Long-time record and continuous high resolution gross primary productivity estimates at continental scales.
Abstract
Gross Primary Productivity (GPP) is an essential variable quantifying the total ecosystem gross carbon uptake through photosynthesis. Due to its importance in Earth monitoring and climate change research, along with many applications in ecology and biodiversity, GPP is routinely estimated at global scales using different operational algorithms combining remotely-sensed data from medium spatial resolution sensors and ancillary meteorological (e.g. air temperature, soil moisture, solar irradiation) information. However, there is a lack of operational global scale GPP products at finer (30m) spatial resolution needed to better resolve plant community scale dynamics. The use of finer resolution satellite information at broad scales to estimate GPP is constrained by extremely high storage and computational needs. These activities also require the creation of consistent mosaics and long time series, which are often plagued by data record gaps due to cloud contamination, present radiometric differences across sensors, scene overlaps, and their inherent sensor noise. To alleviate the above mentioned problems, we have fused similar spectral data from Landsat and MODIS using the HIghly Scalable Temporal Adaptive Reflectance Fusion Model (HISTARFM) algorithm: the method produces monthly gap free high resolution (30 m) surface reflectance data at continental scales with associated well-calibrated data uncertainties. We then utilize this improved Landsat-resolution composited monthly product with daily meteorological information, and in situ eddy covariance GPP estimates to create accurate high spatial resolution GPP estimates over Europe, the contiguous US, and the Amazon basin using both empirical and machine learning approaches. In this work, we capitalize on advancements in cloud computing and parallel processing technologies such as Google Earth Engine to produce long time series of continuous GPP estimates of enormous application potential at higher levels of spatial detail across very broad spatiotemporal scales. The methodology enables more precise carbon studies and understanding of land-atmosphere fluxes, as well as the means for deriving other carbon, heat and energy fluxes at an unprecedented spatio-temporal resolution.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2021
- Bibcode:
- 2021AGUFM.B55E1252M