Remote sensing of plant phenophases and pulsed ecosystem dynamics for wildlife ecological applications

目前，我们对野生动物-栖息地相互作用和自然生态系统功能的其他方面的了解受到我们经常和在大范围内观察复杂环境属性的能力的限制。虽然遥感长期以来一直被确定为支持生态应用的关键技术，但其在多学科项目中的有效利用尚未得到充分发展。这项工作旨在生成必要的工具和处理策略，帮助科学家使用地球轨道卫星的定期观测，在非常大的区域内观察物候模式-周期性重复发生的事件，如春季萌芽、浆果萌发或秋季衰老。要解决的具体挑战包括：(1)确定主要野生动物食用植物达到物候发育不同阶段的累积温度阈值；(2)建立实地观察地点，观察地面年度植物发育周期的进展；(3)开发数字图像处理战略，使技术人员能够从嘈杂的卫星数据中提取物候测量；(4)校准和验证统计模型，使我们能够利用遥感预测森林树冠下特定植物的物候；(5)调查灰熊、麋鹿和驯鹿等野生动物如何应对整个生长季不断变化的食用植物分布模式。这项工作的成果将有助于野生动物的管理和保护，旨在帮助任何在生态背景下处理遥感技术的人。

Our understanding of wildlife-habitat interactions and other aspects of natural ecosystem functioning is currently limited by our capacity to observe complex environmental attributes frequently and over large areas. While remote sensing has long been identified as a key technology for supporting ecological applications, its effective use in multi-disciplinary projects is not yet fully developed. This work is designed to generate the tools and processing strategies necessary to help scientists observe phenological patterns - periodically recurrent events such as spring budburst, berry emergence, or autumn senescence - over very large areas using regular observations from earth-orbiting satellites. Specific challenges to be addressed include (i) determining the cumulative temperature thresholds beyond which key wildlife food plants reach various stages of phenological development; (ii) establishing field sites to observe the progression of annual plant development cycles on the ground; (iii) developing the digital-image-processing strategies that will allow technicians to extract phenological measurements from noisy satellite data; (iv) calibrating and validating statistical models that enable us to predict the phenology of specific plants underneath the forest canopy using remote sensing; and (v) investigating how wildlife such as grizzly bear, elk, and caribou respond to changing patterns of food plant distribution throughout the growing season. The results of this work will contribute to management and conservation of wildlife, and is intended to assist anyone dealing with remote sensing technology in an ecological context.