Interannual variations and trends in global land surface phenology derived from enhanced vegetation index during 1982–2010

Abstract

Land surface phenology is widely retrieved from satellite observations at regional and global scales, and its long-term record has been demonstrated to be a valuable tool for reconstructing past climate variations, monitoring the dynamics of terrestrial ecosystems in response to climate impacts, and predicting biological responses to future climate scenarios. This study detected global land surface phenology from the advanced very high resolution radiometer (AVHRR) and the Moderate Resolution Imaging Spectroradiometer (MODIS) data from 1982 to 2010. Based on daily enhanced vegetation index at a spatial resolution of 0.05 degrees, we simulated the seasonal vegetative trajectory for each individual pixel using piecewise logistic models, which was then used to detect the onset of greenness increase (OGI) and the length of vegetation growing season (GSL). Further, both overall interannual variations and pixel-based trends were examined across Koeppen's climate regions for the periods of 1982-1999 and 2000-2010, respectively. The results show that OGI and GSL varied considerably during 1982-2010 across the globe. Generally, the interannual variation could be more than a month in precipitation-controlled tropical and dry climates while it was mainly less than 15 days in temperature-controlled temperate, cold, and polar climates. OGI, overall, shifted early, and GSL was prolonged from 1982 to 2010 in most climate regions in North America and Asia while the consistently significant trends only occurred in cold climate and polar climate in North America. The overall trends in Europe were generally insignificant. Over South America, late OGI was consistent (particularly from 1982 to 1999) while either positive or negative GSL trends in a climate region were mostly reversed between the periods of 1982-1999 and 2000-2010. In the Northern Hemisphere of Africa, OGI trends were mostly insignificant, but prolonged GSL was evident over individual climate regions during the last 3 decades. OGI mainly showed late trends in the Southern Hemisphere of Africa while GSL was reversed from reduced GSL trends (1982-1999) to prolonged trends (2000-2010). In Australia, GSL exhibited considerable interannual variation, but the consistent trend lacked presence in most regions. Finally, the proportion of pixels with significant trends was less than 1 % in most of climate regions although it could be as large as 10 %.

Characterising the Land Surface Phenology of Europe Using Decadal MERIS Data

“…Figure 1d). Atzberger, Klisch, Mattiuzzi and Vuolo [41], Jeong, Ho, Gim and Brown [39], and Zhang, Tan and Yu [42], and Atzberger, Klisch, Mattiuzzi and Vuolo [41] found similar dates and patterns of variability for all phenophases. However, it should be noted that some differences can be attributed to the inclusion of crops in the present map.…”

“…Although, several studies of both LSP and ground phenology have been published in recent years focused on Europe or the Northern Hemisphere, most of them are related to phenology changes or trends across time, rather than the characterisation of phenological event dates. Apart from the scarce number of studies, there are some technical reasons that make a comparison challenging [42]. The determination of the timing of vegetation phenophases varies greatly according to different factors, including: (i) the methods applied to process the time-series (i.e., filtering or smoothing) [47,52]; (ii) the temporal base information estimate [16]; and (iii) the sensors and vegetation indices used [41,53].…”

Using Moderate-Resolution Temporal NDVI Profiles for High-Resolution Crop Mapping in Years of Absent Ground Reference Data: A Case Study of Bole and Manas Counties in Xinjiang, China

“…Another challenge for crop identification using reference knowledge is the inter-annual phenological changes of the crops caused by the variation of temperature and precipitation among multiple years [43,[59][60][61]. The temporal patterns of NDVI ( Figure 10) have shown inter-annual variation in Bole and Manas.…”

Evaluation of the Plant Phenology Index (PPI), NDVI and EVI for Start-of-Season Trend Analysis of the Northern Hemisphere Boreal Zone

“…The results of VI SOS trends will be shown as two maps (TS trends and TS trends of significant changes) as relatively low amount of significant trends based on per-pixel trend analysis are observed caused by a relatively short period of analysis (2000-2014) and coarse resolution of the earth observation (EO) time-series [70]. …”

“…This is likely due to a short time-period of investigation (as compared to Global Inventory Modeling and mapping Studies (GIMMS) based studies starting in 1982), reduced statistical significance for phenology trends in recent decades [24], outliers in the noisy VI time-series of the high latitudes, and per-pixel based trend estimation method [48,70,83]. A better estimation of significant trends could have been achieved by applying panel analysis [84] instead of pixel-based trend analysis, which is reported to give non or minor significant trends in the high latitudes [85,86].…”