Satellite-derived prefire vegetation predicts variation in field-based invasive annual grass cover after fire

IF 2 3区环境科学与生态学 Q3 ECOLOGY

Applied Vegetation Science Pub Date : 2023-12-21 DOI:10.1111/avsc.12759

Christopher R. Anthony, Cara V. Applestein, Matthew J. Germino

{"title":"Satellite-derived prefire vegetation predicts variation in field-based invasive annual grass cover after fire","authors":"Christopher R. Anthony, Cara V. Applestein, Matthew J. Germino","doi":"10.1111/avsc.12759","DOIUrl":null,"url":null,"abstract":"<div>\n \n \n <section>\n \n <h3> Aims</h3>\n \n <p>Invasion by annual grasses (IAGs) and concomitant increases in wildfire are impacting many drylands globally, and an understanding of factors that contribute to or detract from community resistance to IAGs is needed to inform postfire restoration interventions. Prefire vegetation condition is often unknown in rangelands but it likely affects variation in postfire invasion resistance across large burned scars. Whether satellite-derived products like the Rangeland Analysis Platform (RAP) can fulfill prefire information needs and be used to parametrize models of fire recovery to inform postfire management of IAGs is a key question.</p>\n </section>\n \n <section>\n \n <h3> Methods</h3>\n \n <p>We used random forests to ask how IAG abundances in 669 field plots measured in the 2-3 years following megafires in sagebrush steppe rangelands of western USA responded to RAP estimates of annual:perennial prefire vegetation cover, the effects of elevation, heat load, postfire treatments, soil moisture–temperature regimes, and land-agency ratings of ecosystem resistance to invasion and resilience to disturbance.</p>\n </section>\n \n <section>\n \n <h3> Results</h3>\n \n <p>Postfire IAG cover measured in the field was <math>\n <semantics>\n <mrow>\n <mover>\n <mn>22</mn>\n <mo>¯</mo>\n </mover>\n </mrow>\n <annotation>$$ \\overline{22} $$</annotation>\n </semantics></math>% and RAP-estimated prefire annual herbaceous cover was <math>\n <semantics>\n <mrow>\n <mover>\n <mn>15.7</mn>\n <mo>¯</mo>\n </mover>\n </mrow>\n <annotation>$$ \\overline{15.7} $$</annotation>\n </semantics></math>%. The random forest model had an <i>R</i><sup>2</sup> of 0.36 and a root-mean-squared error (RMSE) of 4.41. Elevation, postfire herbicide treatment, and prefire estimates from RAP for the ratio of annual:perennial and shrub cover were the most important predictors of postfire IAG cover. Threshold-like relationships between postfire IAG cover and the predictors indicate that maintaining annual:perennial cover below 0.4 and shrub cover below <10% prior to wildfire would decrease invasion, at low elevations below 1400 m above sea level.</p>\n </section>\n \n <section>\n \n <h3> Conclusion</h3>\n \n <p>Despite known differences between RAP and field-based estimates of vegetation cover, RAP was still a useful predictor of variation in IAG abundances after fire. IAG management is oftentimes reactive, but our findings indicate impactful roles for more inclusively addressing the exotic annual community, and focusing on prefire maintenance of annual:perennial herbaceous and shrub cover at low elevations.</p>\n </section>\n </div>","PeriodicalId":55494,"journal":{"name":"Applied Vegetation Science","volume":"26 4","pages":""},"PeriodicalIF":2.0000,"publicationDate":"2023-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Applied Vegetation Science","FirstCategoryId":"93","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1111/avsc.12759","RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ECOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Aims

Invasion by annual grasses (IAGs) and concomitant increases in wildfire are impacting many drylands globally, and an understanding of factors that contribute to or detract from community resistance to IAGs is needed to inform postfire restoration interventions. Prefire vegetation condition is often unknown in rangelands but it likely affects variation in postfire invasion resistance across large burned scars. Whether satellite-derived products like the Rangeland Analysis Platform (RAP) can fulfill prefire information needs and be used to parametrize models of fire recovery to inform postfire management of IAGs is a key question.

Methods

We used random forests to ask how IAG abundances in 669 field plots measured in the 2-3 years following megafires in sagebrush steppe rangelands of western USA responded to RAP estimates of annual:perennial prefire vegetation cover, the effects of elevation, heat load, postfire treatments, soil moisture–temperature regimes, and land-agency ratings of ecosystem resistance to invasion and resilience to disturbance.

Results

Postfire IAG cover measured in the field was $\bar{22}$ % and RAP-estimated prefire annual herbaceous cover was $\bar{15.7}$ %. The random forest model had an R² of 0.36 and a root-mean-squared error (RMSE) of 4.41. Elevation, postfire herbicide treatment, and prefire estimates from RAP for the ratio of annual:perennial and shrub cover were the most important predictors of postfire IAG cover. Threshold-like relationships between postfire IAG cover and the predictors indicate that maintaining annual:perennial cover below 0.4 and shrub cover below <10% prior to wildfire would decrease invasion, at low elevations below 1400 m above sea level.

Conclusion

Despite known differences between RAP and field-based estimates of vegetation cover, RAP was still a useful predictor of variation in IAG abundances after fire. IAG management is oftentimes reactive, but our findings indicate impactful roles for more inclusively addressing the exotic annual community, and focusing on prefire maintenance of annual:perennial herbaceous and shrub cover at low elevations.

Abstract Image

查看原文本刊更多论文

卫星衍生的火灾前植被可预测火灾后野外入侵性一年生草覆盖率的变化

目的一年生牧草（IAGs）的入侵和随之而来的野火增加正在影响全球许多旱地，因此需要了解导致或减弱群落对 IAGs 抵抗能力的因素，以便为火后恢复干预措施提供信息。牧场火灾前的植被状况通常不为人知，但它很可能会影响大面积烧毁疤痕的火灾后抵抗入侵能力的变化。像牧场分析平台（RAP）这样的卫星衍生产品能否满足火前信息需求，并用于参数化火灾恢复模型，为IAGs的火后管理提供信息，这是一个关键问题。方法我们使用随机森林来研究美国西部鼠尾草干草原牧场发生特大火灾后 2-3 年内测量的 669 个野外地块中的 IAG 丰度如何响应 RAP 估算的年度：多年生火前植被覆盖度、海拔、热负荷、火后处理、土壤水分-温度机制的影响，以及土地机构对生态系统抗入侵性和抗干扰能力的评级。结果实地测量的火后 IAG 覆盖率为 22 ¯$ \overline{22}$$ %，而 RAP 估算的火灾前年度草本植物覆盖率为 15.7 ¯ $$ \overline{15.7} $$ %。$$ %.随机森林模型的 R2 为 0.36，均方根误差 (RMSE) 为 4.41。海拔高度、火后除草剂处理以及 RAP 对一年生与多年生植物和灌木覆盖率的火前估计是预测火后 IAG 覆盖率的最重要因素。火灾后 IAG 覆盖率与预测因子之间的阈值关系表明，在海拔 1400 米以下的低海拔地区，野火前将一年生与多年生植物覆盖率保持在 0.4 以下、灌木覆盖率保持在 10% 以下将减少入侵。结论尽管已知 RAP 与基于实地估算的植被覆盖度之间存在差异，但 RAP 仍能有效预测火灾后 IAG 丰度的变化。IAG 的管理通常是被动的，但我们的研究结果表明，在低海拔地区，更全面地处理外来一年生植物群落，并注重火前维持一年生和多年生草本植物及灌木的覆盖率，将起到重要作用。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Applied Vegetation Science 环境科学-林学

CiteScore

6.00

自引率

10.70%

发文量

审稿时长

3 months

期刊介绍： Applied Vegetation Science focuses on community-level topics relevant to human interaction with vegetation, including global change, nature conservation, nature management, restoration of plant communities and of natural habitats, and the planning of semi-natural and urban landscapes. Vegetation survey, modelling and remote-sensing applications are welcome. Papers on vegetation science which do not fit to this scope (do not have an applied aspect and are not vegetation survey) should be directed to our associate journal, the Journal of Vegetation Science. Both journals publish papers on the ecology of a single species only if it plays a key role in structuring plant communities.