Characterization of the main land processes occurring in Europe (2000-2018) through a MODIS NDVI seasonal parameter-based procedure.

The identification and recognition of the land processes are of vital importance for a proper management of the ecosystem functions and services. However, on-ground land uses/land covers (LULC) characterization is a time-consuming task, often limited to small land areas, which can be solved using remote sensing technologies. The objective of this work is to investigate how the different MODIS NDVI seasonal parameters responded to the main land processes observed in Europe in the 2000-2018 period; characterizing their temporal trend; and evaluating which one reflected better each specific land process. NDVI time-series were evaluated using TIMESAT software, which extracted eight seasonality parameters: amplitude, base value, length of season, maximum value, left and right derivative values and small and large integrated values. These parameters were correlated with the LULC changes derived from COoRdination of INformation on the Environment Land Cover (CLC) for assessing which parameter better characterized each land process. The temporal evolution of the maximum seasonal NDVI was the parameter that better characterized the occurrence of most of the land processes evaluated (afforestation, agriculturalization, degradation, land abandonment, land restoration, urbanization; R2 from 0.67-0.97). Large integrated value also presented significant relationships but they were restricted to two of the three evaluated periods. On the contrary, land processes involving CLC categories with similar NDVI patterns were not well captured with the proposed methodology. These results evidenced that this methodology could be combined with other classification methods for improving LULC identification accuracy or for identifying LULC processes in locations where no LULC maps are available. Such information can be used by policy-makers to draw LULC management actions associated with sustainable development goals. This is especially relevant for areas where food security is at stake and where terrestrial ecosystems are threatened by severe biodiversity loss.

Crop load affects maximum daily trunk shrinkage of plum trees.

We studied the effects of low fruit load (3-4 fruits cm(-2) of trunk cross-sectional area (TCSA), and high fruit load (6-7 fruits cm(-2) TCSA) on maximum daily trunk shrinkage (MDS) and trunk growth rates (TGR) over two seasons in plum (Prunus salicina Lindell) trees receiving full irrigation or deficit irrigation. Seasonal changes in MDS and TGR were compared with those in midday stem water potential (Psi(s)) and leaf stomatal conductance (g (s)). Crop load increased g (s) in fully irrigated trees approaching harvest. Although crop load did not affect plant water status in either watering regime, there were considerable differences in both MDS and TGR as a function of crop load. Compared with low-cropping [corrected] trees, MDS was 34% higher and TGR was 48% lower in high-cropping [corrected] trees. The differential responses of MDS and Psi(s) to crop load were a consequence of a higher MDS for a given Psi(s) in the high-cropping trees compared with the low-cropping trees. There was a linear increase in MDS with crop load, with a slope of 15.2 microm MPa(-1) per unit increment of crop load. In the fully irrigated trees, day-to-day variations in MDS were related to evaporative demand; however, the slope of the relationship between MDS and evaporative demand increased with crop load, indicating that different reference equations must be used to adjust for tree crop load when using MDS to determine plant water status and irrigation requirements.

Usefulness of diurnal trunk shrinkage as a water stress indicator in plum trees.

We compared seasonal changes in maximum diurnal trunk shrinkage (MDS) with seasonal changes in midday stem water potential (Psi(s)) over three years in plum trees grown in differing drip-irrigated regimes. In well-irrigated trees, day-to-day variations in Psi(s) and MDS were related to evaporative demand. Reference equations were obtained to predict MDS and Psi(s) values for well-irrigated trees as functions of environmental conditions. A decrease in plant water status toward the end of the growing season occurred even in the well-irrigated trees, probably reflecting a reduced volume of soil wetted by the drip irrigation system. Thus, for the prediction of Psi(s), different reference equations are required for the fruit-growth and after-harvest phenological periods. A seasonal change in the relationship between MDS and Psi(s) was observed, which compensated for the decrease in plant water status such that well-irrigated trees had similar MDS values during both the fruit-growth and after-harvest periods. The influence of tree size on the relationship between MDS and Psi(s) was also investigated. For tree trunk diameters ranging between 8 and 13 cm, MDS increased 13% for each cm of increase in trunk diameter, as a result of the thicker phloem tissues of the larger trees. This finding may allow extrapolation of Psi(s) predictions based on empirical relationships with MDS to plum trees of different sizes.