基于微博大数据的气候变化对中国赏樱旅游的影响

研究气候变化对生态系统旅游和休闲功能的影响是旅游地理学的重要领域。本文发展和优化了基于Logistic曲线从微博大数据中提取赏樱旅游活动开展时间的方法,并从2879033条与樱花相关的新浪微博大数据中筛选出587891条有效微博,进而提取和重建了中国2010—2019年赏樱旅游活动时间数据集,该数据集通过了物候站点观测数据和与温度响应关系的验证。在此基础上分析了过去10 a中国21个城市赏樱旅游活动始日、末日和持续期的时空格局,并模拟和预测了SSP2-4.5和SSP5-8.5两种气候情景下,2020—2050年中国赏樱旅游活动时间的变化趋势。结果表明：① 过去10 a,全国大部分城市赏樱旅游活动始日提前（占比61.9%）,末日提前（占比76.2%）,持续期缩短（占比52.4%）。纬度每升高1°,始日、末日分别推迟0.286 d（P < 0.01）和0.394 d（P < 0.01）,持续期缩短0.286 d（P > 0.05）。② 未来气候情景下,大部分城市赏樱旅游活动始日和末日提前,持续期延长。SSP5-8.5情景对赏樱旅游活动的影响比SSP2-4.5情景更明显。本文的方法和框架能够为气候变化对特定旅游活动的影响研究提供基于大数据视角的参考。

The impact of climate change on cherry blossom-viewing tourism based on Weibo big data in China

Studying the impact of climate change on the tourism and leisure functions provided by plant ecosystems is an important field of tourism geography. Obtaining plant phenological timing data and analyzing its change characteristics and mechanism not only provides evidence of the impact of climate change on tourism but also supports the practical work of tourism responding to climate change. Based on the logistic curve fitting method, this paper developed and optimized a method to extract the timing for cherry blossom-viewing tourism, filtered 587891 valid microblogs from 2879033 Sina blogs related to cherry blossoms, and then extracted and reconstructed a dataset of cherry blossom-viewing tourism activities in China from 2010 to 2019, which has been verified by observation site records and the relationship between temperature factors. On this basis, the spatiotemporal pattern of the start timing, end timing, and duration of cherry blossom-viewing activities in the past 10 years were analyzed, and the time trend of cherry blossom-viewing activities from 2020 to 2050 were simulated and predicted under the climatic scenarios of SSP2-4.5 and SSP5-8.5. The results revealed that: (1) In the past 10 years, the start time of cherry blossom-viewing activities in most cities (61.9%) across the country was earlier, so was the end time (76.2%), and the duration was shortened (52.4%). For every 1° increase in latitude, the start time and end time were delayed by 0.286 days (P < 0.01) and 0.394 days (P < 0.01), respectively, and the duration was shortened by 0.286 days (P > 0.05). (2) Under future climate scenarios, the start time and end time of cherry blossom-viewing activities in most cities will be advanced and the duration will be extended. The SSP5-8.5 scenario has a more obvious impact on the cherry blossom-viewing activities than the SSP2-4.5 scenario. The framework and method in this paper can inspire researchers to focus more on the impact of climate change on specific tourism activities using big data methods.