The impact of indicator selection on assessment of global greening
Greening represents a significant increase in vegetation activity. Accurate quantification of global greening in peak growth is vital for quantifying changes in terrestrial productivity. Normalized Difference Vegetation Index (NDVI) is the remote sensing indicator most commonly applied for this purp...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2021-04-01
|
| Series: | GIScience & Remote Sensing |
| Subjects: | |
| Online Access: | http://dx.doi.org/10.1080/15481603.2021.1879494 |
| _version_ | 1827811736410914816 |
|---|---|
| author | Bingwen Qiu Xiongfei Yan Chongcheng Chen Zhenghong Tang Wenbin Wu Weiming Xu Zhiyuan Zhao Chao Yan Joe Berry Wenqing Huang Fangxin Chen |
| author_facet | Bingwen Qiu Xiongfei Yan Chongcheng Chen Zhenghong Tang Wenbin Wu Weiming Xu Zhiyuan Zhao Chao Yan Joe Berry Wenqing Huang Fangxin Chen |
| author_sort | Bingwen Qiu |
| collection | DOAJ |
| description | Greening represents a significant increase in vegetation activity. Accurate quantification of global greening in peak growth is vital for quantifying changes in terrestrial productivity. Normalized Difference Vegetation Index (NDVI) is the remote sensing indicator most commonly applied for this purpose. However, there is limited knowledge about how the applications of other improved or newly available products can impact global greening assessments. This study reports the first systematic investigation of the impact of vegetation indicator selection on global greening in peak growth. It examines the period from 2003 to 2017 using six indicators with a spatial resolution of 500 m: Near Infrared Reflectance of terrestrial vegetation (NIRv) and the coupled diagnostic biophysical modeled Gross Primary Production (PML-GPP), together with NDVI, Enhanced Vegetation Index with two bands (EVI2), Leaf Area Index (LAI) and Moderate Resolution Imaging Spectroradiometer (MODIS) GPP (MOD-GPP) calculated based on MODIS images. Vegetation trends were estimated using the Mann–Kendall test, the unanimous trends by six vegetation indicators were derived, and the concordance ratio in estimated trends by each pair of indicators was investigated among different biomes and their sensitivity to changes in climate were investigated. We found that the estimated greening and browning varied between 12–23% and 2–13% of the vegetated areas, respectively. There was more net greening estimated from NDVI and MOD-GPP (around 19%) compared to the other indicators (8–11%). The concordance ratio between EVI2 and NIRv was much higher (>94%) than other combinations (61–76%) at the global scale. The concordance ratio between one specific indicator and the other indicators gradually increased with its change magnitude. Unanimous results from these six indicators were exhibited in less than two-fifths (38.71%) of vegetated areas. Apart from the EVI2 and NIRv, the concordance ratio varied among different biomes: high in cropland (67–81%), but low in deciduous needle-leaf forest (51–62%) and very low in evergreen broadleaf forest between PML-GPP and others (37–49%). The diverse sensitivity to changes in climate contributed to the discrepancies: greening was more pronounced during warming conditions for the GPP products when compared with the greenness indices. At the global scale, the concordance ratio was higher during cooling conditions among all the indicators except between PML-GPP. Datasets supporing these findings are available online at https://github.com/FuzhouSIRC/Greening_6indicators.2 |
| first_indexed | 2024-03-11T23:08:49Z |
| format | Article |
| id | doaj.art-d51c3c53c50f460f91e36fff0c6b40f6 |
| institution | Directory Open Access Journal |
| issn | 1548-1603 1943-7226 |
| language | English |
| last_indexed | 2024-03-11T23:08:49Z |
| publishDate | 2021-04-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | GIScience & Remote Sensing |
| spelling | doaj.art-d51c3c53c50f460f91e36fff0c6b40f62023-09-21T12:34:16ZengTaylor & Francis GroupGIScience & Remote Sensing1548-16031943-72262021-04-0158337238510.1080/15481603.2021.18794941879494The impact of indicator selection on assessment of global greeningBingwen Qiu0Xiongfei Yan1Chongcheng Chen2Zhenghong Tang3Wenbin Wu4Weiming Xu5Zhiyuan Zhao6Chao Yan7Joe Berry8Wenqing Huang9Fangxin Chen10Fuzhou UniversityFuzhou UniversityFuzhou UniversityUniversity of Nebraska-LincolnKey Laboratory of Agricultural Remote Sensing (AGRIRS), Ministry of Agriculture and Rural Affairs/Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural SciencesFuzhou UniversityFuzhou UniversityFuzhou UniversityCarnegie Institution for ScienceFuzhou UniversityFuzhou UniversityGreening represents a significant increase in vegetation activity. Accurate quantification of global greening in peak growth is vital for quantifying changes in terrestrial productivity. Normalized Difference Vegetation Index (NDVI) is the remote sensing indicator most commonly applied for this purpose. However, there is limited knowledge about how the applications of other improved or newly available products can impact global greening assessments. This study reports the first systematic investigation of the impact of vegetation indicator selection on global greening in peak growth. It examines the period from 2003 to 2017 using six indicators with a spatial resolution of 500 m: Near Infrared Reflectance of terrestrial vegetation (NIRv) and the coupled diagnostic biophysical modeled Gross Primary Production (PML-GPP), together with NDVI, Enhanced Vegetation Index with two bands (EVI2), Leaf Area Index (LAI) and Moderate Resolution Imaging Spectroradiometer (MODIS) GPP (MOD-GPP) calculated based on MODIS images. Vegetation trends were estimated using the Mann–Kendall test, the unanimous trends by six vegetation indicators were derived, and the concordance ratio in estimated trends by each pair of indicators was investigated among different biomes and their sensitivity to changes in climate were investigated. We found that the estimated greening and browning varied between 12–23% and 2–13% of the vegetated areas, respectively. There was more net greening estimated from NDVI and MOD-GPP (around 19%) compared to the other indicators (8–11%). The concordance ratio between EVI2 and NIRv was much higher (>94%) than other combinations (61–76%) at the global scale. The concordance ratio between one specific indicator and the other indicators gradually increased with its change magnitude. Unanimous results from these six indicators were exhibited in less than two-fifths (38.71%) of vegetated areas. Apart from the EVI2 and NIRv, the concordance ratio varied among different biomes: high in cropland (67–81%), but low in deciduous needle-leaf forest (51–62%) and very low in evergreen broadleaf forest between PML-GPP and others (37–49%). The diverse sensitivity to changes in climate contributed to the discrepancies: greening was more pronounced during warming conditions for the GPP products when compared with the greenness indices. At the global scale, the concordance ratio was higher during cooling conditions among all the indicators except between PML-GPP. Datasets supporing these findings are available online at https://github.com/FuzhouSIRC/Greening_6indicators.2http://dx.doi.org/10.1080/15481603.2021.1879494global greeningvegetation indicatorsgppconcordance ratiodiscrepancy |
| spellingShingle | Bingwen Qiu Xiongfei Yan Chongcheng Chen Zhenghong Tang Wenbin Wu Weiming Xu Zhiyuan Zhao Chao Yan Joe Berry Wenqing Huang Fangxin Chen The impact of indicator selection on assessment of global greening GIScience & Remote Sensing global greening vegetation indicators gpp concordance ratio discrepancy |
| title | The impact of indicator selection on assessment of global greening |
| title_full | The impact of indicator selection on assessment of global greening |
| title_fullStr | The impact of indicator selection on assessment of global greening |
| title_full_unstemmed | The impact of indicator selection on assessment of global greening |
| title_short | The impact of indicator selection on assessment of global greening |
| title_sort | impact of indicator selection on assessment of global greening |
| topic | global greening vegetation indicators gpp concordance ratio discrepancy |
| url | http://dx.doi.org/10.1080/15481603.2021.1879494 |
| work_keys_str_mv | AT bingwenqiu theimpactofindicatorselectiononassessmentofglobalgreening AT xiongfeiyan theimpactofindicatorselectiononassessmentofglobalgreening AT chongchengchen theimpactofindicatorselectiononassessmentofglobalgreening AT zhenghongtang theimpactofindicatorselectiononassessmentofglobalgreening AT wenbinwu theimpactofindicatorselectiononassessmentofglobalgreening AT weimingxu theimpactofindicatorselectiononassessmentofglobalgreening AT zhiyuanzhao theimpactofindicatorselectiononassessmentofglobalgreening AT chaoyan theimpactofindicatorselectiononassessmentofglobalgreening AT joeberry theimpactofindicatorselectiononassessmentofglobalgreening AT wenqinghuang theimpactofindicatorselectiononassessmentofglobalgreening AT fangxinchen theimpactofindicatorselectiononassessmentofglobalgreening AT bingwenqiu impactofindicatorselectiononassessmentofglobalgreening AT xiongfeiyan impactofindicatorselectiononassessmentofglobalgreening AT chongchengchen impactofindicatorselectiononassessmentofglobalgreening AT zhenghongtang impactofindicatorselectiononassessmentofglobalgreening AT wenbinwu impactofindicatorselectiononassessmentofglobalgreening AT weimingxu impactofindicatorselectiononassessmentofglobalgreening AT zhiyuanzhao impactofindicatorselectiononassessmentofglobalgreening AT chaoyan impactofindicatorselectiononassessmentofglobalgreening AT joeberry impactofindicatorselectiononassessmentofglobalgreening AT wenqinghuang impactofindicatorselectiononassessmentofglobalgreening AT fangxinchen impactofindicatorselectiononassessmentofglobalgreening |