Decoupling silicate weathering from primary productivity – how ecosystems regulate nutrient uptake along a climate and vegetation gradient

In addition to the supply of primary minerals and water flow the presence and growth rate of land plants are thought to drive rock weathering. While doubtlessly plants and their associated below-ground microbiota possess the tools for considerable weathering work, the quantitative evaluation of their impact relative to the common abiogenic weathering 10 processes remains poorly known. Here we report on a strategy to decipher the relative impact of these two drivers. We did so by quantifying weathering rates and nutrient uptake along the “EarthShape” transect in the Chilean Coastal Cordillera where landscapes are subjected to a substantial north to south gradient in precipitation and vegetation growth, whereas rock type is granitoid throughout and tectonic process rates do not differ much along the gradient. We quantified the bio-available fraction of nutritive elements in regolith and we measured 87Sr/86Sr isotope ratios in the different compartments of the Earth’s Critical 15 Zone (bedrock, regolith, bio-available fraction in saprolite and soil, and vegetation) to identify the sources of mineral nutrients to plants. We thereby budgeted inventories, gains, and losses of nutritive elements in and out of these ecosystems, and quantified mineral nutrient recycling. We found that the weathering rates do not increase with precipitation from north to south along the climate gradient. Instead, the simultaneous increase in biomass growth rate is accommodated by faster nutrient recycling. The absence of an increase in weathering rate in spite of a five-fold increase in precipitation leads us to hypothesize 20 that the presence of plants can negatively impact weathering through inducing secondary-mineral formation and by fostering a microbial community that is adapted for nutrient-recycling rather than nutrient-acquisition through weathering.


Introduction
Ever since the emergence of land plants, their dependence on mineral-derived nutrients has impacted rock weathering. This impact results from mechanical weakening of rock through roots and microbial symbionts (e.g. mycorrhizal fungi; Blum et al., 25 2002;Brantley et al., 2017;Hasenmueller et al., 2017;Minyard et al., 2012;Quirk et al., 2014;van Schöll et al., 2007), and from a variety of biogeochemical processes that alter the weatherability of minerals. These mechanisms include root respiration releasing protons and CO2 which lowers the soil pH, exudation of organic ligands through roots hence increasing the solubility of nutrients by complexation, and the uptake, uplift, and recycling of pore fluids and nutrients from solution (e.g. Berner et al., https://doi.org/10.5194/bg-2020-69 Preprint. Discussion started: 9 March 2020 c Author(s) 2020. CC BY 4.0 License. biological cycle with the long-term, largely geological and hydrological driven cycle fostering nutrient inputs to and outputs from terrestrial ecosystems (Powers et al., 2015;Vitousek et al., 1998). Such models have recently been adapted by 65 geochemistry (Uhlig and von Blanckenburg, 2019). In these, the so-called "organic nutrient cycle" comprises a set of strategies for efficient nutrient re-utilization through microbial mineralization from plant litter and organic matter and entails rapid nutrient turnover. However, ecosystems in sloping landscapes permanently lose these organic-derived nutrients by plant-litter erosion (Heartsill Scalley et al., 2012;Scatena and Lugo, 1995) as well as litter and root decomposition and their subsequent export in solution (Baskaran et al., 2017;Chaudhuri et al., 2007;Moore et al., 2005;Silver and Miya, 2001). Atmospheric wet 70 and dry deposition serves as a potential replacement in tropical (e.g. Boy and Wilcke, 2008;Chadwick et al., 1999;Dosseto et al., 2012), and desert (Wang et al., 2014) ecosystems which are subject to slow erosion rates and high atmospheric input.
However, in most sloping landscapes, the dominant nutrient-replacing mechanism is provided by the "geogenic nutrient pathway" which counterbalances losses by a slow but steady release of nutrients through chemical weathering of rock (Buendía et al., 2010;Cleveland et al., 2013;Uhlig and von Blanckenburg, 2019). Even though these geogenic fluxes are minor 75 compared to the organic nutrient cycle they sustain ecosystem nutrition over the long-term (millennial) time scales. The importance of this geogenic nutrient source to sustain ecosystem nutrition with macronutrients (e.g. Ca, K, Mg, and P) over longer time scales has recently been demonstrated by several studies across a variety of biomes (Hahm et al., 2014;Schuessler et al., 2018;Uhlig et al., 2017;Uhlig and von Blanckenburg, 2019). Uhlig and von Blanckenburg (2019) found that long-term ecosystem nutrition is ensured by the adjustment of the nutrient re-utilization loop towards higher efficiency that compensates 80 for a lower nutrient supply from chemical weathering.
However, whether the geogenic nutrient pathway is sufficiently effective to prevent development of nutrient limitation over the millennial scale depends on the supply of fresh rock into the weathering zone, the bio-availability of the nutrients released, and whether plant roots and the associated mycorrhizal fungi can access them. Thus, any exploration of these links remains inconclusive without constraining the loci of nutrient source in regolith and its stoichiometry in comparison to plant demand. 85 In this study, we explored weathering, nutrient uptake, and nutrient recycling along the "EarthShape" (Oeser et al., 2018) climate and vegetation gradient in the Chilean Coastal Cordillera. The four sites range from ~26°S to ~38°S and while lithology (granitoid), tectonic uplift, and erosion rates (10 to 40 t km −2 yr −1 ; Schaller et al., 2018) are broadly similar, precipitation forms a gradient from 10 mm yr −1 in the north to 1100 mm yr −1 in the south (Ministerio de Obras Públicas, 2017). Similarly, net 90 primary productivity (NPP) increases from 30 to 520 gC m −2 yr −1 (Werner et al., 2018). Along this gradient we quantified the degree (using chemical analyses of rock and regolith, Oeser et al., 2018) and rates (using cosmogenic 10 Be, Schaller et al., 2018) of weathering and nutrient uptake (using NPP and the chemical composition of the major plant species at each site).
Sequential extraction on bulk regolith was used to identify the stoichiometry of the main plant-available elements in the regolith (sum of saprolite and soil) in addition to phosphorus (Brucker and Spohn, 2019). We further utilized radiogenic 87 Sr/ 86 Sr 95 isotope ratios in bulk rock, regolith, and plant samples along with those in the bio-available fraction to identify the sources of mineral nutrients. Thus, we were able to identify gains and losses of nutritive elements in and out of these ecosystems and to https://doi.org/10.5194/bg-2020-69 Preprint. Discussion started: 9 March 2020 c Author(s) 2020. CC BY 4.0 License.
Catchment-wide denudation rates inferred from cosmogenic nuclides (in situ 10 Be) are lowest in Pan de Azúcar (7.7 ± 0.7 t km −2 yr −1 ) and Santa Gracia (9.2 ± 0.8 t km −2 yr −1 ), highest in La Campana (200 ± 22 t km −2 yr −1 ), and intermediate in Nahuelbuta (27.4 ± 2.4 t km −2 yr −1 ) is lower (van Dongen et al., 2019). At the soil pit scale, Schaller et al. (2018) report a trend in total soil denudation rates, interpreted as soil production rates, from Pan de Azúcar (8 -11 t km −2 yr −1 ) to Santa Gracia (16  140 -22 t km −2 yr −1 ) with a peak in La Campana (54 -69 t km −2 yr −1 ) and 18 -48 t km −2 yr −1 in Nahuelbuta. The elevated catchment-wide rates at La Campana are attributed to faster erosion processes dominating the entire catchment that is characterized by higher hill slopes than in the other areas (e.g. mean slope 23° in La Campana, 9° in Nahuelbuta;van Dongen et al., 2019). They are thus towards the lower end of global cosmogenic nuclide-derived soil production rates (Dixon et al., 2012). The relative consistency of these rates along the climate gradient are ascribed to uniform tectonic forces acting in the 145 whole study area (e.g. Blanco-Chao et al., 2014;Melnik, 2016).
The architecture of the regolith profiles, their chemistry, mineralogy, and the physical properties of soils, saprolite, and the rocks beneath have been extensively described by Bernhard et al. (2018), Dal Bo et al. (2019, Oeser et al. (2018)  .
The regolith profiles in Pan de Azúcar are located between 330 and 340 m a.s.l. on steep (25 -40°; Table 1) hill slopes. The 150 soils on the north-and on the south-facing slope were classified by Bernhard et al. (2018) as Regosols with only shallow A and B horizons of ∼20 -30 cm thickness, lacking any kind of organic layer. In this area, the processes disintegrating rock and developing regolith are mainly physical weathering attributed to a combination of insolation-and salt weathering (Oeser et al., 2018). The regolith profiles in Santa Gracia are situated at almost 700 m a.s.l. on mild sloping hills (15 -25°; Table 1). The soils on the north-and on the soul-facing slope are a Leptosol and a Cambisol, respectively (Bernhard et al., 2018). Distinct 155 O-horizons are not apparent. The Ah horizons in both profiles reach depths of 10 cm and the transition from the mineral soil (Bw) into saprolite occurs at 25 -30 cm depth. Oeser et al. (2018) attribute this study sites' high degree of elemental depletion (i.e. τ, almost 60% loss of Ca and Na relative to bedrock; Fig. A1; Table S2) despite low precipitation to the low abundance of quartz and the high abundance of readily weatherable plagioclase and mafic minerals. The regolith profiles in La Campana, located at 730 m a.s.l. and on mild sloping hills (12 -23°), are classified as Cambisols. The O-horizon is ∼5 cm thick and is 160 followed by a Ah horizon, extending up to 40 cm depth (Bernhard et al., 2018). Here, the mineral-soil layer turns into saprolite at approximately 110 cm in both profiles (Table 1). The elemental depletion of for example Ca increases accordingly from https://doi.org/10.5194/bg-2020-69 Preprint. Discussion started: 9 March 2020 c Author(s) 2020. CC BY 4.0 License.
∼45% at the profiles' bottom towards ∼70% at their top and can be classified as depletion (north-facing) or depletion and enrichment profiles (south-facing, Fig. A1; Table S2; Brantley and Lebedeva, 2011), respectively. The regolith in Nahuelbuta are situated on gently sloping hills (∼15°) at about 1200 m a.s.l. (Table 1). Bernhard et al. (2018) have classified the soils on 165 the north-and south-facing slope as umbric Podsol and orthodystric Umbrisol, respectively. Here, the Ah horizons measure up to 50 cm (greater thickness on the south-facing slope) and are overlain by an organic horizon of 5.5 cm thickness. In the two regolith profiles, the soil-saprolite transition is at 100 and 120 cm depth, respectively. The coarse-grained saprolite disaggregates readily. These profiles are characterized by highly heterogeneous weathering patterns caused by the incorporation of the metamorphic basement at various parts (e.g. Oeser et al., 2018;Hervé, 1977). 170 A comprehensive summary of the eight regolith profiles' characteristics and major plant types is given in Table 1.

Sampling
Regolith samples were collected in a continuous sequence of depth increments from bottom to top. Increments amount to a thickness of 5 cm for the uppermost two samples, 10 cm for the 3rd sample from top, and increase to 20 cm thickness for the 175 4th sample onwards. To account for the dependence on solar radiation, two regolith profiles on adjacent hillslopes (north-and south-facing) were sampled at each study site.
The underlying unweathered bedrock has not been reached in either of the regolith profiles and the depth to bedrock remains unknown. Thus, bedrock samples were collected from nearby outcrops. This sample set comprises the 20 bedrock samples already reported in Oeser et al. (2018) and 15 additional bedrock samples (in total 12 in Pan de Azúcar,8 in Santa Gracia,10 180 in La Campana, and 5 in Nahuelbuta) from within the respective study sites.
Vegetation samples from representative shrubs and trees of each study site have been sampled. The sample set comprises stem-, twig-, and leaf-samples from mature plants of the prevailing species Nolana mollis (Pan de Azúcar), Asterasia sp., Cordia decandra, Cumulopuntia sphaerica, and Proustia cuneifolia (Santa Gracia), Aristeguietia salvia, Colliguaja odorifera, Cryptocarya alba, and Lithraea caustica (La Campana), Araucaria araucana, Nothofagus antarctica, and Chusquea coleu 185 (Nahuelbuta), respectively. These samples were either taken using an increment borer (stem samples) or plant scissors (leaf and twig samples) equipped with a telescopic arm to reach the higher parts of trees. In addition, the litter layer in La Campana and Nahuelbuta was sampled.

Chemical composition of regolith and bedrock 190
The concentration of major and trace elements in bedrock and regolith samples were determined using a X-Ray Fluorescence spectrometer (PANalytical AXIOS Advanced) at the section for "Inorganic and Isotope Geochemistry", GFZ German https://doi.org/10.5194/bg-2020-69 Preprint. Discussion started: 9 March 2020 c Author(s) 2020. CC BY 4.0 License.
Research Centre for Geosciences. A detailed description of the analytical protocols and sample preparation is given in Oeser et al. (2018).

Chemical composition of vegetation 195
Major and trace element concentrations of vegetation samples were determined using a Varian 720-ES axial ICP-OES at the Helmholtz Laboratory for the Geochemistry of the Earth Surface (HELGES), GFZ German Research Centre for Geosciences (von Blanckenburg et al., 2016) with relative uncertainties smaller than 10%. Prior to analysis, all samples were oven-dried at 120°C for 12 hrs. Subsequently, leaves were crushed and homogenized. About 0.5 g of leaf and 1 g of woody samples were digested in PFA vials using a microwave (MLS start) and ultra-pure concentrated acid mixtures comprising H2O2 and HNO3,200 HCl and HNO3, and HF. In some plant samples Si-bearing precipitates formed upon evaporation after digestion. These sample cakes were redissoled in a mixture of concentrated HF and HNO3 to ensure complete dissolution of Si prior to analysis. As some Si might have been lost by evaporation in this process, we have not included these plant samples (indicated by a * in Table S5) for the compilation of the plants' Si budget. With each sample batch, the international reference material NIST SRM 1515 Apple leaves and a procedural blank was processed. 205

Extraction of the bio-available fraction and its chemical analyses
The bio-available fraction of regolith samples was extracted using a sequential extraction procedure adapted from Arunachalam et al. (1996), He et al. (1995), andTessier et al. (1979). The sequential extraction was performed in parallel on two regolith aliquots, of which the supernatants were pooled together for analyses. About 2 g of dried and sieved (<2 mm) sample material were immersed in 14 ml either 18 MΩ Milli-Q H2O (water-soluble fraction) or in 1M NH4Oac (exchangeable 210 fraction), thus maintaining a sample to reactant ratio of about 1:7. Each extraction step was performed with mild agitation.
After each extraction, the mixture was centrifuged for 30 min at 4200 rpm and the supernatant was pipetted off. The remaining sample was then rinsed with 10 ml Milli-Q H2O and centrifuged again (4200 rpm, 30 min). This rinse solution was added to the supernatant. Subsequently, the supernatants were purified using a vacuum-driven filtration system (Millipore ® ; 0.2 μm acetate filter), evaporated to dryness, and redissolved with ultra-pure concentrated acid mixtures comprising H2O2, HNO3, and 215 HCl. With each sample batch, international reference materials (NIST SRM 2709a San Joaquin soil, CCRMP TILL-1) along with a procedural blank were processed.
The water-soluble fraction is comprised of elements contained in soil water in the form of free ions and ions which form complexes with soluble organic matter. It represents the most labile soil compartment and thus is most accessible to plants (e.g. He et al., 1995). This fraction was accessed by suspending the samples for 24 hrs in Milli-Q H2O at room temperature. 220 The exchangeable fraction constitutes of elements that form weak electrostatic bonds between the hydrated surfaces of phyllosilicates (i.e. clays and micas), oxyhydroxide minerals (e.g. boehmite, diaspore, goethite, lepidocrocite, ferrihydrite), and organic matter. This fraction was extracted by suspending the samples in a mechanical end over end shaker at room temperature in 1 M NH4OAc for 2 hrs at 60 rpm. Note that none of the further extraction steps described in Tessier et al. (1979) https://doi.org/10.5194/bg-2020-69 Preprint. Discussion started: 9 March 2020 c Author(s) 2020. CC BY 4.0 License.
has been applied to the regolith samples in this study as they are believed to make a negligible contribution to the bio-available 225 fraction.
The element concentrations of the water-soluble and exchangeable fraction were determined using a Varian 720-ES axial ICP-OES at HELGES, following the analytical procedures described in Schuessler et al. (2016) with relative uncertainties estimated at smaller than 10%.
The soil-P fractions were determined by Brucker and Spohn (2019). In this case, the bio-available fraction refers to the 230 inorganic products of the modified Hedley sequential P fractionation method of Tiessen and Moir (1993), specifically the water-extractable Pi and labile Pi which was extracted by using 0.5 M NaHCO3.

87 Sr/ 86 Sr isotope ratios
The radiogenic Sr isotope ratio was determined on bulk bedrock and regolith, the bio-available fractions of saprolite and soil, and on the different plant organs at each study site. 235 Upon sample digestion (bulk samples) or sequential extraction (bio-available fraction), Sr was separated from matrix elements using 200 μl Sr-Spec resin. Matrix elements were removed by elution with 2.5 ml 3 M and 2 ml 7.5 M HNO3. Subsequently, Sr was eluted with 4 ml of 18 Ω Milli-Q H2O. Any organic crown-ether which has been released from the Sr-spec resin was removed after evaporation and subsequent redissolution of the Sr fraction in 1 ml of a 1:1 mixture of concentrated H2O2 and HNO3. This mixture was cooked in a tightly closed beaker at 150°C for at least 12 hrs. Within each sample batch, a minimum 240 of one standard reference material and a procedural blank were processed. 87 Sr/ 86 Sr was measured in a 50-ng g −1 pure Sr solution in 0.3 M HNO3 using a multi collector inductively coupled plasma mass spectrometer (MC-ICP-MS, Thermo Neptune) in medium mass resolution. The MC-ICP-MS was equipped with an APEX-Q (ESI) desolvating nebulizer with uptake rates of 70 µl min -1 and a nickel sampler cone. Radiogenic Sr isotope ratios were determined over one block of 20 cycles with an integration time of 16 seconds each. The sequence of a sample run was 245 comprised of 10 to 12 blocks, where each block comprised a blank, four samples, and five SRM 987 which were not processed through chemistry. Blank correction of samples and reference material during the sequence was less than 0.4% of the sample signal. The intensities of the ion beams on the masses 82 Kr (L4), 83 Kr (L3), 84 Sr (L2), 85 Rb (L1), 86 Sr (central Faraday Cup), 87 Sr (H1) and 88 Sr (H2) were monitored using Faraday collectors equipped with 10 11 Ω resistors. Isobaric interference on the masses 84, 86, and 87 were corrected for with the Kr and Rb isotope ratios measured prior to the sequence run. To correct for 250 any natural and instrumental isotope fractionation, the measured 87 Sr/ 86 Sr ratio was normalized to the 88 Sr/ 86 Sr ratio of 8.375209 (Nier, 1938's value) by using an exponential law. Finally, the 87 Sr/ 86 Sr ratios were corrected for a session offset.

Parameterizing geogenic and biogenic element fluxes in a terrestrial ecosystem
The parameterization of the "geogenic nutrient pathway" and the "organic nutrient cycle" (Fig. 2) to characterize element fluxes into, within, and from the Critical Zone and its ecosystem components, including the derivation of the corresponding 255 equations has been thoroughly described in Uhlig and von Blanckenburg (2019). Because throughout this article we make use of this framework we only briefly summarize the metrics, which are shown in Table 2. Calculation and parameters used for these metrics are presented Appendix A.

Results
We structure the presentation of our results in the following sequence: (1) the element fluxes of the geogenic nutrient pathway; 260 (2) the availability of elements in regolith to plants; and (3) the plants' chemical composition along with the element fluxes that link the geogenic nutrient pathway to the organic nutrient cycle.
We focus the detailed presentation of these results on the two most important mineral nutrients to plants P and K, and further provide data on the plant-essential elements Ca, K, Mg, and Mn as well as plant-beneficial elements Al, Fe, Na, and Si (Marschner, 1983). In this presentation, we treat both Sr and Ca as plant-essential elements due to their similar (bio-) chemical 265 behavior (e.g. Blum et al., 2012;Faure and Mensing, 2005;Faure and Powell, 1972;Poszwa et al., 2002).

Degree of weathering and elemental gains and losses
The chemical depletion fraction (CDF; Table 2, Eq. 5) and elemental mass transfer coefficient (τ; Table 2, Eq. 6) disclose the total and the element-specific loss, respectively, of soluble elements relative to bedrock. Thus, both metrics quantify the degree 270 of weathering. The average CDF of the shallowest mineral soil (combined analysis of north-and south-facing profiles) in Pan de Azúcar, Santa Gracia, La Campana, and Nahuelbuta amounts to 0.03, 0.54, 0.50, and 0.25, respectively ( Fig. 3; Table S2).
At all four sites, the elemental losses ( Fig. A1; Table S2) can be attributed to a "kinetically limited weathering regime" . This means that the erosion rate is at a sufficient level to continuously replenish the nutrient stock held in weatherable minerals that transit vertically through the weathering profile. 275 Systematic differences in chemical depletion pattern (i.e. CDF and τ) that arose from different combinations of (micro-) climate and biota between north-and south-facing slopes are not discernible or, as is the case of La Campana, caused by anomalous high Zr concentrations throughout the entire north-facing profile. Moreover, we found that neither CDF nor τ systematically differ between Santa Gracia, La Campana, and Nahuelbuta.
A comprehensive presentation of these data can be found as Appendix in Fig. A1 and in the supplementary Table S2 (Oeser 280 and von Blanckenburg, 2020).

Elemental chemical weathering fluxes
W regolith X (Table 2, Eq. 3) quantifies the elemental release fluxes from rock and regolith by weathering and thus the weathering supply of nutrients to ecosystems. The weathering release for the most plant-essential rock-derived mineral nutrient phosphorus W regolith P in Pan de Azúcar, Santa Gracia, La Campana, and Nahuelbuta amounts to 1.3, 12, 19, and 11 mg m −2 yr −1 , 285 respectively, and those of the similarly plant-essential element potassium (W regolith K ) to 30, 80, 840, and 100 mg m −2 yr −1 (Fig. 4, Table 3). W regolith X of the plant-beneficial elements Al, Na, and Si along with Fe and Sr follow the trend described for K and P. The weathering supply fluxes of both P and K as well as the plant-beneficial elements in Santa Gracia and Nahuelbuta are thus in a similar range despite severe differences in MAP, NPP, and vegetation cover. W regolith Ca and W regolith Mg do deviate from this general pattern such that these elements' highest weathering-release fluxes are attained in Santa Gracia followed by 290 La Campana, Nahuelbuta, and Pan de Azúcar. These elevated weathering release fluxes of Ca and Mg in Santa Gracia are attributed to the bedrocks' initial mineralogy, and their high Ca and Mg concentration (Table S1).

Availability of mineral nutrients to ecosystems
The maximum amount of nutrients present can be assessed by determining their inventory in bulk regolith (I bulk X ; Table 2, Eq. 8). For most elements, apart from K and Si, the inventory of potentially available elements is by far greatest in Santa Gracia 295 (Table 4). I bulk X in the other three study sites are mostly at a similar level. Note that only a minor fraction of this bulk nutrient inventory is readily bio-available.
The element concentration in the bio-available fraction are some orders of magnitude below the concentrations in bulk regolith Table S3). Bio-available P in saprolite (I bio-av, sap P ) is virtually absent in Pan de Azúcar and amounts to 21, 39, and 23 g m −2 in Santa Gracia, La Campana, and Nahuelbuta, respectively (Table 4). I bio-av, sap K equals 253 in the northernmost, 300 and 23, 70, and 19 g m −2 in the other study sites further south, respectively. Bio-availability of the remaining mineral nutrients in saprolite is high in Santa Gracia and La Campana, and low in Pan de Azúcar and Nahuelbuta (Table 4). Note that I bio-av, sap K was calculated over the uppermost 1 m of saprolite, whereas in fact the zone of mineral nutrient extraction might extend deeper.
Bio-availability of most elements in soil, bar a few exceptions, increase from Pan de Azúcar to La Campana. However, despite featuring the thickest soils, bio-availability is lowest in Nahuelbuta. Exceptions from this general trend are P and K. I bio-av, soil P 305 amounts to 3.3 g m −2 in Pan de Azúcar and 22 g m −2 in Santa Gracia and is as high as 28 and 31 g m −2 in La Campana and Nahuelbuta, respectively (Table 4). I bio-av, soil K amounts to 53, 38, 90, and 38 g m −2 in Pan de Azúcar, Santa Gracia, La Campana, and Nahuelbuta, respectively. Thus, K is almost equally available to plants in the four study sites.

Plant element composition and nutrient-uptake fluxes
Average elemental concentrations in bulk plants decrease from Pan de Azúcar towards Nahuelbuta. The exception is P, whose 310 average concentration increases from 290 μg g −1 in Pan de Azúcar to 1400 μg g −1 in Nahuelbuta (Table 5). The average K concentration amounts to 6900, 6400, 12000, and 5400 μg g −1 along the EarthShape climate and vegetation gradient from north to south. Thus, in Pan de Azúcar, Santa Gracia, and Nahuelbuta, average K concentrations are in a similar range, whereas in La Campana, K concentration in plants is almost twice as high compared to the other three sites.
The nutrient-uptake fluxes of the two most important rock-derived mineral nutrients to plants, P and K, increase steadily from 315 north to south, such that U total P amounts to 5, 70, 170, and 350 mg m −2 yr −1 and U total K to 110, 500, 2000, and 1400 mg m −2 yr −1 in Pan de Azúcar, Santa Gracia, La Campana, and Nahuelbuta, respectively (Table 3).
U total X of the plant-essential elements Ca, K, Mg, Mn, P, and Sr exceed W regolith X up to several times. U total X and W regolith X are close to identical for Mg, Mn, and Sr in La Campana ( Fig. 4; Table 3). U total X of the remaining plant-beneficial elements are, apart from Fe and Na in Pan de Azúcar, always lower than their release by weathering. 320

87 Sr/ 86 Sr isotope ratios
Radiogenic Sr isotope ratios on bulk bedrock and regolith samples disclose mineral-weathering reactions and the incorporation of external sources into the regolith profiles. Moreover, 87 Sr/ 86 Sr in the bio-available fraction and plants reveal the plants' mineral nutrient sources.
In Pan de Azúcar, the 87 Sr/ 86 Sr ratio of bulk regolith ranges from 0.723 to 0.737 and is distinct from average bedrock (0.726 ± 325 0.002; Fig. 5, Table 6). The Sr isotope ratios differ significantly between the two regolith profiles which can be attributed to the varying degrees of atmospheric deposition ( 87 Sr/ 86 Srseaspray = 0.7092; Pearce et al., 2015). 87 Sr/ 86 Sr ratios in the bio-available fraction of saprolite and soil deviate by 0.02 from those of bulk bedrock and regolith but do not vary considerably between saprolite and soil, and the north-and south-facing slopes, respectively. Bulk plant samples yield 87 Sr/ 86 Sr ratios of 0.710 and are thus indistinguishable from the 87 Sr/ 86 Sr ratio in the bio-available fraction (0.710 ± 0.001; Fig. 5, Table 6). 330 In Santa Gracia, the 87 Sr/ 86 Sr ratios in both bedrock and the regolith profiles do not deviate significantly from each other  Table 6). 335 The bulk regolith 87 Sr/ 86 Sr ratio in La Campana ranges from 0.7051 in the N-facing to 0.7055 in the south-facing regolith profile. This ratio is a substantially lower than in bedrock (0.7061 ± 0.0003; Fig. 5, Table 6) which can be attributed to the loss of a mineral with a high 87 Sr/ 86 Sr isotope ratio (e.g. biotite) beneath the depth of the sampled regolith profiles. The radiogenic Sr composition of the bio-available fraction in saprolite and soil amounts to 0.7051 and 0.7053, respectively, and is within the range of bulk regolith. Here, the difference in the 87 Sr/ 86 Sr ratio between the north-and the south-facing regolith profile 340 increases from bulk regolith to the bio-available fraction in saprolite and is highest in the soil bio-available fraction. The average 87 Sr/ 86 Sr ratio in plants is 0.7059 and can be as high as 0.7063 in Cryptocaria alba (Table S7) and is thus slightly higher than in the soil and saprolite bio-available fraction. All these ratios are lower than in bulk bedrock.
In Nahuelbuta the radiogenic Sr isotope ratio in bedrock (0.716 ± 0.007) is in good agreement to those reported by Hervé et al. (1976) for the granitoid basement (0.717). However, the large spread among the bedrock samples is remarkable and denotes 345 to the high petrological and chemical variability of the Nahuelbuta mountain range (e.g. Hervé, 1977). Thus, 87 Sr/ 86 Sr in regolith is subject to a high variability as well (Fig. 5, Table 6 & S2). The 87 Sr/ 86 Sr ratios in both bio-available fractions in Nahuelbuta are restricted to a relatively narrow range in both regolith profiles, and amount to 0.711 ± 0.002 and are thus indistinguishable from the ratios in plants (Fig. 5, Table 6). Individual plants' radiogenic Sr signature are distinct from each other and reflect the slope's bio-available fraction they grow on. 350

The source of mineral nutrients
The radiogenic Sr composition of the bio-available fractions in saprolite and soil along with those of bulk plant serves as a proxy for the plants' nutrient sources. In all four study sites, plants take up their Sr from the bio-available fraction. In the arid site Pan de Azúcar this Sr pool is formed by Sr deposition from atmospheric sources (e.g. up to 93% seaspray contribution; 355 Table 6). In the mediterranean and humid sites of La Campana and Nahuelbuta, respectively, the bio-available Sr is supplied by release from rock and regolith through weathering. In semi-arid Santa Gracia, we found a combination of both sources (up to 43% seaspray contribution; Table 6). In La Campana we found evidence for a deep nutrient source (located between the bottom of the regolith profile and unweathered rock) in the elemental-depletion pattern (Fig. A1). Here, deep-rooting plants (e.g. Lithraea caustica; Canadell et al., 1996) bypass the bio-available fraction of saprolite and soil and take up Sr with a higher 360 proportion of radiogenic 87 Sr which has been released through biotite weathering beneath the regolith profiles.

Are nutrient sources setting plant stoichiometry?
We expand the analysis of the source and demand of mineral nutrients, which in the preceding section was based solely on Sr as plant-essential element, to all other plant-essential (P, K, Ca, Mg, Mn) and plant-beneficial (Al, Fe, Na, Si) elements. We note that the ecosystems in the EarthShape study sites are mostly N-limited (Stock et al., 2019), yet we consider the essential 365 mineral nutrients to be potentially co-limiting. To do so, we first evaluate whether "ecological stoichiometry" has been attained by the plants. Ecological stoichiometry suggests that, similar to ocean microbial biomass, an optimum C:N:P ratio is attained -the "Redfield ratio" (e.g. Redfield, 1934). However, it is less clear whether such stoichiometrical pattern is also valid for terrestrial ecosystems, and for other elements besides C, N, and P (Cleveland and Liptzin, 2007;Sardans et al., 2011). As pointed out by several authors (e.g. Elser et al., 2010;Geider and La Roche, 2002;Hillebrand et al., 2014;White et al., 2006), 370 photoautotrophs only exploit a weak stoichiometric homeostasis. Hence, the elemental composition in plants (i.e. ecological stoichiometry; Sterner and Elser, 2003) is mainly set by the nutrient-supplying reservoirs ('you are what you root in model'; Elser et al., 2010), and only to a second order by homeostasis. Evidence for this suggestion was provided by Uhlig and von Blanckenburg (2019) for a large number of mineral nutrients. In that study, it was shown that the plant-available fraction in saprolite and soil is well correlated with the ecological stoichiometry in two forested ecosystems and that deviations from that 375 trend may have been driven by demand.
As an evaluation of the hypothesis that the nutrient reservoir sets plant stoichiometry we normalized both the mineral nutrient concentrations in plants (Table 5) and those in the bio-available fraction in saprolite and soil (Table S3) by the most plantessential mineral nutrient P (Fig. 6). In this analysis, an element X that plots on the 1:1 line would fulfill the entirely hypothetical condition of being perfectly co-limited and that its relative composition is set by the relative composition in the 380 plant-available fraction (i.e. "optimum" stoichiometric range). Such pattern would be similar to an ocean Redfield ratio in which a C:N:P ratio is identical between seawater and marine micro-biota (Capek et al., 2018;Moore et al., 2013). Elements that plot beneath the 1:1 line are available in excess in the bio-available fraction relative to P. In contrast, elements that plot above this line would be enriched in plants by their homeostasis relative to P over the X:P ratio of the bio-available fraction.
Hence, this means that this element would be limiting plant growth rather than P. 385 Four main features characterize the ecosystems' stoichiometry along that climate transect: (1) Indeed, all elements plot close to the 1:1 line while the deviations from the line exhibit an only weak control by homeostasis. Thus, the release of elements by weathering and supply by seaspray (in Pan de Azúcar) is setting the composition of the bio-available fraction, which to a first order is setting plant stoichiometry.
(2) At all sites K is taken up from the bio-available fractions in preference over P.
According to the rationale of this diagram (Capek et al., 2018;Moore et al., 2013), K would be the limiting mineral nutrient 390 rather than P, as it appears to be in higher demand than P. Evidence against this possibility are plant P concentrations that increase along the gradient, hinting at possible P limitation. This might be caused by the P fraction in the studied 1-2m of regolith shifting from the plant-available apatite fraction to less available secondary mineral P from north to south (Brucker and Spohn, 2019). In contrast, concentrations of plant-available K are rather uniform (Table S3). (3) The plant-beneficial elements (Na, Fe, Al, Si) are available in surplus only in La Campana and Nahuelbuta. Thus, at these sites a sorbed pool of 395 these elements, which are initially released by weathering exists that is not strongly plant-utilized, whereas at the (semi-)arid sites plants make use of the full weathering release. (4) X:P ratios increasingly approach the "optimum" stoichiometric range with increasing NPP. They do this more pronounced in soil than in saprolite. So how do these gradients on stoichiometry develop together with NPP even though weathering rates are rather uniform between sites?

Nutrient recycling as buffering mechanism 400
The recycling of the most plant-essential mineral nutrients is the key mechanism enabling differences in NPP despite nearconstant weathering rates. As seen in Figure 4 elemental uptake rates U total X of the most plant-essential nutrients P, K, and Ca increase in spite of rather uniform release rates from Santa Gracia to Nahuelbuta by weathering W regolith X . This recycling is quantified by the recycling factor Rec 6 (Table 2, Eq. 7; Table 7; note that in this discussion we use the Rec X calculated for W regolith X from weathering only. In Table 7 and Fig. A4 we also show Rec X including atmospheric inputs). For example, K is 405 recycled 3 times through biomass after release by weathering at the arid site and 15 times at the humid site. P is recycled 4 times in the arid site, and 30 times in the humid site. Ca shows a similar, albeit weaker difference. Moreover, we found evidence that recycling of K through plants (Fig. 4) increases its availability in soil relative to saprolite over the entire climate gradient (Fig. 6).
Furthermore, with increasing recycling efficiency (Table 7) the nutrient pools in the bio-available fraction are increasingly 410 dominated by the pool of recycled nutrients, thus shifting the stoichiometry in the bio-available fraction with increasing proportion of recycling successively towards the stoichiometry in vegetation (Fig. 6). In other words, over the course of several recycling loops, the chemical composition of the bio-available fraction and biota eventually approaches a ratio close to the relative requirement of the ecosystem for the different nutrients (Vitousek et al., 1998), ultimately resulting in a virtually almost perfect stoichiometric yield (Sterner and Elser, 2003). P also obtains an increasingly biocycled speciation where the total 415 organic P fractions increase strongly along the gradient (Brucker and Spohn, 2019). In Nahuelbuta Ca, K, Mg, and Sr are taken up in excess over P. Here, organic-bound P is not as easily available, whereas the plant-availability of these elements might increase due to the higher potential for solubilization at the lower pH in soil (soil pH = 4 -4.5;Bernhard et al., 2018). Thus, the amplitude of recycling varies from nutrient to nutrient and site to site (Table 7). In the arid Pan de Azúcar, nutrients are primarily being recycled after photodegradation of shrubs (e.g. Gallo et al., 2006;Day et al., 2015). In the remainder sites 420 Rec 6 increases from Santa Gracia to Nahuelbuta and is highest for the plant-essential elements Ca, K, and P (Table 7). Thus, despite having the smallest nutrient inventory of bio-available nutrients but highest NPP of these sites, the ecosystem of Nahuelbuta can at least partially maintain its nutrient requirements through efficient nutrient recycling.
Particular conditions of plant nutrition prevail at the arid sites. In Pan de Azúcar and Santa Gracia elemental concentrations in plants are exceptionally high (bar P and K; Table 5), even though these sites' nutrient-availability through weathering and 425 water availability is generally low (Table 3 & S2). This elevated mineral-nutrient storage is typical for plants growing in infertile habitats. Through such intermediate storage, plants accumulate an internal nutrient pool that is available for homeostasis when growth conditions eventually improve during, for example, rare rain events (e.g. Chapin III, 1980;Chapin III et al., 2011;Vitousek et al., 1998). High amounts of Al and Na are being incorporated into the plants' tissues in the arid and semi-arid sites, elements that tend to be harmful and hinder plant growth when concentration is high (e.g. Delhaize and Ryan, 430 1995;Kronzucker and Britto, 2011). However, Al-toxicity is prevented in these plants by accumulation of similarly high amounts of Si that compensates the effects of Al (Liang et al. (2007). The Na concentrations in N. mollis in Pan de Azúcar are up to three orders of magnitude higher compared to the other study sites. This exceptional high Na concentration is typical of the metabolism of N. mollis which is known be covered with salt glands on their leaves, aiding to retrieve water by directly condensing moisture from unsaturated air Mooney et al., 1980). 435

How the organic and the geogenic nutrient pathway set the size of the bioavailable pool
Both our data on ecological stoichiometry and the radiogenic Sr isotopic composition suggest that the bio-available pool in saprolite and soil feeds plants. In none of our sites is the bio-available nutrient pool entirely depleted (Table 4). The concentrations of the most plant-essential and bio-available mineral nutrients K, Ca, and Mg in saprolite are highest in the arid site, lower in the semi-arid and mediterranean site, and lowest in the humid-temperate site. The element concentration in the bio-available fraction translates into the size of the inventory (note, however, that the inventory can in fact be larger than the 1 m inventory that we have used for its calculated to allow a comparison. This is suggested by the elevated 87 Sr/ 86 Sr ratios in plants at La Campana suggesting extraction of a pool beneath the bio-available upper saprolite). Regardless, given that neither these pool sizes nor the concentrations of the bio-available elements correlate with elemental release rates W regolith X , gradients in pool sizes cannot solely be explained by differences in weathering rate. Further, the differences in the pool size cannot be 445 explained by the lack of fresh primary minerals. All four sites represent kinetically limited weathering regimes (Oeser et al., 2018) such that primary minerals are permanently available for dissolution and in turn for replenishing the bio-available pool.
We thus consider next pedogenic properties as pool size-controlling factors.
The decrease of soil pH from 8 at the arid to 4 at humid site (Bernhard et al., 2018) might cause the decrease in the bioavailable divalent base cations Mg, Ca, and Sr. However, the decrease in pH could in turn be the result of the loss of these 450 elements and thus their pH buffering capacity. Another possibility are the higher concentrations of soil organic carbon (Bernhard et al., 2018) at the mediterranean and the humid sites compared to the arid sites. We regard this cause as unlikely as those elements like Al, Fe(III), and P that are readily complexed and thus lost as organic complexes are higher in the humid and mediterranean sites than the other two sites. The cation exchange capacity CECeff (Table 1; Bernhard et al., 2018) exceeds the element concentration in the bio-available fraction in the semi-arid (85 µmolC g -1 vs. 35 µmol g -1 ), mediterranean (80 µmolC 455 g -1 vs. 40 µmol g -1 ), and humid-temperate (30 µmolC g -1 vs. 6 µmol g -1 ) site. Thus, exchange sites are available in excess and do not set the maximum pool size. As a result where fluid residence times are long the bio-available fraction, formed by precipitation and sorption from pore fluids, is likely close to equilibrium with regolith fluid (Maher and Chamberlain, 2014).
Only at sites with high MAP regolith fluids might be diluted and deviate from a chemostatic behavior (e.g. Godsey et al., 2019), and these fluid concentrations will be below equilibrium concentrations. No discharge data is available for the 460 EarthShape sites, but with MAP slightly > 1000 mm yr -1 a dilution effect might be in effect at Nahuelbuta for elements like Mg and Ca that tend to switch from chemostatic to dilution behavior at high discharge (e.g. Godsey et al., 2019). At Nahuelbuta these are also the elements with the lowest bio-available inventory and are thus possibly present in concentrations below equilibrium there.
The third possibility is that the inventory of a bio-available nutrient pool is set by the rate at which an element is extracted into 465 plants from regolith relative to its replenishment by weathering. For a given bio-available reservoir an increase in uptake rate ( Fig. 4; Table 3) will decrease the turnover time, or residence time of mineral-derived nutrients in the soil bio-available fraction (T 789:;<.>98?,@ 6 ; Table 8). For all important mineral nutrients, we see a decrease in turnover time of the bio-available soil pool with respect to uptake from the arid to the humid sites. Short turnover times with respect to uptake also reflect fast recycling.
The humid-temperate ecosystem in the south of our ecological gradient is dominated by this "organic nutrient cycle". 470 Yet the organic nutrient cycle alone cannot be separated from its geogenic counterpart, because as discussed above the regolith nutrient pool is subject to losses by desorption and runoff, and these we consider to be quantified by the elemental weathering release flux W regolith X . Another loss trajectory not included in W regolith X might be erosion of solid plant matter (Uhlig et al., 2017).
To prevent the bio-available nutrient stocks from eventually running into depletion over longer timescales, the pool needs to be replenished. This replenishment takes place either by exogenous inputs (e.g. Boy and Wilcke, 2008;Porder et al., 2007;475 Vitousek, 2004;Vitousek et al., 2010), or by weathering of primary minerals at depth (Uhlig et al., 2017;Uhlig and von Blanckenburg, 2019). This replenishment is the "geogenic nutrient pathway". In the arid Pan de Azúcar, where weatheringrelease fluxes are low, these pools are being substantially replenished by the deposition of atmospheric sources (up to 93%; Table 6). In the other study sites the bio-available pools are being replenished by weathering of rock and regolith. The timescales T 789:;<.>98?,A 6 over which this replenishment from weathering occurs do not deviate much between semi-arid, 480 mediterranean, and humid-temperate climates, bar a few exceptions (Table 8). For Ca, Mg, and Sr these turnover times are within the same range as turnover times with respect to plant uptake, meaning these elements reflect a delicate balance between supply and nutrient demand. For P and K they are sometimes orders of magnitude longer than their turnover times with respect to plant uptake. For example, the inventory of K in the bio-available soil pool at Nahuelbuta is turned over every 30 years between soil and plants, but it takes 400 years to replenish their external losses by weathering. The turnover times of the bio-485 available fraction in saprolite with respect to weathering (calculated here for an inventory of the uppermost meter), decrease from the semi-arid, mediterranean, to the humid-temperate site for the elements Ca, Mg, K, and Sr (but not for P). The weathering turnover times in the south are lower because of the lower bio-available inventories, given that supply rates by weathering do not differ as much as the inventories.
A key observation of these considerations is the decrease in the saprolites' bio-available pool in K, Mg, and Ca from the dry 490 to wet sites, the increase in the organic nutrient cycle as reflected by decreasing turnover times with respect to plant uptake and increasing recycling ratios, all in the face of similar weathering supply rates of nutrients. We speculate that soil chemical properties and water discharge are of subordinate importance in setting the size of the bio-available regolith pool. Instead, we consider plants and their nutrient demands to primarily control its size. If true, biological mechanisms need to be in place that regulate this delicate balance between nutrient replenishment by weathering and plant uptake in the ecosystems of Santa 495 Gracia, La Campana, and Nahuelbuta, in order to prevent the nutrient pools from entire depletion.

Is weathering modulated by biota?
According to the conventional paradigm one would expect high weathering-release fluxes at sites at which MAP and hence runoff is high because weathering rate is proportional to runoff for the chemostatic elements that comprise the bulk of the weathering flux, amongst them Si that contributes roughly half of the flux (e.g. Godsey et al., 2019; Maher and Chamberlain, 500 2014). High NPP is thought to raise W regolith X as well (e.g. Berner et al., 2003;Brantley et al., 2011;Buss et al., 2005;Kelly et al., 1998;Porder, 2019;Schwartzmann, 2015). Indeed, plants and their associated micro-biota do enhance weathering rates in a direct and indirect way. (1) Deep plant roots can increase regolith thickness . (2) Land plants regulate the hydrological cycle and hence modulating runoff and total weathering fluxes (Drever and Zobrist, 1992). (3) Ectomycorrhizal fungi can actively extract nutrients such as P, K, Ca, Mg, and Fe from minerals (Finlay et al., 2009;505 Rosenstock, 2009;Rosling et al., 2004) distant from the root, and even during dry conditions and thereby increase mineral dissolution kinetics. (4) Through the respiratory release of soil CO2 and excretion of organic complexing agents plants, hyphae and their associated microbiota can increase the solubility limits of cations by a factor of up to <10 (Winnick and Maher, 2018). However, we see no increase in weathering rate along the MAP and NPP gradient, and, with the exception of arid Pan de Azúcar, not even an increase in weathering intensity as denoted by CDF (Fig. 3) or the depletion (t) of Ca, Mg, or K 510 (Fig. A1). These indices are similar between the other three sites even though they are sensitive indicators of primary mineral dissolution. We are faced with the situation that the increased plant cover and NPP not only fails to accelerate weathering. We speculate that the effect of vegetation might even compensate for a potential increase in weathering that would be caused by the increase in MAP, essentially damping the geogenic pathway.
To evaluate why biota and runoff do not accelerate weathering, we focus on the two study sites in semi-arid (Santa Gracia) 515 and humid-temperate climate (Nahuelbuta). In these sites, soil weathering rates (5 -10 t km −2 yr −1 ; Table 1) and weatheringrelease fluxes (W regolith X ; Fig. 4, Table 3) of most nutrients are on par despite massive differences in vegetation cover, primary productivity (NPP), and even precipitation (MAP; Table 1). By comparison, the La Campana site features higher weathering rates simply because of its steeper topography denudation rates are higher (e.g. Oeser et al., 2018;Schaller et al., 2018;van Dongen et al., 2019). We evaluate bio-weathering by the interactions (1) to (4) introduced above. 520 (1) Roots deepening regolith thickness. While a detailed survey of rooting depth along the EarthShape gradient has not been done, deep roots were not observed in Santa Gracia whereas in (La Campana and) Nahuelbuta individual roots reached up to several meters depth into saprolite. A and B horizons in Santa Gracia are shallow (20 -40 cm), whereas they are deep in Nahuelbuta (80 -100 cm; Oeser et al., 2018;Bernhard et al., 2018). We do not know the depth of the weathering front which appears to be at least a dozen of meters depth or more at both sites. Thus, deep rooting can benefit plant growth by making a 525 higher nutrient inventory available.
(2) Canopy and roots converting precipitation into evapotranspiration. In sites with higher vegetation cover, water vapor is being recycled and does not immediately enter runoff. By providing canopy, trees can modulate infiltration whereas they turn water back into transpiration (Ibarra et al., 2019). For example Ibarra et al. (2019) have shown that total runoff can decrease by up to 23% as vegetation cover raise from barely vegetated towards highly vegetated sites. Such an effective water recycling 530 hence decreases total runoff and potentially reduces weathering-release fluxes in the highly vegetated sites of La Campana and Nahuelbuta. However, along the EarthShape transect this potential total 23% reduction is a minor effect when considering the 100-fold increase in precipitation over the entire gradient. A larger effect might be presented by roots if they provide preferential flowpaths such that infiltrating water either bypasses the regolith matrix available for weathering . However, given the deep weathering front at the EarthShape sites that is likely beneath rooting depth we consider this 535 effect to be minor too, or in the contrary might even increase deep weathering. Thus, we consider the impact of plants on the hydrology that is relevant to weathering to be minor along the EarthShape gradient.
(3) Increase in primary mineral dissolution by ectomycorrhizal fungi. As yet we have no direct observations on nutrient foraging by fungi and other microbes in regolith from the EarthShape sites as obtained on other mountain sites in Chile (Godoy https://doi.org/10.5194/bg-2020-69 Preprint. Discussion started: 9 March 2020 c Author(s) 2020. CC BY 4.0 License. and Mayr, 1989). Parameters of total microbial biomass in saprolite show no increase along the gradient: total gene copy have 540 similar ranges from Santa Gracia to Nahuelbuta, and DNA amounts even decrease slightly (Oeser et al., 2018). Common strategies of microbial symbionts by tree roots suggest that energy investment into nutrient recycling from leaf litter is more advantageous than dissolving primary mineral (Andrino et al., 2019). Thus, we would expect that mycorrhiza predominantly aids recycling in La Campana and Nahuelbuta. In Santa Gracia, however, the absence of a litter layer may prompt the subsurface fungal network to invest into primary mineral dissolution, adding microbial weathering to total weathering at that 545 that site.
(4) Increasing solubility by release of soil CO2 and organic complexing agents. Process (4) mainly increases the equilibrium limit of elemental solubility. If dissolution is not kinetically limited, we would indeed expect higher solute concentrations with higher soil CO2, and hence higher dissolution rates of primary minerals (Winnick and Maher, 2018). Plants potentially impose a negative feedback onto this dependence by interfering into the silicon cycle. Because silicon is the most abundant element 550 in felsic rock and regolith (besides oxygen) it exerts the major control on the total weathering fluxes. The conversion of dissolved silicon into biosilica by uptake formation of e.g. phytoliths would aid kaolinite formation (Lucas, 2001). However, the Si uptake flux at all sites is negligible compared to the Si weathering flux: W regolith Si amounts to 2100 and 2000 mg m −2 yr −1 in Santa Gracia and Nahuelbuta, respectively (Table 3). Si concentration in the above-ground living ecosystem of Santa Gracia (1400 μg g −1 ) exceeds the Si concentration in Nahuelbuta (110 μg g −1 ; Table 5). This leads to an uptake of Si from soil solution 555 (U total Si ) of 100 mg m −2 yr −1 in Santa Gracia and 30 mg m −2 yr −1 in Nahuelbuta (Table 3) which is only 2 -5% of the Si release flux. We can therefore exclude plant Si uptake and recycling of Si as a factor impacting weathering rates. A key observation is provided by the analysis of pedogenic oxides (i.e. dithionite-extractable Al, Si extracted by oxalate, dithionite, and pyrophosphate; Oeser et al., 2018) and cation exchange capacity (Bernhard et al., 2018). These analyses suggest high amounts of amorphous precipitates and secondary minerals in the regolith of Nahuelbuta but not in Santa Gracia. We thus argue that Si 560 is effectively captured in these barely soluble secondary minerals and in turn, W regolith Si in Nahuelbuta is subdued in spite of elevated solubility of primary minerals due to increased CO2 respiration by roots.
Ecosystems thus exert substantial control over weathering by both directly and indirectly modulating processes. These processes can either enhance or reduce weathering fluxes and result, in combination with effective recycling loops of plantlitter material, in well-balanced nutrient cycles. The silicate weathering fluxes become effectively decoupled from the ultimate 565 nutrient demands of biota. Our empirical evidence provided here suggests that the combination of recycling and negative feedbacks on weathering by secondary solid formation within the regolith counterbalance weathering rates in areas of high vegetation and biomass growth from what they would be in the absence of high biomass density.

Conclusions
We found that even though the four EarthShape study sites feature a steep north-south gradient in MAP and NPP, none such 570 a gradient was apparent for weathering rates and weathering intensity between the study sites situated in semi-arid, https://doi.org/10.5194/bg-2020-69 Preprint. Discussion started: 9 March 2020 c Author(s) 2020. CC BY 4.0 License. mediterranean, and humid-temperate climate. Thus, the ecosystems with high NPP maintain their nutrient supply by increasing recycling rather than increasing weathering. We consequently find that the "organic nutrient pathway" intensifies, whereas the "geogenic nutrient pathway" stays steady despite increasing MAP and NPP.
Ecological stoichiometry in plants and Sr isotopes reflects the bio-available fraction in saprolite and soil. At the sites with high 575 NPP, the bio-available fraction approaches a plant-dominated ecological stoichiometry signature by intense recycling. Besides nitrogen, the first mineral nutrient to be limit plant growth might be K rather than P. However, we found that deep-rooting plants can bypass this shortage K in the upper regolith by deep K-uptake from beneath the regolith profiles.
In fact, the presence of plants might compensate a potential weathering increase along the gradient downward by regulating the hydrological cycle, fostering secondary mineral formation, and a microbial community specializing on nutrient-recycling 580 rather than nutrient-acquisition through weathering. Due to this nutrient buffering by recycling, higher plants today may not be a big driver in the global silicate-weathering cycle.

Weathering indices (CDF & τ)
Zr, Ti, and Nb are commonly used to estimate mass losses to the dissolved form during weathering (Eqs. 5 & 6) as they are presumed to be the least mobile elements during weathering (Chadwick et al., 1990;White et al., 1998). The suitability of these elements for the EarthShape study sites has been evaluated and thoroughly discussed on a site to site basis in Oeser et al. (2018). Based on possible Ti-mobility in some samples and the fact that Zr is used as a reference element in the majority of 590 weathering and soil production studies worldwide (e.g. Fisher et al., 2017;Green et al., 2006;Hewawasam et al., 2013;Riebe and Granger, 2013;Riebe et al., 2001;Schuessler et al., 2018;Uhlig et al., 2017), Zr was taken as immobile reference element in this study.
The calculations of these weathering indices rely on a good approximation of the regolith profiles' initial substrate (i.e. chemical bedrock composition). Thus, any regolith sample with a Zr concentration that was lower than the mean of 595 unweathered bedrock by more than one standard deviation (1SD) was excluded from further consideration. These samples potentially originate from a chemically distinct bedrock or highlight another type of substrate heterogeneity (e.g. a pegmatitic vein) as a lower Zr concentration cannot be due to weathering. Saprolite samples were also excluded from our data set if Cr and Ti concentrations were twice those of unweathered bedrock (+ 1SD). Elevated concentrations of these elements imply the presence of mafic precursor rock such as commonly present in bedrocks' mafic enclaves. All such excluded samples are 600 marked grey in Figs. 3 & A1.
The concentration of K throughout the entire regolith profiles in Santa Gracia is three-fold higher than K contained in local bedrock samples (Oeser et al., 2018). We thus assume that the K concentration in the bedrock samples of Santa Gracia as determined by Oeser et al. (2018) underestimates the actually occurring K concentration of local bedrock. Thus, τ K has been calculated using published values on K and Zr concentration from a study nearby (Miralles González, 2013). 605

Weathering fluxes
To estimate elemental release fluxes from regolith (Eq. 3) for each study site, the most negative τ-values from the shallowest mineral-soil sample of each regolith profile were used (red-circled symbols in Fig. A1), resulting in an integrated net-elemental release over each entire regolith profile. Ultimately, the elemental chemical weathering flux (W regolith X ) at each study site has been averaged. Because some fraction of this flux might not be in the reach of the plants' roots (e.g. if a fraction is lost into 610 deep groundwater) an upper estimate of the nutrient fluxes from rock into vegetation results. W regolith X is reported in Table 3.

Ecosystem nutrient uptake fluxes
Total ecosystem nutrient uptake fluxes (U total X ) have been evaluated using Eq. 4 and are reported in Table 3. Net primary productivity (NPP), has been derived from a dynamic vegetation model (LPJ-GUESS) simulating vegetation cover and 615 composition during the Holocene (Werner et al., 2018) and is reported in Table 1. Biomass production was estimated from NPP(C) by assuming that dry biomass consists of 50 wt% carbon. To obtain the element-specific uptake rate U X GrowthRatePlant is multiplied with the bulk concentration of X in the plants [X]Plant. To estimate an elemental bulk ecosystem composition, we applied the dimensionless organ growth quotients GL/GS (leaf growth relative to stem growth) and GL/GR (leaf growth relative to root growth) in accordance with Niklas and Enquist (2002). However, the sampling and analyses of 620 roots was excluded in this study, because of the difficulties that arose in associating roots to a specific tree or shrub. For elemental analysis this difficulty is compounded by the difficulties encountered during the roots' purification from soil particles. We thus adapted the organ growth quotients from the work of Niklas and Enquist (2002), such that we only differentiate between the growth rate of leaves and stem, respectively, and the differences of these quotients between angiosperms and gymnosperms. Ultimately, elemental bulk ecosystem composition (Table 5) has been determined by 625 weighting the averaged elemental composition for each sampled plant for their relative abundance in the respective ecosystem.

Inventories
The inventories for the bio-available fraction (I bio-av. X ) and in bulk regolith (I bulk X ) have been calculated using Eq. 8 and are reported in Table 4. I bio-av. X was determined for both the bio-available fraction in soil (comprised of the A and B horizon; 630 I bio-av, soil X ) and saprolite (I bio-av, sap X ). For the calculation of all inventories we used the soils' bulk density determined by X is comprised of elements contained in fine-earth material and in fragmented rocks and coarse material (e.g. core stones). We derive the relative amount of coarse material of each depth increment from Bernhard et al.
(2018) and allocate them the bedrocks' chemical composition (Table S1). If information on either bulk density or the relative amount of coarse material was not available, the respective horizons' average has been used for the calculation of I i,j X . 635 Ultimately, we do not know how far the saprolite extents to depth in each of the eight regolith profiles. Thus, for comparison purposes, we calculated the inventories of the bio-available fraction in saprolite (I bio-av, sap X ) and in bulk regolith (I bulk X ) to the depth of the respective regolith profile and normalized this value to the arbitrary value of 1 m.

Nutrient recycling factor 640
The nutrient recycling factor Rec 6 was calculated using Eq. 7 and is reported in Table 7. This factor, however, might represent an underestimate as it is parameterized with W regolith X and some fraction of which will bypass nutrient uptake by plants if it is drained directly via groundwater into streams. Moreover, Rec 6 might also be underrated for some elements as it does not account for stem-flow fluxes and throughfall. According to e.g. Wilcke et al. (2017), these fluxes are generally highest for K compared to other plant-essential elements. 645 In Pan de Azúcar, where atmospheric deposition (Dep dry X and Dep wet X ) has been reported (e.g. increasing τ-values towards the profiles top despite bio-lifting of elements and field observation; Oeser et al., 2018) we need to consider these inputs in addition to the weathering release fluxes (W regolith X ). Thus, to account for all potential sources of elements available for plant uptake, the nutrient recycling factor in Pan de Azúcar has been determined using: Atmospheric deposition fluxes have been estimated by determining the absolute difference between the lowest τ-value in the shallowest mineral-soil sample and the highest τ-value in the soil profile above it. Further, we assume that elemental gains (i.e. increasing τ-values) in the regolith profiles are attributed solely to atmospheric deposition. We test these estimates for atmospheric depositional fluxes by placing the elemental gains in proportion to the initially determined weathering release fluxes (W regolith X , Eq. 3; Table 3).

Uncertainty estimation on nutrient fluxes
The uncertainties on the nutrient fluxes of W regolith X and U total X were estimated performing Monte Carlo simulations in which 20 000 random data sets were sampled within the standard deviation of all input parameters using a Box-Muller transformation (Box and Muller, 1958). The uncertainties used for the simulation were the SD of W regolith X , the respective regolith profile's denudation rate (D , Table 1), the SD of the bedrocks' element concentration of interest, and 3% relative uncertainty on the element concentration in regolith samples. In the case of U total X the SD of the respective study site's NPP and the SD of the chemical composition of the weighted plants (Table 5) were used.

Sample availability
All sample metadata are already available on a public server using unique sample identifiers in form of the "International Geo Sample Number" (IGSN). 665 9 Author contributions R.A. Oeser conducted field sampling, analyzed samples, interpreted data, and wrote text. F. von Blanckenburg designed the study, selected the study sites, interpreted data, and wrote text.

Competing financial interests
The authors declare no competing financial interests. 670

Additional information
Supplementary data tables are available at GFZ data services (Oeser and von Blanckenburg, 2020).

Acknowledgements
We acknowledge CONAF in Chile for providing us with the opportunity to work in the national parks of Pan de Azúcar, La Campana, and Nahuelbuta. We also thank CEAZA for facilitating access to the Reserva Natural Santa Gracia. We are grateful 675  and Biochemistry, 89, 109-122, 10.1016/j.soilbio.2015.06.029, 2015. Delhaize, E., and Ryan, P. R.: Aluminum Toxicity and Tolerance in Plants, Plant Physiology, 107, 315-321, 1995: Phosphorus in forest ecosystems: New insights from an ecosystem nutrition perspective, Journal of Plant Nutrition andSoil Science, 179, 129-135, 10.1002/jpln.201500541, 2016. Lee, J.-E., andBoyce, K.: Impact of the hydraulic capacity of plants on water and carbon fluxes in tropical South America, J Geophys Res, 115, 10.1029/2010jd014568, 2010. Lenton, T. M., Crouch, M., Johnson, M., Pires, N., and Dolan, L.: First plants cooled the Ordovician, Nature Geoscience, 5, 86-89, 10.1038/ngeo1390, 2012 Mechanisms of silicon-mediated alleviation of abiotic stresses in higher plants: a review, Environ Pollut, 147, 422-428, 10.1016/j.envpol.2006.06.008, 2007 The Role of Plants in Controlling Rates and Products of Weathering: Importance of Biological Pumping, Annual Review of Earth and Planetary Sciences, 29, 135-163, 2001.   . Uncertainty bars show 1SD. Differences in nutrient recycling factors for the plant essential elements Ca, K, Mg, and P among the four study sites are highlighted in the lower panels. Note that here we use the Rec X calculated for W regolith X from silicate weathering only. In Table 7 Table 5 for weighting parameters). Whiskers span 90% of the 1035 respective data set. On the boxes' right-hand site, the differences between north-and south-facing regolith profiles are depicted. Note that bulk regolith samples in Nahuelbuta with anomalously low Zr concentrations have been excluded from this analysis as they are suspected to comprise a different parent rock. Y-axis covers broader range in Pan de Azúcar and Nahuelbuta than Santa Gracia and La Campana.

Campana, and Nahuelbuta (from left to right) for plant-essential (closed symbols) and plant-beneficial elements (open symbols).
Weathering-release fluxes for Ca, K, Mg, Na, and Sr in Pan de Azúcar have been complemented by atmospheric depositional fluxes such that the total amount of available nutrients increase by 95, 22, 18, 12, and 10%, respectively. Grey contour lines emphazize the nutrient recycling factor (Rec X ), which is the ratio of U total X to W regolith X . Uncertainty bars show 1SD. Differences in nutrient recycling factors for the plant essential elements Ca, K, Mg, and P among the four study sites are highlighted in the lower panels.   Total mass fluxes (in e.g. t km −2 yr −1 ) Eq.
(1) D = E + W denudation rate; the sum of chemical weathering and physical erosion E physical erosion; physical removal of primary and secondary minerals along with biogenic material Eq.
(2) W = D × CDF chemical weathering rate; net-chemical release flux from minerals as some fraction of which is being incorporated into secondary minerals and pedogenic (hydr-)oxides GPP gross primary production; gross carbon input into biomass NPP net primary productivity; net-carbon fixation by biomass Inventory of element X in the bio-available fraction in soil I bio-av. sap 2 Inventory of element X in the bio-available fraction in saprolite I bulk 2 Inventory of element X in bulk regolith

Elemental turnover times (in e.g. yr)
Eq. (9) T i,j X = I i X j Turnover time of element X in compartment i with respect to input or output flux j; the ratio of total stock of element X in i to input or output flux j T bio-av, U -Turnover time of element X in the forest floor with respect to uptake into trees; mean time a nutrient rest in the forest floor before re-utilization by forest trees T bio-av, W 2 Turnover time of element X in the bio-available fraction in regolith with respect to adsorption onto clay minerals; mean time over which the inventory of the bio-available fraction is replenished by chemical silicate weathering in the absence of other gains or losses https://doi.org/10.5194/bg-2020-69 Preprint. Discussion started: 9 March 2020 c Author(s) 2020. CC BY 4.0 License.  Table 4 Inventories of plant-essential and plant-beneficial elements in bulk regolith and the bio-available fraction in soil and saprolite. Apart from phosphorus, the accessibility of these elements was determined using a sequential extraction method described Table 5 Chemical composition of the above ground living plants. Plant organs have been weighted according to Niklas and Enquist (2002), using the plant organs' relative growth rate (see Appendix A). Relative growth rates and relative abundance of the different plant species can be found in this table's footnotes. The unweighted chemical composition of each plant organ is listed in Table S5 (Oeser and von Blanckenburg, 2020). Table 6 Average 87 Sr/ 86 Sr ratio for bulk bedrock, bulk regolith, and the bio-available fraction in saprolite and soil. 87 Sr/ 86 Sr in bulk plants are weighted by the plant organs' relative growth rate and relative species abundance in the respective ecosystem (see Table  5). Radiogenic Sr composition for each single specimen are reported in Tables S2 (bulk regolith samples Pearce et al. (2015) † Standard deviation corresponds to species-to-species differences in 87 Sr/ 86 Sr ‡ Potential seaspray input into the bio-available fraction derived from a simple two-component mixing equation using bulk bedrock and seaspray as end-members. Substantial seaspray incorporation into the bio-available fraction in La Campana and Nahuelbuta is very unlikely (see text for discussion), therefore not shown. Table 7 Nutrient recycling factors in Pan de Azúcar, Santa Gracia, La Campana, and Nahuelbuta. Shown in brackets are the Rec X prior correction for atmospheric deposition. Table 8 Turnover times for the soil and saprolite bio-available fraction with respect to the release by weathering and turnover times for bio-available fraction in soil with respect to uptake into plants.