Search

Extreme wet events in central Australia triggered large vegetation responses that contributed greatly to large global land carbon sink anomalies. There remain significant uncertainties on the extent to which these events over dryland vegetation can be monitored and assessed with satellite data. In this study, we investigated the vegetation responses of the major Australian semiarid biomes to two extreme wet events utilizing multi-satellite observations of (1) solar-induced chlorophyll fluorescence (SIF), as a proxy for photosynthetic activity and (2) the enhanced vegetation index (EVI), as a measure of canopy chlorophyll or greenness. We related these satellite observations with gross primary productivity (GPP) estimated from eddy covariance tower sites, as a performance benchmark. The C3-dominated Mulga woodland was the most responsive biome to both wet pulses and exhibited the highest sensitivity to soil moisture. The C4-dominated Hummock grassland was more responsive to the 2011 "big wet" event, relative to the later 2016-2017 wet pulse. EVI swiftly responded to the extreme wet events and showed markedly amplified seasonal amplitude, however, there was a time lag as compared with SIF during the post-wet period, presumably due to the relatively slower chlorophyll degradation in contrast with declines in photosynthetic activity. Despite a robust linear SIF-GPP relationship (r2 ranging from 0.59 to 0.85), the spatially coarse SIF derived from the Global Ozone Monitoring Experiment-2 (GOME-2) yielded high retrieval noise over the xeric biomes, hindering its capacity to capture thoroughly the dryland vegetation dynamics in central Australia. Our study highlights that synchronous satellite observations of greenness and fluorescence can potentially offer an improved understanding of dryland vegetation dynamics and can advance our ability to detect ecosystem alterations under future changing climates.

A comprehensive understanding of the effects of agricultural management on climate–crop interactions has yet to emerge. Using a novel wavelet–statistics conjunction approach, we analysed the synchronisation amongst fluxes (net ecosystem exchange NEE, evapotranspiration and sensible heat flux) and seven environmental factors (e.g., air temperature, soil water content) on 19 farm sites across Australia and New Zealand. Irrigation and fertilisation practices improved positive coupling between net ecosystem productivity (NEP = −NEE) and evapotranspiration, as hypothesised. Highly intense management tended to protect against heat stress, especially for irrigated crops in dry climates. By contrast, stress avoidance in the vegetation of tropical and hot desert climates was identified by reverse coupling between NEP and sensible heat flux (i.e., increases in NEP were synchronised with decreases in sensible heat flux). Some environmental factors were found to be under management control, whereas others were fixed as constraints at a given location. Irrigated crops in dry climates (e.g., maize, almonds) showed high predictability of fluxes given only knowledge of fluctuations in climate (R2 > 0.78), and fluxes were nearly as predictable across strongly energy- or water-limited environments (0.60 < R2 < 0.89). However, wavelet regression of environmental conditions on fluxes showed much smaller predictability in response to precipitation pulses (0.15 < R2 < 0.55), where mowing or grazing affected crop phenology (0.28 < R2 < 0.59), and where water and energy limitations were balanced (0.7 < net radiation ∕ precipitation < 1.3; 0.27 < R2 < 0.36). By incorporating a temporal component to regression, wavelet–statistics conjunction provides an important step forward for understanding direct ecosystem responses to environmental change, for modelling that understanding, and for quantifying nonstationary, nonlinear processes such as precipitation pulses, which have previously defied quantitative analysis.

Projecting the likely change of potential evapotranspiration (ETp) under future climate scenarios is crucial for quantifying the impacts of climate change on the hydrologic cycle and aridity conditions. However, there are different sources of uncertainty in projecting future ETp that may arise from global climate models (GCMs), emission scenarios, and multiple ETp models used. In this study, we developed three random forest-based (RF-based) ETp models with solar radiation and air temperature at eight climatic stations in southeastern Australia. With Penman model as the benchmark, their performance was firstly compared with four empirical models (Jensen-Haise, Makkink, Abtew, and Hargreaves), which requires the same meteorological inputs. In general, the RF-based ETp models showed better performance in ETp estimates across all stations, with coefficients of determination (R2) ranging from 0.68 to 0.92, root mean square errors (RMSE) ranging from 0.58 mm day−1 to 1.46 mm day−1, and relative mean bias errors (rMBE) ranging from −16.10% to 9.73%. The RF-based and empirical models were then used to project future ETp for the eight stations based on statistically downscaled daily climatic data from 34 GCMs under two different representative concentration pathways (RCP4.5 and RCP8.5). All models indicated that ETp was likely to increase at the eight stations. The ensemble increases of mean ETp across eight stations ranged from 33 mm year−1 (2.1%, 2040s) to 129 mm year−1 (9.2%, 2090s) and from 43 mm year−1 (2.8%, 2040s) to 248 mm year−1 (17.6%, 2090s) under RCP4.5 and under RCP8.5, respectively. In addition, we also quantified uncertainties in ETp projections originating from ETp models, GCMs, RCPs, and their combined effects using the analysis of variance (ANOVA) method. Results showed that RCP-related uncertainty contributed the most to projected ETp uncertainty (around 40% for most stations) while GCM-related and ETp model-related uncertainties accounted for roughly equal amounts of projected ETp uncertainty (10%–30%). This study demonstrated the better performance of RF-based ETp models. It is advisable to use multiple ETp models driven by various GCMs under different RCPs to produce reliable projections of future ETp.

Quantifying global photosynthesis remains a challenge due to a lack of accurate remote sensing proxies. Solar-induced chlorophyll fluorescence (SIF) has been shown to be a good indicator of photosynthetic activity across various spatial scales. However, a global and spatially challenging estimate of terrestrial gross primary production (GPP) based on satellite SIF remains unresolved due to the confounding effects of species-specific physical and physiological traits and external factors, such as canopy structure or photosynthetic pathway (C3 or C4). Here we analyze an ensemble of far-red SIF data from OCO-2 satellite and ground observations at multiple sites, using the spectral invariant theory to reduce the effects of canopy structure and to retrieve a structure-corrected total canopy SIF emission (SIFtotal). We find that the relationships between observed canopy-leaving SIF and ecosystem GPP vary significantly among biomes. In contrast, the relationships between SIFtotal and GPP converge around two unique models, one for C3 and one for C4 plants. We show that the two single empirical models can be used to globally scale satellite SIF observations to terrestrial GPP. We obtain an independent estimate of global terrestrial GPP of 129.56 ± 6.54 PgC/year for the 2015–2017 period, which is consistent with the state-of-the-art data- and process-oriented models. The new GPP product shows improved sensitivity to previously undetected ‘hotspots’ of productivity, being able to resolve the double-peak in GPP due to rotational cropping systems. We suggest that the direct scheme to estimate GPP presented here, which is based on satellite SIF, may open up new possibilities to resolve the dynamics of global terrestrial GPP across space and time.