Photosynthesis trend in terrestrial biosphere using MODIS GPP time series data during 2000-2015

Changing trends in ecosystem photosynthesis can be quantified using satellite observations of Normalized Difference Vegetation Index (NDVI), Fraction of Photosynthetically Active Radiation (fPAR) and Gross Primary Production (GPP). However, the estimation of trends from NDVI, fPAR and GPP time series differs substantially depending on analyzed satellite dataset, the corresponding spatiotemporal resolution, and the applied statistical method. The global terrestrial GPP described as the total amount of carbon dioxide assimilated to the terrestrial biosphere by vegetation in photosynthesis. In other words, GPP is an essential flux of the net ecosystem exchange of CO2 between the atmosphere and terrestrial ecosystems. Therefore, GPP plays a key role in the global and terrestrial carbon cycle. The utilized data in this research are NASA Moderate Resolution Imaging Spectroradiometer (MODIS) Gross Primary Production and the Climatic Research Unit (CRU) meteorological data station. The MODIS photosynthesis model is based on the light use efficiency logic for calculating GPP. In this research, GPP dataset based on satellite observations and meteorological data has been used to estimate photosynthesis trend at a spatial resolution of 0.5-degree grid cell in terrestrial ecosystems from 2000 to 2015. Satellite remote sensing can provide continuous, repetitive, and consistent observations of dynamic changes in terrestrial ecosystem structure and function over large areas; it has become a more and more important tool for monitoring land surface properties.The objective of this research is to assess the trends of GPP using Mann-Kendall proxies at 90% confidence level and identify their key driving factors. This test enables the investigation of long-term GPP tendencies, without assuming that a given dataset follows a normal distribution. The Mann-Kendall test could apply to annual, seasonal, and monthly time series data. Generally, time series can be decomposed in a trend, seasonal, and remainder component. In time-series data, seasonality is the existence of variations that happen at particular regular intervals less than a year. Seasonal fluctuations may be caused by various causes, such as weather and consists of periodic, repetitive, and generally regular and predictable patterns in the time series. Seasonal fluctuation is an average that can be used to compare an actual observation relative to what it would be if there were no seasonal variation. After that, the spatial distribution of the linear regression of the GPP and meteorological data (temperature and precipitation) was calculated for each grid cell in the terrestrial biosphere for 2000~2015. Linear regression analyses are models that involve one independent variable (e.g. temperature or precipitation) and one dependent variable (GPP). Although earlier studies were carried out at the global or regional scales, these results cannot be easily matched with this investigation.  According to the results, despite the GPP fluctuations, the dominant trend, waiving the disturbance processes, is no-trend. The positive trends can be found in the southern parts of Africa, tropical regions in Asia and America which show increasing trends in GPP. The spatial patterns of the climatic controls on the annual variability of GPP is consistent with previous studies. The results showed that, in the high latitudes, temperature is clearly the dominant and limiting driver on GPP/photosynthesis. Stronger correlations between GPP and temperature than precipitation were observed.