-
The study area was located in the Banlao and Banhong Townships, where the Asian elephant is distributed in the west area and surroundings of the Cangyuan part of Nangunhe National Nature Reserve (WA-S-CNNNR) (98°55′41″E–99°04′48″E, 23°11′16″N–23°19′34″N), Lin-cang City, Yunnan Province, China. It has an altitude between 640–1451 m. The region below 900 m is of sub-humid monsoon climate in North tropical valley, which is hot and dominated by latosols with strong evaporation. Main crops planted by local people were rice, corn, banana and rubber, with chicken, duck, pig and cattle being the primary livestock (Tang et al., 2015).
Information about Asian elephants’ distribution region was extracted from the literature (Liu et al., 2016), officers and rangers of Nangunhe NNR when determining our study area. The 32 settlements nearby or within the distribution region of the Asian elephants (Liu et al., 2016) were chosen to conduct interviews. These settlements belonged to seven villages: Shangbanlao, Yingpan, Xinzhai and Panang of Banlao Township, and Mangku, Nanban and Banhong of Banhong Township. The settlements visited are individually marked on Fig. 1. The two areas (Banlao and Banhong Township) differed in the extent of tourism development and festival celebrations, which encouraged us to investigate whether there were differences in the attitudes of the local people regarding elephants and we, therefore, incorporated ‘Area’ as a predictor variable in our data analysis.
-
The questionnaire, which categorized family as a unit, was divided into three parts: basic information, status of HEC with attitude towards conservation and resource utilization. In the basic information part, questions were issued about participants’ ethnic groups, family size, family annual income, family members’ quantity in non-primary industries (primary industries include crop-plantation, forestry, animal husbandry and fishery), main crops cultivated, and primary livestock raised. In the HEC part, respondents were asked if they had seen wild Asian elephants in the past three years and what kind of conflict arose from encountering Asian elephants. Respondents were also inquired if they had used any methods to avoid or reduce their losses caused by conflicts with Asian elephants. If the answer was ‘yes’, respondents were then asked to further explain the methods they had adopted and other methods they considered adequate. In the part regarding attitudes towards conservation and utilization, we investigated respondents’ perceptions towards Asian elephants and the developments of tourism, and what values they perceive in elephants.
The interviews were conducted in October 2018 when researchers were divided into two groups. Each group was assigned with a local ranger to help explain the questionnaire to local people in their local dialect. The number of families interviewed was dependent on the size of the settlement being concerned. If the settlement was a regular rural group (administrative unit in China), we would interview 10 families unless there were not enough ones at home. If the settlement was incorporated by multiple groups with a greater population than one regular group, we would interview 11–20 families there. Respondents were chosen randomly and each one completed a questionnaire. Many people were unable to finish the questionnaire independently due to illiteracy, in which case the researchers would read out the questions with the interpretation of rangers for better understanding. Researchers also communicated more casually with respondents, which provided more information on what they experienced and their customs. New options would be added when respondents gave answers not initially listed in the preset options.
A total of 328 questionnaires were retrieved after the interviews, among which 327 were considered valid. There were 178 valid questionnaires from Banlao Township and 149 valid questionnaires from Banhong Township, respectively.
-
Data were logged into Excel 2003 and organized for further analysis in R (R Core Team, 2019). Family annual income, quantity of family members in non-primary industries, attitudes towards elephants, perceptions of elephants’ values and attitude towards tourism development were listed in a rank from low to high (Table 1). The rank for positive part of ‘Attitude towards elephants’ was based on how voluntary the attitude was, and was according to respondents’ reason about their positive attitude towards elephants. In our interviews, we questioned ‘Why do you think elephants are good?’ when respondents said they have positive attitude towards elephants. We found the culturally-motivated respondents introduced their cultures first and told us the positive meanings brought by elephants, while some only attributed to the government and the laws protecting elephants, hence these animals must be good. Therefore, we ranked a lower value for those positive attitudes based on a sense of legal enforcement than those based on cultural values. We ranked the highest score to another positive attitude that referred to the feeling of love towards elephants, without obligation from culture values or regulations from law.
Item Option Rank Family annual income < 10000 yuan (RMB) (< 1511.17 US$) 1 10000–50000 yuan (1511.17–7555.84 US$) 2 50000−100000 yuan (7555.84–15111.68 US$) 3 >100000 yuan (> 15111.68 US$) 4 Quantity of family members in non-primary industries 0 0 1 1 2 2 > 2 3 Attitude towards elephants Very bad (hate elephants and cannot endure to coexist) 1 Bad (dislike elephants but can tolerate the coexistence) 2 Somewhat bad (have average attitude towards elephants but think they have no value) 3 Neutral 4 Reasonably good (be willing to conserve elephants for being law-abiding) 5 Good (consider elephants as holy and auspicious animals because of the culture rather than from powerful influences like the government and the laws) 6 Very good (would like to protect elephants from the heart without any other influence) 7 Perception of the values of elephants No value 1 Existence is the only value 2 Valuable (economic value, cultural value, ecological value, scientific value, etc.) 3 Attitude towards tourism development It is not achievable 1 It is achievable but the respondent will not participate in it 2 It is achievable and the respondent wants to participate in it 3 Table 1. The assigned items, options and ranks in the questionnaire for human-elephant interactions in the west area and surroundings of the Cangyuan part of Nangunhe National Nature Reserve, Yunnan, China
Pearson’s Chi-squared test was used in R (R Core Team, 2019) to examine the relationship among ‘Area’,‘Family size’, ‘Annual income’ , ‘Quantity of family members in non-primary industries’ and ‘Experiencing loss or not’, which all have a significant correlation with one another (Table 2), indicating that they were not assembled by the same variable combination. A group of 16 logistic regression models was then applied, each model employing either the zero model or one of the 15 variable combinations (Table 3). There were three response variables: ‘Attitude towards elephants’, ‘Perception of the values of elephants’, and ‘Attitude towards tourism development’.
P-value Area Family size Annual income Quantity of family members in non-primary industries Experiencing loss or not Area / 0.63 0.06 0.12 0.19 Family size / 0.01 ** 4.76 × 10−4 *** 0.87 Annual income / 5.20 × 10−3 *** 0.45 Quantity of family members in non-primary industries / 0.83 Experiencing loss or not / Notes: ** Significant at 1% level; *** Significant at 0.1% level Table 2. Correlation of predictor variables used in logistic regression model
Sequence number Variable combination 1 Area 2 Family size 3 Annual income 4 Quantity of family members in non-primary industry 5 Experiencing loss or not 6 Area + Family size 7 Area + Annual income 8 Area + Quantity of family members in non-primary industry 9 Area + Experiencing loss or not 10 Experiencing loss or not + Family size 11 Experiencing loss or not + Annual income 12 Experiencing loss or not + Quantity of family members in non-primary industry 13 Area + Experiencing loss or not + Family size 14 Area + Experiencing loss or not + Annual income 15 Area + Experiencing loss or not + Quantity of family members in non-primary industry Table 3. The variable combinations used in the logistic regression models
We used package ‘MASS’ (Venables and Ripley, 2002) in R (R Core Team, 2019) to establish 16 Ordinal Logistic Regression Models (OLRM) for each response variable. These included models with the 15 variable combinations and a zero model. The package ‘brant’ (Schlegel and Steenbergen, 2020) was used to test if each model met the parallel regression assumption, and if OLRM was suitable for the models that did (Zhao et al., 2009). ‘Parallel regression assumption holds’ were accepted if the P-value for the omnibus test was higher than 0.05.
Models that met the parallel regression assumption were ranked according to Akaike’s Information Criterion (AIC), using the R package ‘MuMIn’ (Bartoń, 2022). The most representative models were then analyzed further. Among models for ‘Attitude towards elephants’, the best predictor variable combination was ‘Area + Experiencing loss or not + Annual income’ (AICc = 941.5, weight = 0.98). However, for the models addressing ‘Perception of the values of elephants’, while all models met the parallel regression assumption, the weight of the first ranked model was 0.37, so we did not analyze it further. Among models for ‘Attitude towards tourism development’, four models did not meet the parallel regression assumption and the remainder did not contain a representative model therefore, a Binomial Logistic Regression Model (BLRM) was used to analyze attitude towards tourism development.
We established BLRM for two response variables: ‘If people thought tourism development was achievable’ (with all data), and ‘If people would like to participate in tourism development’ (with the data of respondents who thought tourism development achievable). We used the ‘GLM’ function in R (R Core Team, 2019) to establish 16 BLRMs for each response variable. The 16 models included 15 with variable combinations and a zero model. The parameter of family used was ‘binomial’. We ranked these models using the package ‘MuMIn’ (Bartoń, 2022).
Among models for ‘If people thought tourism development was achievable’, the weight of the first model in the rank was 0.19, followed by the zero model. Therefore, we did not do any further analysis. Among models for ‘If people would like to participate in tourism development’, the best variable combination model was ‘Area + Annual income’ (AICc = 270.0, weight = 0.61), which we analyzed further.
In further analysis, we calculated the odds ratio (OR) and 95% confidence interval (CI) in combination with P-values to assess how predictor variables affect response variables (Kabacoff, 2011).
-
Nearly all respondents were ethnically Wa people (97.86%, n = 320) who have lived in the study area for decades and retained their unique culture.
Families generally consisted of four (27.83%, n = 91) or five (24.77%, n = 81) members. The biggest family included 10 members (0.31%, n = 1) while the smallest included one member (0.92%, n = 3). Over half of the interviewed families earned an annual income between 10 000 yuan (RMB) and 50 000 yuan (57.10%, n = 185), followed by 126 families (38.89%) earning less than 10 000 yuan per year.
Most families had none family member working in non-primary industries (38.53%, n = 126), and the proportion of family decreased as the number of family members working in non-primary industries increased. Most of the interviewed families working in primary industry at various degrees, regardless of whether they engaged in other vocations or not. Only two families’ members were not engaged in farming at all, where they were not planting any crops or other plantation, and one of these two families did not raise any livestock. According to the questionnaire, the three most widely-planted crops in both Banlao and Banhong were tea, rubber tree and corn. Those with the largest area planted in Banlao were rubber tree, sugarcane and corn, while in Banhong, the widest were tea, rubber trees and corns. The three livestock species most widely-raised in Banlao and Banhong were chickens, pigs and cattle, which also owned the largest population in these two townships.
-
Interactions between Asian elephants and local people have been frequent. 17.48% (n = 57) of respondents have seen wild elephants in the past three years and 25.99% (n = 85) have suffered a loss due to these elephants. Panang suffered the most losses including 57.38% (n = 35) of interviewed families, with Mangku following at 39.36% (n = 37). There were 15.00% (n = 6) interviewed families in Nanban, 12.82% (n = 5) interviewed families in Shangbanlao, 4.88% (n = 2) interviewed families in Xinzhai which suffered losses caused by elephants over the past three years. None interviewed family experienced human-elephant conflict over the past three years in Yingpan and Banhong. Six types of HEC were reported in the area: crop raiding (71.76% of respondents reported losses, n = 61), damages to rubber plantations (28.24%, n = 24), damages to tea or coffee plantations (7.06%, n = 6), trampling field ridges (3.53%, n = 3), increased farming costs (16.47%, n = 14) and stepping in fishponds (1.18%, n = 1).
Local people have tried to avoid losses by various methods, though yielded little success. Seventy-five families have tested 12 kinds of method to protect their possessions from Asian elephants (Table 4). ‘Giving tribute to elephants’ was one of these methods, as a retained custom of the Wa people. They gave tribute, often sugarcane, bananas or other foods assumably preferred by elephants in the mountains on certain days every year; some people also paid tributes when elephants came into their fields, hoping these sacred animals would not bring too much losses.
Method Number and percentage / % of people
that have used this methodNumber and percentage / % of people
that think this method is effectiveBuilding normal fences 2 / 2.67 4 / 5.56 Building electric fences 1 / 1.33 0 Digging ditches around the fields 1 / 1.33 0 Cultivating bees 1 / 1.33 0 Lighting up the fields 1 / 1.33 0 Approaching and driving elephants away 2 / 2.67 5 / 6.94 Making big noises 57 / 76.00 29 / 40.28 Threatening elephants by fires 7 / 9.33 5 / 6.94 Threatening elephants by gun noises 1 / 1.33 0 Guarding the field all the time 4 / 5.33 2 / 2.78 Flashlights 5 / 6.67 0 Giving tributes to elephants 17 / 22.67 11 / 15.28 Table 4. Methods adopted by local communities to mitigate human-elephant conflicts in the west area and surroundings of the Cangyuan part of Nangunhe National Nature Reserve, Yunnan, China
Moreover, since elephants were considered holy and auspicious animals in the study area, some people even stated that they allowed elephants to forage in their fields. This phenomenon appeared to be more common in Banlao than in Banhong Township (Banlao: 16.28%, n = 28; Banhong: 2.68%, n = 4).
-
Table 5 illustrates respondents’ individual attitudes towards conservation and utilization. There were 10 kinds of values that respondents thought elephants possessed: value to nature, value to humans, value to the nation, ornamental value, value to research, commercial value, bringing eco-compensation, bringing attention, ivory and leather (some people used them years ago), and faeces as fertilizers (elephants leave faeces after foraging in the fields, which can make good fertilizers).
Item Option Banlao Banhong Attitude towards elephants Very bad (hate elephants and cannot endure to coexist) 1 / 0.56 3 / 2.01 Bad (dislike elephants but can tolerate the coexistence) 1 / 0.56 4 / 2.68 Somewhat bad (have average attitude towards elephants but think they have no value) 2 / 1.12 3 / 2.01 Neutral 16 / 8.99 28 / 18.79 Reasonably good (be willing to conserve elephants for being law-abiding) 29 / 16.29 48 / 32.21 Good (considering elephants as holy and auspicious animals because of the culture rather than from powerful influences like the government and the laws) 64 / 35.96 32 / 21.48 Very good (would like to protect elephants from the heart without any other influence) 65 / 36.52 30 / 20.13 Blank 0 1 / 0.67 Perception of the values of elephants No value 4 / 2.25 7 / 4.70 Existence is the only value 35 / 19.66 37 / 24.83 Valuable (economic value, cultural value, ecological value, scientific value, etc.) 138 / 77.53 102 / 68.46 Blank 1 / 0.56 3 / 2.01 Attitude towards tourism development It is not achievable 14 / 7.87 11 / 7.38 It is achievable but the respondent will not participate in it 42 / 23.60 17 / 11.41 It is achievable and the respondent wants to participate in it 119 / 66.85 115 / 77.18 Blank 3 / 1.69 6 / 4.03 Table 5. Respondents’ attitudes towards conservation and utilization (number/percentage (%) in the west area and surroundings of the Cangyuan part of Nangunhe National Nature Reserve, Yunnan, China
People’s attitudes towards elephants in Banlao Township were 2.75 times more positive than in Banhong Township, when the other two variables were kept constant (Table 6). However, if people suffered loss due to elephants in the past three years, a worse attitude towards elephants was reported. The degree of people’s attitudes towards elephants would become two times better if they did not suffer loss due to elephants in the past three years. Higher annual income also contributed to generating a more positive attitude towards elephants. People’s attitudes towards elephants were 2.09 times higher if their annual income grew to the next level, given the other two variables were kept constant (Table 6).
Variable OR 95% CI P-value Lower Higher Area (Banlao) 2.75 1.82 4.18 1.73 × 10−6*** Experiencing loss or not (Yes) 0.50 0.32 0.79 3.29 × 10−3*** Annual income 2.09 1.46 3.00 5.45 × 10−5*** Note: *** Significant at 0.1% level Table 6. The odds ratio (OR), 95% confidence interval (CI), and P-value of variables in the best model for ‘Attitude towards elephants’ in the west area and surroundings of the Cangyuan part of Nangunhe National Nature Reserve, Yunnan, China
The willingness to participate in tourism development was 2.95 times higher in Banhong Township than in Banlao Township. Among those who felt that the development of tourism was feasible, the people of Banhong Township preferred to participate in it. The willingness to participate in tourism development was 3.38 times higher among people with higher annual income compared to those with lower annual income (Table 7).
Variable OR 95% CI P-value Lower Higher Area (Banlao) 0.34 0.17 0.64 1.18 × 10−3*** Annual income 3.38 1.90 6.22 5.68 × 10−5*** Note: *** Significant at 0.1% level Table 7. The odds ratio (OR), 95% confidence interval (CI), and P-value of variables in the best model for The willingness to participate in tourism development in the west area and surroundings of the Cangyuan part of Nangunhe National Nature Reserve, Yunnan, China
Inferring Human-elephant Coexistence Based on Characteristics of Human-elephant Interactions in Nangunhe of Yunnan, China
doi: 10.1007/s11769-023-1332-6
- Received Date: 2021-05-17
- Accepted Date: 2021-11-08
- Available Online: 2023-03-06
- Publish Date: 2023-03-05
-
Key words:
- Asian elephant (Elephas maximus) /
- Wa people /
- human-elephant conflict /
- coexistence /
- Nangunhe National Nature Reserve /
- Yunnan /
- China
Abstract: Human-wildlife conflict (HWC) negatively impacts both humans and wildlife. Attitudes of local residents have been critical in promoting wildlife conservation. It is therefore necessary to understand the characteristics of HWC and identify influential factors on attitudes towards conservation to implement conservation strategies efficiently. This research focused on features of human-elephant interactions, while attitudes and values regarding the small population of Asian elephants (Elephas maximus) in Nangunhe National Nature Reserve (NNR), Yunnan, China. The total of 327 valid questionnaires were gathered around the area where Asian elephants were distributed. Logistic regression models were employed to analyze the correlations among five predictor variables (‘Area’, ‘Family size’, ‘Annual income’, ‘Quantity of family members in non-primary industries’ and ‘Experiencing loss or not’) and three response variables (‘Attitude towards elephants’, ‘Perception of the values of elephants’ and ‘Attitude towards tourism development’). The study area was densely forested with tea plants, rubber trees, corns and sugarcane. There, 25.99% of respondents reported the experience of human-elephant conflict (HEC), with crop raiding and cash crop damages being the major conflict types. To demonstrate respect for elephants and to mitigate HEC, a unique custom called ‘Giving tribute to elephants’ was developed long ago. Respondents’ township with an official annual festival of ‘Giving Tribute to Elephants’ (odds ratio (OR) = 2.75, P = 1.73 × 10−6) and higher annual income (OR = 2.09, P = 5.45 × 10−5) significantly contributed to forming a more positive attitude towards elephants, whereas HEC itself have contributed to a more negative attitude (OR = 0.50, P = 3.29 × 10−3). Therefore, we propose that: 1) reducing human-elephant conflict by testing multiple mitigation measures and adopting the most effective one of them; 2) enhancing local livelihoods through the development of ecological products and ecotourism; and 3) preserving and developing the Wa culture in this region. The study area deserves more attention and further research to explore and obtain endorsement from the public to achieve coexistence between human and wildlife.
Citation: | WANG Jiahui, CHEN Ying, SUN Yakuan, LYU Zhuoluo, SHI Kun, 2023. Inferring Human-elephant Coexistence Based on Characteristics of Human-elephant Interactions in Nangunhe of Yunnan, China. Chinese Geographical Science, 33(2): 363−376 doi: 10.1007/s11769-023-1332-6 |