Publications

Drying and rewetting (D/W) of soil have significant impacts on soil organic matter (SOM) turnover. We hypothesised that frequent D/W cycles would release t

A disturbance due to selective logging causes the release of carbon from forest soils. Here, we examine this effect under Nothofagus stands in Andisols of southern Chile (39°38′S, 72°02′W, >4000mmyr−1). We postulate that soil carbon destabilization caused by forest management is counteracted by two contrasting carbon stabilization mechanisms linked to pedogenic processes: (1) the formation of allophane-imogolite-like materials (i.e., short-range ordered aluminosilicate particles with a specific Al/Si molar ratio, for example in the present study 3.8) and (2) formation of Al- and Fe-carbon complexes. The aim of this study was to compare the effects of selective logging on the organic carbon storage of forest Andisols. We estimated that carbon storage occurred in the uppermost 1 m of soils under evergreen (EGF) and secondary growth mixed deciduous forests (DMF). The selective dissolution method was used to estimate the mechanisms of soil carbon storage. The soil carbon storage under the unmanaged EGF did not differ significantly from that of the stands measured 10 or 50years after selective logging. Likewise, the soil carbon storage under DMF was similar to that under stands measured 2, 6 or 50 years after selective logging. Overall, soil carbon storage was 24% lower under DMF than under EGF, which was attributed to the faster decomposition of organic matter produced by deciduous forests. The soil pH of the Ah-horizon was significantly lower (4.5–5.1) than that of the subsoil (5.5–6.0) due to the leaching of cations, leading to the development of less allophanic properties (i.e., an aluandic horizon). This effect increased the formation of organo-mineral complexes and reduced the content of allophane-imogolite-like precursors (e.g., non-crystalline Si and Al). As a result, an inverse relationship was found between metals (Al, Fe) complexed with carbon and the content of non-crystalline Si and Al, both of which were within single horizons and the entire soil profile in general. We conclude that disturbances caused by forest management did not alter carbon storage in these Andisols, which can be explained by (1) the potential mechanisms of carbon stabilization studied here and (2) humid climate, which protected the bare horizon from irreversible drying after the management of these ecosystems, thus preventing potential carbon loss.

In crop models, yield depends on complex interactions between biomass production, crop development and assimilate partitioning, particularly with crop species exhibiting indeterminate patterns of development and submitted to limiting soil resources. The present study aimed at: (i) evaluating the adaptability of the STICS model based on the boxcartrain approach for the simulation of the fruit development and growth of tomato cultivars with different growth habits and (ii) evaluating the hypothesis that both carbon and nitrogen limitations alter fruit setting. Four experiments were carried out to test the model, one on a vine type tomato cultivar grown in a greenhouse and three on bush type tomato cultivars grown in open field, at the experimental station Panguilemo that belongs to Talca University in Chile. Total dry matter, nitrogen uptake, leaf area index and fruit number were satisfactorily simulated (model efficiency (EF)>82%), whereas the fruit dry matter production was predicted with a lower precision (EF=63%). The simulations showed that the indeterminate growth features incorporated in the STICS model were appropriate to simulate the crop growth and development for both the vine and bush types of tomato cultivars. Introducing a stress nitrogen index in the algorithm to calculate fruit set improved the quality of prediction of the fruit number.

Soil bacterial communities play a critical role in shaping soil stability and formation, exhibiting a dynamic interaction with local climate and soil depth. We employed an innovative DNA separation method to characterize microbial assemblages in low-biomass environments such as deserts and distinguish between intracellular DNA (iDNA) and extracellular DNA (eDNA) in soils. This approach, combined with analyses of physicochemical properties and co-occurrence networks, investigated soil bacterial communities across four sites representing diverse climatic gradients (i.e., arid, semi-arid, Mediterranean, and humid) along the Chilean Coastal Cordillera. The separation method yielded a distinctive unimodal pattern in the iDNA pool alpha diversity, increasing from arid to semi-arid climates and decreasing in humid environments, highlighting the rapid feedback of the iDNA community to increasing soil moisture. In the arid region, harsh surface conditions restrict bacterial growth, leading to peak iDNA abundance and diversity occurring in slightly deeper layers than the other sites. Our findings confirmed the association between specialist bacteria and ecosystem-functional traits. We observed transitions from Halomonas and Delftia, resistant to extreme arid environments, to Class AD3 and the genus Bradyrhizobium, associated with plants and organic matter in humid environments. The distance-based redundancy analysis (dbRDA) analysis revealed that soil pH and moisture were the key parameters that influenced bacterial community variation. The eDNA community correlated slightly better with the environment than the iDNA community, whereas the iDNA community was more sensitive to changes in soil physicochemical parameters. Soil depth was found to influence the iDNA community significantly but not the eDNA community, which might be related to depth-related metabolic activity. Our investigation into iDNA communities uncovered deterministic community assembly and distinct co-occurrence modules correlated with unique bacterial taxa, thereby showing connections with sites and key environmental factors. The study additionally revealed the effects of climatic gradients and soil depth on living and dead bacterial communities, emphasizing the need to distinguish between iDNA and eDNA pools.

<p>Besides nitrogen, phosphorus (P) is the major limiting nutrient of terrestrial primary productivity, with major P stocks being bound in soils. Stocks, speciation, and bioavailability of soil P differ among ecosystem types and with rock weathering status, which are both driven by climatic conditions. Microorganisms and plants have developed a range of strategies to mobilize P from organic and inorganic sources, e.g. expression of extracellular phosphatases and excretion of low-molecular-weight organic acids (LMWOA). However, the impact of precipitation, vegetation type, and soil P speciation on plant P acquisition strategies is not well understood, yet.</p><p>A semi-desert-to-humid-temperate-rainforest ecosystem sequence was investigated. Soil samples were taken from three sampling sites, all developed on granodiorite, comprising a precipitation gradient (66 mm a<sup>-1</sup> to 1469 mm a<sup>-1</sup>) along the Chilean Coastal Cordillera. Small-scale gradients (mm) from single roots to bulk soil in three depths were sampled to examine changes in P speciation, enabling the identification of local P depletion by plant roots and differences in P-speciation between rhizosphere and non-rhizosphere soil. Phosphorus speciation was examined by X-ray absorption near edge structure analysis. LMWOA as biotic weathering agents, and acid phosphatase kinetics as proxy for organic P recycling, were quantified. The aim was to disentangle the impact and functions of roots and associated microorganisms on driving agents of P-cycling.</p><p>Rhizosphere P speciation in soil changed considerably along the precipitation gradient from mainly primary P minerals in the semi-desert ecosystem to a dominance of organic P species in the humid-temperate rainforest. Contents of organically bound P were higher in root proximity compared to bulk soil in the humid-temperate rainforest soils (320 mg kg<sup>-1 </sup>and 70 mg kg<sup>-1</sup>, respectively) and in the topsoil of the Mediterranean woodland ecosystem (134 mg kg<sup>-1 </sup> and 62 mg kg<sup>-1</sup>, respectively).  In contrast, the rhizosphere soil was depleted in sesquioxide-adsorbed P in comparison to root-free bulk soil.</p><p>The content of LMWOA was correlated with inorganic P in soils of the semi-desert ecosystem, indicating intensive LMWOA exudation for biogenic P weathering of primary and secondary minerals. Under temperate rainforest LMWOA content, phosphatase activity, and microbial biomass carbon exhibited a negative correlation with secondary inorganic P forms but were positively linked to organic P species. We therefore conclude that P nutrition in this ecosystem relies less on weathering of P bearing minerals by LMWOA but is mainly based on organic P sources.</p><p>In terms of process understanding, these findings clearly show that LMWOA fundamentally change their role in the rhizosphere depending on the nutrient acquisition strategy of the respective ecosystem, which is affected by mean annual precipitation. While LMWOA facilitate biogenic weathering of P bearing minerals in the semi-desert, they mainly contribute to P recycling in the humid-temperate rainforest by preventing its precipitation and sorption. We conclude that P acquisition and cycling depends on the nutritional constrains of the given ecosystem: from biological weathering of inorganic P forms in the semi-desert driven by LMWOA and plant uptake to intensive P recycling from organic forms in the humid-temperate rainforest.</p>