondents were asked to complete the same set of questions after viewing each photo.

The survey investigated the public’s aesthetic and recreational preferences separately. Based on a literature review, aesthetic preference was categorized into three sub-factors—aesthetic quality [55,56], historicity [28,57,58], and locality [57,59,60]—each associated with 1–5 Likert scale-designed questionnaire items (

Table 1. Questionnaire design for investigating aesthetic preference and recreational preference.

Dimension	Factors	Survey Questions
aesthetic preference	aesthetic quality	Do you think this scene is beautiful or not? Very unbeautiful = 1; Unbeautiful = 2; Average = 3; Beautiful = 4; Very beautiful = 5
	historicity	Do you think this scene evokes a sense of history and heritage? Not at all = 1; Barely = 2; Average = 3; Somewhat = 4; Strongly = 5
	locality	Does this scene appear unique to you, reflecting distinct local characteristics? Not at all = 1; Barely = 2; Average = 3; Somewhat = 4; Strongly = 5
recreational preference	walking	Which of the following recreational activities do you think are suitable to be carried out or developed in this scene? (Multiple choices allowed) walking, jogging, biking, swimming, boating, fishing, no suitable activities
	jogging
	biking
	swimming
	boating
	fishing

). For recreational preference, respondents were asked to select from six types of recreational activities based on the literature review [21,61]. If an activity was selected, it was recorded as ‘1’; otherwise, it was recorded as ‘0’. The average values of the sub-factors for both aesthetic and recreational preferences were calculated to obtain overall scores for the two dimensions.

In addition to the above, the questionnaire inquired about the respondents’ social backgrounds, encompassing basic demographic characteristics such as gender, age, education level, and income. After considering existing literature that highlights the influence of rural experience [62] and cognitive differences [57], we added two additional questions: (1) Have you ever lived in a rural area? Respondents were asked to indicate whether they had ever lived in a rural area and whether the duration of their stay exceeded two years. (2) Before taking this survey, were you aware of historical irrigation structures or WHIS? Respondents were required to choose from four options as follows: ‘Very aware’, ‘Somewhat aware’, ‘Heard of but not aware’, ‘Completely unheard of’.

2.5. Measurement of Landscape Characteristics

Based on the relevant literature [33,44,48], we selected 13 factors to quantify the landscape characteristics of various photographs (

Table 2. Landscape characteristic factors and measurement methodology.

Type	Factors	Measurement Methodology
natural elements	percentage of water	calculated by proportion
	percentage of vegetation
	percentage of trees
	percentage of farmland
	percentage of bare land
built elements	percentage of hardscape
	percentage of canal
	percentage of hydraulic facilities *
	percentage of road
	percentage of buildings
	percentage of other facilities *
naturalness	naturalness of bank	entirely artificial revetment with no greenery = 0 partially greened artificial revetment = 1 fully greened revetment = 2 fully greened and naturally shaped revetment = 3
	naturalness of vegetation	artificial vegetation = 1 semi-artificial and semi-natural vegetation = 2 entirely natural vegetation = 3

* The hydraulic facilities include sluice stations and bridges, while other facilities encompass utility poles, signs, and other visible built infrastructure.

). These factors integrate commonly used indicators in landscape characteristic evaluations [44,48,63], designed to measure the proportion of various natural elements (water, vegetation, trees, farmland, bare land) and built elements (hard landscapes, canals, hydraulic facilities, roads, buildings, and other facilities). Additionally, two factors—naturalness of bank and vegetation—were added to assess landscape configuration patterns, which are key considerations in evaluating waterfront landscape characteristics [46,64,65], as well as in heritage and rural landscape studies [38,50]. Graduate students majoring in landscape architecture manually delineated the boundaries of various elements in AutoCAD, subsequently calculating the corresponding areas and their proportions in the photographs. The quantification of naturalness was achieved through assessments conducted by ten landscape architecture experts for each photograph.

2.6. Investigation of Intervention Strategies with Varying Costs

Based on the stepwise regression analysis results between landscape characteristics and aesthetic and recreational preferences, we identified four main landscape characteristics affecting public preferences: percentage of water, naturalness of bank, naturalness of vegetation, and percentage of road. Since the proportion of water denotes the grade and width of the HICs, these are usually not altered for aesthetic and recreational purposes. Therefore, the survey questionnaire established three sets of scenarios based on the three HIC grades: main, branch, and tertiary. Each set consists of eight different intervention scenarios formed based on the other three landscape characteristics (Figure 4). These intervention strategies were coded (0 or 1) for subsequent statistical analysis, with banks categorized as natural (0) or hardened (1), vegetation as natural (0) or regular (1), and roads as absent (0) or present (1). These varying scenarios were visually represented through photomontage method [49,52] in Photoshop cc2018 created on actual photographs to simulate the different interventions.

Through discussions with three local experts, we ascertained the actual construction costs of HICs. Based on this, we determined the relative costs required for each intervention strategy, setting the lowest at 0 yuan (RMB) and the highest at 100 yuan (RMB), which is often used in related research [66]. The meaning of relative costs was explained in the survey text, and the relative costs of different intervention strategies were prominently marked on the images of each scenario.

The survey was distributed to participants via the Internet. In the questionnaire, respondents are first briefly introduced to the situation of the Li Canal-Gaoyou Irrigation District, as well as the concepts of aesthetics, recreation, and costs. Subsequently, participants view images of three sets of scenarios and are asked to complete the same two questions after viewing each set of photos: (1) Considering both cost and aesthetics, select and rank the best three scenarios. (2) Considering both cost and recreation, select and rank the best three scenarios. Based on the ranking, the first place was recorded as ‘3’, second was recorded as ‘2’, third was recorded as ‘1’, and otherwise, ‘0’.

2.7. Statistical Analysis

Descriptive analysis was employed to compare the differences in aesthetic preferences and recreational preferences among their respective sub-factors. Subsequently, Spearman’s correlation was employed to assess the relationship between landscape characteristics and both aesthetic and recreational preferences, providing an initial observation of which landscape characteristics are associated with public preferences. We further explored how landscape characteristics relate to aesthetic and recreational preferences through stepwise regression analysis. Thirdly, to investigate the impact of social backgrounds on public preference, we employed one-way ANOVA to compare the differences in average aesthetic and recreational preferences across 18 photographs among groups from various social backgrounds. Fourth, the relationship between the cost of various intervention strategies and aesthetic and recreational preferences was assessed using Spearman’s correlation.

3. Results

3.1. Descriptive Analysis of Aesthetic and Recreational Preferences

The study collected a total of 498 questionnaires, of which 33 were invalid due to either excessively short response times or inconsistent selections within the questionnaire (such as selecting “no suitable activities” while also choosing other types of activities). A total of 465 valid questionnaires were collected, including 373 online questionnaires from non-local respondents and 92 offline questionnaires from local residents, of whom 66 (70.97%) had experience using the irrigation canals. Cronbach’s alpha was used to assess the reliability of the data. The reliability of the other items ranged from 0.816 to 0.949, suggesting high reliability of the research data (>0.8). Overall, the data demonstrate high reliability and are suitable for further analysis.

Comparing the aesthetic preference evaluations of 18 photographs (Figure 5), the 16th photograph scored the highest (average score = 4.092), followed by the 14th (average score = 3.948) and the 13th (average score = 3.912). The 11th photograph scored the lowest (average score = 3.112), closely followed by the 18th (average score = 3.232) and the 4th (average score = 3.275). However, overall, respondents consider these landscapes to possess aesthetic value (score > 3.0).

Examining the average performance of the three aesthetic dimensions across all 18 photographs, aesthetic quality scored the highest (average score = 3.781), followed by locality (average score =3.563), with historicity scoring the lowest (average score = 3.467).

Comparing the recreational preference of 18 photographs (Figure 6), the 16th photograph scored the highest (average score = 0.574), followed by the 14th (average score = 0.567) and the 13th (average score = 0.558). The 2nd photograph scored the lowest (average score = 0.236), closely followed by the 1st (average score = 0.243) and the 11th (average score = 0.258).

Examining the average preferences for six recreational activities across all 18 photographs, walking (average score = 0.725) emerged as the most preferred activity, followed by jogging (average score = 0.477). The least preferred recreational activity was swimming (average score = 0.129).

3.2. Relationship between Landscape Characteristics and Both Aesthetic and Recreational Preference

Table 3. Result of Spearman correlation analysis between landscape characteristics and aesthetic preference and recreational preference.

	Aesthetic Preference	Aesthetic Quality	Historicity	Locality	Recreational Preference	Walking	Jogging	Biking	Swimming	Fishing	Boating
percentage of water	0.399	0.152	0.544 *	0.438	0.626 **	−0.24	0.11	0.076	0.890 **	0.913 **	0.902 **
percentage of vegetation	−0.517 *	−0.387	−0.467	−0.513 *	−0.439	−0.186	−0.399	−0.353	−0.433	−0.307	−0.407
percentage of trees	−0.034	−0.205	0.096	0.001	0.174	−0.015	0.013	−0.039	0.169	0.255	0.178
percentage of farmland	−0.316	−0.07	−0.392	−0.352	−0.566 *	0.142	−0.235	−0.177	−0.632 **	−0.693 **	−0.657 **
percentage of bare land	−0.535 *	−0.729 **	−0.475 *	−0.508 *	−0.37	−0.118	−0.368	−0.385	−0.21	−0.15	−0.226
percentage of hardscape	0.470 *	0.348	0.414	0.485 *	0.451	0.560 *	0.637 **	0.674 **	−0.072	−0.071	−0.024
percentage of canal	0.434	0.214	0.529 *	0.469 *	0.569 *	−0.056	0.168	0.103	0.685 **	0.703 **	0.722 **
percentage of hydraulic facilities	0.255	0.212	0.317	0.304	0.342	0.108	0.218	0.172	0.34	0.278	0.349
percentage of road	0.252	0.059	0.126	0.246	0.454	0.729 **	0.761 **	0.799 **	−0.309	−0.271	−0.292
percentage of buildings	0.344	0.22	0.375	0.389	0.333	0.177	0.321	0.347	0.124	0.193	0.23
percentage of other facilities	0.201	0.216	0.29	0.231	0.117	0.239	0.232	0.275	−0.082	−0.082	0.026
naturalness of bank	−0.479 *	−0.693 **	−0.339	−0.451	−0.203	−0.383	−0.296	−0.319	0.03	0.149	0.086
naturalness of vegetation	−0.538 *	−0.668 **	−0.399	−0.501 *	−0.074	−0.306	−0.241	−0.186	−0.023	0.185	0.065

* p < 0.05, ** p < 0.01.

displays the results of the Spearman correlation analysis between landscape characteristics and both aesthetic and recreational preferences.

Regarding the relationship between aesthetic preference and landscape characteristics, there is a significant negative correlation with the percentage of vegetation, percentage of bare land, naturalness of bank, and naturalness of vegetation. Conversely, there is a significant positive correlation with the percentage of hardscape. The three sub-factors of aesthetic preference exhibit varied relationships with landscape characteristics as follows: (1) Aesthetic quality demonstrates a significant negative correlation with the percentage of bare land, naturalness of the bank, and naturalness of vegetation. (2) Historicity exhibits a significant positive correlation with the percentage of canal and water, and a significant negative correlation with the percentage of bare land. (3) Locality displays a significant positive correlation with the percentages of hardscape and canal, and a significant negative correlation with the percentage of vegetation, percentage of bare land, and naturalness of vegetation.

Regarding the relationship between recreational preference and landscape characteristics, recreational preference shows a significant positive correlation with the percentages of canal and water, and a significant negative correlation with the percentage of farmland. Although the six recreational activities vary in their relationships with landscape characteristics, they exhibit similar correlations and can be grouped into two categories as follows: (1) Walking, jogging, and biking each show a significant positive correlation with the percentages of hardscape and road. (2) Swimming, boating, and fishing each demonstrate a significant positive correlation with the percentages of canal and water, and a significant negative correlation with the percentage of farmland.

Table 4. Result of stepwise regression.

Dependent Variable	Independent Variable	Unstandardized Coefficients		Standardized Coefficients	t	p	Collinearity Statistics
		B	Std. Error	Beta			VIF	Tolerance
aesthetic preference (R2 = 0.769, Adj R2 = 0.72)	Constant	3.71	0.089	-	41.728	0.000 **	-	-
	percentage of water	1.455	0.31	0.614	4.699	0.000 **	1.037	0.964
	naturalness of bank	−0.095	0.032	−0.463	−2.95	0.011 *	1.494	0.669
	naturalness of vegetation	−0.133	0.056	−0.37	−2.392	0.031 *	1.455	0.687
aesthetic quality (R2 = 0.753, Adj R2 = 0.7)	Constant	4.056	0.116	-	34.84	0.000 **	-	-
	percentage of water	1.295	0.406	0.432	3.193	0.007 **	1.037	0.964
	naturalness of bank	−0.132	0.042	−0.506	−3.119	0.008 **	1.494	0.669
	naturalness of vegetation	−0.195	0.073	−0.43	−2.686	0.018 *	1.455	0.687
historicity (R2 = 0.722, Adj R2 = 0.685)	Constant	3.333	0.062	-	54.048	0.000 **	-	-
	percentage of water	1.732	0.31	0.775	5.593	0.000 **	1.037	0.964
	naturalness of bank	−0.102	0.027	−0.524	−3.778	0.002 **	1.037	0.964
locality (R2 = 0.797, Adj R2 = 0.753)	Constant	3.421	0.059	-	57.707	0.000 **	-	-
	percentage of water	1.257	0.259	0.599	4.844	0.000 **	1.053	0.95
	percentage of hardscape	1.359	0.469	0.355	2.897	0.012 *	1.03	0.97
	naturalness of bank	−0.109	0.023	−0.595	−4.798	0.000 **	1.058	0.945
recreational preference (R2 = 0.861, Adj R2 = 0.819)	Constant	0.194	0.035	-	5.61	0.000 **	-	-
	percentage of water	0.571	0.184	0.566	3.102	0.008 **	3.116	0.321
	percentage of canal	0.415	0.167	0.466	2.491	0.027 *	3.277	0.305
	percentage of road	2.034	0.619	0.364	3.283	0.006 **	1.149	0.87
	naturalness of bank	−0.037	0.01	−0.425	−3.791	0.002 **	1.174	0.852
walking (R2 = 0.554, Adj R2 = 0.429)	Constant	0.735	0.027	-	27.241	0.000 **	-	-
	percentage of road	2.046	0.825	0.461	2.479	0.026 *	1.03	0.971
	naturalness of bank	−0.032	0.013	−0.459	−2.47	0.026 *	1.03	0.971
jogging (R2 = 0.778, Adj R2 = 0.73)	Constant	0.224	0.052	-	4.327	0.001 **	-	-
	percentage of canal	0.794	0.154	0.716	5.148	0.000 **	1.216	0.822
	percentage of road	5.047	0.924	0.724	5.461	0.000 **	1.108	0.903
	naturalness of bank	−0.042	0.015	−0.383	−2.849	0.013 *	1.138	0.879
biking (R2 = 0.782, Adj R2 = 0.735)	Constant	0.157	0.06	-	2.64	0.019 *	-	-
	percentage of canal	0.819	0.178	0.635	4.611	0.000 **	1.216	0.822
	percentage of road	6.159	1.064	0.76	5.788	0.000 **	1.108	0.903
	naturalness of bank	−0.051	0.017	−0.403	−3.029	0.009 **	1.138	0.879
swimming (R2 = 0.905, Adj R2 = 0.892)	Constant	0.014	0.017	-	0.848	0.41	-	-
	percentage of water	0.986	0.083	0.96	11.839	0.000 **	1.037	0.964
	naturalness of bank	−0.028	0.007	−0.312	−3.844	0.002 **	1.037	0.964
fishing (R2 = 0.878, Adj R2 = 0.87)	Constant	0.112	0.03	-	3.795	0.002 **	-	-
	percentage of water	1.717	0.16	0.937	10.707	0.000 **	1	1
boating (R2 = 0.92, Adj R2 = 0.909)	Constant	−0.014	0.025	-	−0.534	0.601	-	-
	percentage of water	1.66	0.127	0.97	13.056	0.000 **	1.037	0.964
	naturalness of bank	−0.043	0.011	−0.291	−3.921	0.001 **	1.037	0.964

* p < 0.05, ** p < 0.01.

identifies the key landscape characteristics that influence aesthetic and recreational preferences and their sub-factors.

Aesthetic preference is influenced by the percentage of water, naturalness of the bank, and naturalness of vegetation. Aesthetic quality is affected by the percentage of water, naturalness of bank, and naturalness of vegetation. Historicity is influenced by the percentage of water and the naturalness of the bank. Locality is determined by the percentage of water, percentage of hardscape, and the naturalness of bank.

Recreational preference is influenced by the percentage of canal, percentage of water, naturalness of the bank, and percentage of road. Walking is mainly influenced by the naturalness of the bank and percentage of road. Jogging and biking share similar influencing factors, including the percentage of canal, naturalness of the bank, and percentage of road. Swimming and boating are both influenced by the percentage of water and naturalness of the bank. Fishing is primarily influenced by the percentage of water.

3.3. Social Background and Public Preference

Table 5. Result of the one-way ANOVA.

	N	Aesthetic Preference	Aesthetic Quality	Historicity	Locality	Recreational Preference	Walking	Swimming	Fishing	Boating
Female	254	3.58 ± 0.76	3.77 ± 0.70	3.44 ± 0.86	3.52 ± 0.81	0.39 ± 0.17	0.72 ± 0.25	0.48 ± 0.28	0.43 ± 0.27	0.12 ± 0.17
Male	211	3.63 ± 0.80	3.79 ± 0.75	3.50 ± 0.90	3.61 ± 0.86	0.38 ± 0.18	0.73 ± 0.25	0.47 ± 0.30	0.42 ± 0.28	0.14 ± 0.18
<20	24	3.35 ± 0.71 b	3.63 ± 0.62	3.14 ± 0.76 B	3.28 ± 0.80 b	0.39 ± 0.16	0.74 ± 0.25	0.53 ± 0.27	0.43 ± 0.25	0.08 ± 0.11
20~29	102	3.47 ± 0.78 b	3.69 ± 0.72	3.30 ± 0.87 B	3.44 ± 0.87 b	0.38 ± 0.15	0.70 ± 0.25	0.46 ± 0.28	0.40 ± 0.25	0.11 ± 0.15
30~39	162	3.55 ± 0.69 b	3.74 ± 0.66	3.37 ± 0.83 B	3.53 ± 0.72 b	0.39 ± 0.16	0.73 ± 0.23	0.46 ± 0.27	0.43 ± 0.26	0.13 ± 0.18
40~49	86	3.72 ± 0.85 ab	3.90 ± 0.76	3.65 ± 0.93 AB	3.62 ± 0.92 ab	0.39 ± 0.18	0.74 ± 0.25	0.51 ± 0.30	0.43 ± 0.28	0.14 ± 0.18
50~59	67	3.83 ± 0.80 a	3.92 ± 0.80	3.75 ± 0.86 A	3.80 ± 0.82 a	0.39 ± 0.19	0.77 ± 0.26	0.49 ± 0.33	0.42 ± 0.35	0.16 ± 0.20
>60	24	3.74 ± 0.91 ab	3.82 ± 0.82	3.71 ± 0.95 AB	3.70 ± 1.00 ab	0.36 ± 0.25	0.62 ± 0.35	0.44 ± 0.33	0.44 ± 0.32	0.14 ± 0.19
Junior high school and below	37	4.07 ± 0.88 A	4.20 ± 0.82 A	3.99 ± 0.94 A	4.01 ± 0.91 A	0.30 ± 0.21 C	0.62 ± 0.35 B	0.41 ± 0.33	0.35 ± 0.31	0.11 ± 0.18 bc
High school/Technical secondary school	39	3.86 ± 0.89 A	4.02 ± 0.81 A	3.79 ± 0.93 A	3.76 ± 0.98 AB	0.33 ± 0.17 BC	0.73 ± 0.30 AB	0.45 ± 0.32	0.40 ± 0.35	0.06 ± 0.09 c
Junior college	59	3.83 ± 0.79 A	3.90 ± 0.75 AB	3.76 ± 0.87 A	3.84 ± 0.83 A	0.43 ± 0.18 A	0.79 ± 0.24 A	0.51 ± 0.29	0.46 ± 0.30	0.18 ± 0.19 a
Bachelor’s degree	293	3.48 ± 0.72 B	3.69 ± 0.68 B	3.32 ± 0.84 B	3.44 ± 0.79 B	0.40 ± 0.16 AB	0.74 ± 0.23 A	0.49 ± 0.28	0.43 ± 0.26	0.13 ± 0.17 b
Master’s degree or above	37	3.47 ± 0.64 B	3.67 ± 0.64 B	3.30 ± 0.75 B	3.42 ± 0.66 B	0.36 ± 0.20 ABC	0.62 ± 0.27 B	0.40 ± 0.29	0.39 ± 0.25	0.14 ± 0.23 ab
<5000 RMB	128	3.81 ± 0.85 A	4.00 ± 0.75 A	3.69 ± 0.96 A	3.75 ± 0.93 a	0.38 ± 0.18	0.74 ± 0.28	0.50 ± 0.30	0.41 ± 0.30	0.12 ± 0.16
5000–10,000 RMB	228	3.56 ± 0.75 B	3.75 ± 0.71 B	3.43 ± 0.86 B	3.51 ± 0.81 b	0.39 ± 0.17	0.74 ± 0.24	0.48 ± 0.30	0.44 ± 0.28	0.14 ± 0.18
10,000–30,000 RMB	100	3.47 ± 0.68 B	3.60 ± 0.65 B	3.32 ± 0.77 B	3.47 ± 0.73 b	0.38 ± 0.16	0.69 ± 0.22	0.45 ± 0.27	0.42 ± 0.26	0.13 ± 0.18
>30,000 RMB	9	3.22 ± 0.76 B	3.46 ± 0.72 B	2.99 ± 0.85 B	3.20 ± 0.85 b	0.29 ± 0.12	0.56 ± 0.31	0.31 ± 0.23	0.31 ± 0.15	0.04 ± 0.05
No, I have always lived in the city	75	3.43 ± 0.80 a	3.62 ± 0.74 a	3.28 ± 0.91 a	3.40 ± 0.87 a	0.35 ± 0.17	0.69 ± 0.26	0.43 ± 0.30	0.41 ± 0.29	0.09 ± 0.15
Yes, but for less than 2 years	80	3.47 ± 0.74 b	3.68 ± 0.68 ab	3.28 ± 0.87 b	3.44 ± 0.79 ab	0.37 ± 0.16	0.68 ± 0.26	0.44 ± 0.27	0.40 ± 0.25	0.13 ± 0.17
Yes, for more than 2 years	310	3.68 ± 0.77 b	3.85 ± 0.72 b	3.56 ± 0.86 b	3.64 ± 0.83 b	0.40 ± 0.17	0.74 ± 0.25	0.50 ± 0.29	0.43 ± 0.28	0.14 ± 0.18
Completely unheard of	55	3.38 ± 0.83 B	3.61 ± 0.81 B	3.24 ± 0.93 B	3.28 ± 0.87 C	0.36 ± 0.18	0.72 ± 0.28	0.43 ± 0.31	0.41 ± 0.31	0.09 ± 0.15
Heard of but not aware	221	3.55 ± 0.78 B	3.75 ± 0.72 B	3.39 ± 0.88 B	3.51 ± 0.83 BC	0.38 ± 0.17	0.73 ± 0.24	0.47 ± 0.29	0.41 ± 0.28	0.13 ± 0.17
Somewhat aware	157	3.66 ± 0.71 B	3.80 ± 0.66 B	3.54 ± 0.80 B	3.63 ± 0.78 B	0.39 ± 0.17	0.71 ± 0.26	0.49 ± 0.28	0.44 ± 0.27	0.13 ± 0.18
Very aware	32	4.10 ± 0.81 A	4.18 ± 0.78 A	4.05 ± 0.89 A	4.08 ± 0.85 A	0.43 ± 0.17	0.76 ± 0.25	0.55 ± 0.32	0.47 ± 0.30	0.18 ± 0.16
Not local	373	3.52 ± 0.77 B	3.71 ± 0.71 B	3.38 ± 0.88 B	3.49 ± 0.83 B	0.37 ± 0.16 B	0.71 ± 0.26 b	0.46 ± 0.28 B	0.41 ± 0.27 B	0.12 ± 0.16 B
Local	92	3.93 ± 0.73 A	4.08 ± 0.70 A	3.84 ± 0.80 A	3.86 ± 0.78 A	0.45 ± 0.18 A	0.78 ± 0.22 a	0.55 ± 0.30 A	0.49 ± 0.29 A	0.18 ± 0.22 A

Significance at the 0.05 level is indicated by lowercase letters (^abc), and at the 0.01 level by uppercase letters (^ABC). Factors showing no significance are not marked. For ease of reading and layout, jogging and biking, which did not demonstrate significant differences, have been omitted from all results.

presents the results of the one-way ANOVA. As indicated in the table, gender does not have a significant impact on public preferences. Age significantly influences aesthetic preferences, with older individuals tending to give higher ratings for historicity and locality, and demonstrating a higher aesthetic preference. Education level significantly affects both aesthetic and recreational preferences, with highly educated individuals tending to give lower evaluations of aesthetic preference, including its three sub-factors. Highly educated individuals also prefer activities—swimming, fishing, and boating—showing a higher recreational preference. Monthly income, experience of rural life, and knowledge of historical irrigation structures all significantly influence aesthetic preferences, where individuals with higher incomes tend to give lower evaluations of aesthetic preference, including the three sub-factors. People with more extensive rural living experience and greater knowledge of historical irrigation structures tend to give higher evaluations of aesthetic preference, including the three sub-factors. Non-local and local respondents showed significant differences in all preference elements, with locals consistently scoring higher on all elements compared to non-locals.

3.4. Relationship between Cost and Both Aesthetic and Recreational Preference

We distributed a total of 219 online surveys and after excluding responses with excessively short completion times, we collected 198 valid questionnaires.

We compared the aesthetic and recreational preference scores for 24 scenarios across three sets (Figure 7). Within the aesthetic preferences, the most popular scenario among tertiary canals was ‘road + hardened banks + regular vegetation’, for branch canals it was ‘hardened banks’, and for main canals, it was ‘hardened banks’ and ‘hardened banks + regular vegetation’. The least popular for all three types of channels was ‘no road + natural banks + natural vegetation’. In terms of recreational preferences, the most favored scenario for tertiary canals was ‘hardened banks + regular vegetation’, for branch canals it was ‘regular vegetation + road’, and for main canals, ‘road + hardened banks + regular vegetation’. The scenarios with the lowest scores in all three channel types were ‘no road + natural banks + natural vegetation’.

Table 6. Result of Spearman correlation analysis between landscape characteristic, costs, and public preferences.

	Aesthetic Preference	Recreational Preference
Naturalness of vegetation (0 natural −1 regular)	0.151	0.072
Naturalness of bank (0 natural −1 hardened)	0.482 *	0.12
Percentage of road (0 absent −1 present)	0.199	0.746 **
Cost	0.541 **	0.523 **

* p < 0.05, ** p < 0.01.

shows the relationship between landscape characteristic, costs, and public preferences. Aesthetic preferences are significantly positively correlated with the naturalness of banks and costs. Recreational preferences show a significant positive correlation with roads and costs.

4. Discussion

4.1. Aesthetic Preference and Recreational Preference

Previous research has discovered aesthetic and recreational values in some spontaneously occurring activities associated with HICs [16]. This study confirms that the aesthetic and recreational values of HICs are widely recognized by the public, suggesting that this type of built heritage possesses significant value that warrants robust protection and enhancement.

Among recreational activities, the public shows a preference for activities such as walking and jogging, attributed to the distinct linear characteristics of irrigation canals, proven suitable for these activities [67,68,69]. Swimming is the least preferred activity, possibly because people generally prefer swimming in larger bodies of water such as rivers or lakes [70,71].

4.2. Landscape Characteristics and Aesthetic Preference

In stepwise regression analysis, factors predicting aesthetic preference include the percentage of water, naturalness of the bank, and naturalness of vegetation. This suggests that larger canals generally have greater aesthetic appeal, possibly because larger water bodies often enhance the aesthetic experience [54]. Respondents preferred built environment elements over natural features, such as artificial banks and vegetation, which contradicts some existing studies related to landscape preferences [36]. This may be because artificially modified HICs more effectively convey their aesthetic value as heritage. Similar findings have been reported in other studies between nature and heritage [50]. Additionally, regular planting patterns, such as hedgerows, are considered characteristic of agricultural landscapes [47,72], contributing to a unique rural aesthetic.

Percentage of vegetation did not emerge as a key variable in explaining locality or other aesthetic preference factors. However, it is noteworthy that correlation analysis revealed that extensive vegetation and a lower proportion of hardscape can actually diminish aesthetic evaluations, contrary to other research findings [73]. From a sub-factor perspective, this could be linked to locality. Vegetation, in comparison to hard landscapes, tends to evoke a weaker sense of locality, which diminishes the overall aesthetic preference evaluations. Additionally, correlation analysis has shown that larger areas of bare land are associated with lower aesthetic quality, a finding supported by previous research [74]. These bare lands are subject to seasonal influences such as vegetation withering and fluctuations in water levels, reminding us to be aware of the negative impact of seasonal changes on the quality of the HIC landscape, especially during winter and dry seasons.

According to the stepwise regression results for aesthetic preference sub-factors, the naturalness of the bank and the percentage of water are crucial in predicting all sub-factors. More artificial vegetation is associated with higher perceived aesthetic quality. However, the naturalness of vegetation is not included among the predictive factors for historicity and locality, possibly because both naturally and artificially planted forms of vegetation can evoke a sense of history [75]. The percentage of hardscape is also a critical factor in predicting locality. This may be attributed to built structures more likely evoking a sense of heritage or place than natural elements [26,76].

4.3. Landscape Characteristics and Recreational Preference

Based on stepwise regression results, recreational preference is primarily influenced by the canal itself, specifically through the percentage of canal, percentage of water, and the naturalness of the bank. Additionally, the percentage of road also predicts recreational preference. This may be attributed to the activities provided relying on the canal itself or the linear activity space it offers [67]. Correlation analysis also revealed that extensive farmland can make people feel that activities are restricted, possibly because people perceive farmland as inaccessible and unsuitable for recreation compared to grasslands, woodlands, and paved areas. In China, the policies regulating prime farmland are very strict, which may create conflicts between farmland and rural recreation. It is necessary to focus on coordination in planning and management, and consider measures such as integrated development [77].

In terms of specific recreational activities, individuals tend to prefer well-maintained vegetation landscapes, aligning with previously discussed aesthetic preferences. Other recreational activities are primarily influenced by the characteristics of the canal itself and the adjacent roads. Among these, the percentage of canal can predict jogging and biking, likely because these activities rely more on the linear space provided by the canal than walking. The percentage of water predicts swimming, boating, and fishing, as these activities depend heavily on extensive water surfaces. The naturalness of the bank predicts walking, swimming, and boating, likely because individuals prefer these activities in a more controlled environment rather than a highly natural or wild setting. The percentage of road predicts activities such as walking, jogging, and biking, which is logical. Therefore, in the multifunctional development of canals, it is advisable to provide waterfront roads tailored to different linear activities based on the scale of the HICs, such as designating wider roads for bike paths alongside main canals.

4.4. Social Background and Both Aesthetic and Recreational Preference

The higher ratings for aesthetic preferences among the elderly population may be attributed to their strong sense of place attachment, which enables them to observe and recognize the historical value and uniqueness of rural landscapes and historical irrigation structures more readily, thereby resulting in higher evaluations of historicity and locality. Individuals who have extensive rural living experience and possess more knowledge of historical irrigation structures also tend to give higher aesthetic ratings. This suggests that rural living experience and an understanding of rural and historical irrigation structures can enhance aesthetic appreciation, consistent with previous research [57,78]. This underscores the importance of increasing the introduction and promotion of the historical irrigation structures, especially to urban visitors, including media campaigns, local science outreach, and the use of signage and interpretive systems, for the multifunctional development of rural HICs. Additionally, the preference for fishing among middle-aged and younger adults warrants attention, as also noted in related studies [79]. This may represent a significant recreational activity to attract young people to rural areas.

Studies have found that individuals with higher education levels tend to rate aesthetic preferences lower, which is inconsistent with existing research [62]. This discrepancy may be attributed to their broader aesthetic perspectives and experiences, leading to higher evaluation standards. Highly educated individuals appear more capable of envisaging the possibilities of activities, thus giving higher ratings for recreational preferences, possibly due to their more diverse entertainment experiences and demands, especially in relatively rare recreational activities such as swimming, fishing, and boating. These findings highlight the importance of high-quality landscapes and diverse recreational activities in attracting high-income and highly educated populations, warranting consideration in future planning and management.

There are significant differences in preferences for all elements between locals and non-locals, with locals consistently giving higher scores than non-locals. This may be attributed to a sense of cultural identity or pride that locals have towards HICs. They may also have expectations for more changes to attract non-locals to the area. During the on-site questionnaire survey, many locals expressed pride and envisioned activities taking place, indicating that HICs also enjoy support from local stakeholders from the bottom up. This support suggests that the conservation and utilization of HICs positively contribute to local identity and residents’ well-being.

4.5. Cost and Both Aesthetic and Recreational Preference

After considering cost factors, the most significant landscape characteristic affecting public aesthetic preferences is the naturalness of banks, with respondents showing a preference for hardened banks. When cost factors are considered, the public places the greatest significance on the presence of roads in their recreational preferences, favoring scenarios that include roads. Furthermore, costs are positively correlated with public aesthetic and recreational preferences, indicating that the public is willing to pay certain costs for better aesthetic and recreational experiences. Compared to scenarios where cost is not considered, the naturalness of vegetation does not significantly correlate with aesthetic preferences, and the naturalness of banks is not significantly correlated with recreational preferences, suggesting that people are not necessarily willing to pay to change the naturalness of vegetation or banks when considering costs. This implies that in practical planning and management, managers might prioritize constructing hardened banks for aesthetic preferences and roads for recreational preferences, especially when costs are limited.

4.6. Limitations

This study focuses on the potential for aesthetic and recreational development of HICs in rural areas, offering assessments and recommendations based on public preferences. However, in the actual development process, other stakeholders such as management bodies and cultural tourism enterprises should also be considered, along with a variety of factors such as agricultural production and ecological conservation, all of which require further research. This paper explores recreational preferences involving various activity types, some of which may be subject to restrictions in practice, such as limitations imposed by management authorities on the scale and scope of activities like boating and fishing. Therefore, the implementation of recreational activities should also consider different stakeholders and management requirements, providing a direction for future research. Additionally, the study concentrates mainly on HICs in the water network plains of southern China, and the conclusions may not be applicable to other types of historical irrigation structures, such as those in mountainous or arid regions. Significant potential remains for related research in these areas.

5. Conclusions

In summary, the study demonstrates that HICs provide significant aesthetic and recreational value to the public. Overall, the differences between aesthetic quality, historicity, and locality are not significant, whereas the variations among different recreational activities are more significant. Larger water bodies, manicured vegetation, and artificial banks enhance aesthetic preference. Additionally, percentage of hardscape contributes to historicity and locality. Larger canals, water bodies, and artificial canal banks enhance recreational preference. The public prefer jogging and biking along larger canals, opting for broader water bodies for swimming, boating, and fishing. They also favor walking, swimming, and boating along artificial canal banks. A larger proportion of roads encourages walking, jogging, and biking. Older individuals, those with lower incomes, lower education levels, individuals with extensive rural living experience, and those with greater knowledge of historical irrigation structures tend to give higher evaluations of aesthetic preference. Middle-aged and younger adults show a preference for fishing, while highly educated individuals exhibit a higher recreational preference. Local residents have a higher aesthetic and recreational preference for HICs compared to non-locals. When considering costs, hardened banks and the provision of roads are more favored by the public.

This research offers valuable insights and recommendations for the conservation and utilization of HICs in rural areas. Firstly, the aesthetic and recreational values of canals warrant protection and utilization. This not only attracts non-locals to rural areas but also influences the local residents’ sense of place identity and well-being. Aesthetic quality, historicity, and locality all require consideration in planning, while the arrangement of recreational activities should be tailored to local conditions. Secondly, in contrast to urban landscapes, preserving built environment and artificial structures, instead of excessive naturalization, can provide better aesthetic and recreational experiences. The canal itself significantly impacts individuals’ experiences, particularly its size and water volume. Tailored development and utilization strategies should be formulated for various canal scales. If costs are limited, interventions involving banks and roads can be prioritized. Thirdly, it is essential to promote and popularize heritage and rural knowledge. Providing higher quality landscapes and a more diverse range of recreational activities (especially swimming, boating, and fishing) can attract more affluent individuals, thereby boosting the rural tourism economy.