Book/Dissertation / PhD Thesis

FZJ-2017-00934

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png On model and measurement uncertainty in predicting land surface carbon fluxes

Post, H. (Corresponding author)FZJ*

2016
Forschungszentrum Jülich GmbH Zentralbibliothek, Verlag Jülich
ISBN: 978-3-95806-190-3

Please use a persistent id in citations: http://hdl.handle.net/2128/13643

Abstract: The Net Ecosystem Exchange (NEE) of CO$_{2}$ between the land surface and the atmosphere refers to the difference of photosynthetic CO$_{2}$ uptake and CO$_{2}$ release via ecosystem respiration. NEE is an important indicator for the net carbon source or sink function of an ecosystem and a crucial variable for understanding and predicting feedback mechanisms between climate and ecosystem change. NEE is typically measured by micrometeorological methods like eddy covariance (EC). At continental or global scales, land surface models (LSMs) such as the Community Land Model (CLM) are commonly used to predict NEE and other fluxes by simulating the coupled carbon, nitrogen, water and energy cycle of the land surface. In order to support future decision making in climate politics and environmental planning, it is important to improve LSM carbon flux predictions at regional scales. A central goal of this PhD work was therefore to combine measured EC data and CLM to estimate NEE for the Rur catchment area. For the last decade, model-data fusion approaches like parameter estimation have increasingly been applied to reduce the uncertainty of carbon flux estimates, because both EC measurements and LSM predictions are uncertain. In order to use EC data in meaningful model-data fusion or LSM evaluation approaches, an estimate of the measurement uncertainty is required. Thus, in the first part of the thesis, the NEE measurement uncertainty was studied for one grassland site in Germany located in the Rur catchment. At present, many uncertainty estimation approaches exist, but none are generally accepted and applied. The classical two-tower approach, which is based on the standard deviations of the fluxes measured simultaneously at two nearby EC towers, is one of the most well-known approaches. It provides linear regression functions between the flux magnitude and the random error, which are commonly adopted by scientists for a fast estimation of the random error. In previous studies, the (classical) two-tower approach has yielded robust uncertainty estimates, but care must be taken to meet the often competing requirements of statistical independence (non-overlapping footprints) and ecosystem homogeneity when choosing an appropriate tower distance. Thus, an extension of the classical two-tower approach is proposed here that corrects systematic differences of the NEE fluxes measured synchronous at the two EC tower stations. The role of the tower distance was investigated with help of a roving station separated between 8 m and 34 km from a permanent EC grassland station. For evaluation, uncertainty estimates obtained from a different, raw-data based method were used as reference. The herein introduced correction for systematic flux differences applied to weather-filtered data substantially reduced the overestimation of the two-tower based NEE measurement uncertainty for all distances (except 8 m) by 79% (34 km distance) to 100% (95 m distance). Results indicated that the sensitivity of the two-tower approach to the tower distance was reduced, which enhances the applicability of the extended two-tower approach. [...]

Note: RWTH Aachen, Diss., 2016

---

Contributing Institute(s):

1. Agrosphäre (IBG-3)

Research Program(s):

1. 255 - Terrestrial Systems: From Observation to Prediction (POF3-255) (POF3-255)

Appears in the scientific report 2016

---

Database coverage:
;

---