The Median Isn’t the Message: Elucidating Soil Nutrient Hot Spots

doi:10.21203/rs.3.rs-3126146/v1

Download PDF

Research Article

The Median Isn’t the Message: Elucidating Soil Nutrient Hot Spots

https://doi.org/10.21203/rs.3.rs-3126146/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 13 Jan, 2024

Read the published version in Biogeochemistry →

You are reading this latest preprint version

Soil nutrient distribution is heterogeneous in space and time, potentially altering nutrient acquisition by trees and microorganisms. Ecologists have distinguished “hot spots” (HSs) as areas with enhanced and sustained rates of nutrient fluxes relative to the surrounding soil matrix. We evaluated the spatial and temporal patterns in nutrient flux HSs in two mixed-conifer forest soils by repeatedly sampling the soil solution at the same spatial locations (horizontally and vertically) over multiple seasons and years using ion exchange resins incubated in situ. The climate of these forests is Mediterranean, with intense fall rains occurring following summers with little precipitation, and highly variable winter snowfall. Hot spots formed most often for NO₃^- and Na⁺. Although nutrient HSs often occurred in the same spatial location multiple times, HSs persisted more often for NO₃^-, NH₄⁺, and PO₄^3- and were more transient for Na⁺, Ca²⁺, and Mg²⁺. Sampling year (annual precipitation ranged from 701 to 1216 mm) impacted the occurrence of HSs for most nutrients, but season was only significant for PO₄^3-, NH₄⁺, NO₃^-, and Na⁺, with HSs forming more often after fall rains than spring snowmelt. The frequency of HSs significantly decreased with soil depth for all nutrients, forming most commonly immediately below the surficial organic horizon. Although HSs accounted for less than 17% of the sampling volume, they were responsible for 56-88% of PO₄^3-, NH₄⁺, and NO₃^- fluxes. Our results suggest that macronutrient HSs have a disproportional contribution to soil biogeochemical structure, with implications for vegetation nutrient acquisition strategies and biogeochemical models.

biogeochemical cycles

Mediterranean-type climate

mixed-conifer forest

nutrient fluxes

Sierra Nevada

soil nutrient heterogeneity

Understanding the ecological significance, spatial and temporal patterns, and biotic and abiotic controls of soil nutrient heterogeneity remains a frontier in terrestrial ecosystem science (Lovett et al. 2005; Turner and Gardner 2015). Plants can exhibit a morphological response to nutrient-rich patches by increasing root growth, or a physiological response by increasing specific root nutrient uptake rates (Robinson 1994). These responses can provide an interspecific competitive advantage to species that have the capacity to exploit these nutrient-rich patches (Robinson et al. 1999; Hodge et al. 1999; Wang et al. 2018), lead to increased shoot biomass relative to belowground biomass (Day et al. 2003), and impact biodiversity and structure at the community level (Wijesinghe et al. 2005; Maestre et al. 2005; Maestre and Reynolds 2007; Palacios et al. 2011). Plants that utilize nutrient rich patches can also outcompete soil microorganisms for these resources over the long-term (Kaye and Hart 1997; Hodge et al. 2000, 2004). Although plants clearly respond and exploit nutrient patches, we currently have little information, in situ, on the distribution of nutrient patches in space and how they change over time.

Areas of elevated biogeochemical activity in soil have been recognized by the terrestrial ecological community for a long time as so-called “hot spots” (HSs; Parkin 1987; McClain et al. 2003; Vidon et al. 2010). Hot spots have been identified quantitatively by their positive skew (Kurunc et al. 2011), and as high statistical outliers in nutrient pool sizes, transformation rates, or fluxes (e.g., (Johnson et al. 2010; Darrouzet-Nardi and Bowman 2011). Biogeochemical activity can also be “hot” temporally; these so-called “hot moments” (HMs) are defined by a high rate of biogeochemical activity occurring at a specific location for a short period of time relative to the rate observed in the bulk soil over longer time periods (McClain et al. 2003; Vidon et al. 2010). More recently, Bernhardt et al. (2017) recommended using the new term of “ecosystem control points” to better recognize that HS and HM are integrated in time and space; however, in this paper, we present and discuss our findings using the HS-HM nomenclature because they occur more commonly in the literature, while still acknowledging spatiotemporal dynamics.

Many investigations of HSs assess the dynamics of only one or two nutrients (Bernhardt et al. 2017) over a relatively short timeframe (i.e., within one year; Gibson 1986; Kleb and Wilson 1997; Johnson et al. 2014; Barcellos et al. 2018), and often use destructive sampling techniques when conducted in the field (i.e., Bruckner et al. 1999 Charley and West 1977; Kleb and Wilson 1997; Farley and Fitter 2001; Washburn and Arthur 2003; Mueller et al. 2008; Harms and Grimm 2008; but see Bilyera et al. 2020; Kratz et al. 2022). Without repeatedly sampling the same location over time, it is impossible to know if HSs are elevated for prolonged periods of time or if they only occur periodically (i.e., as HMs). Additionally, without evaluating several nutrients simultaneously, the relative importance of HS phenomena to overall biogeochemical cycling within an ecosystem cannot be ascertained. Before we can incorporate HSs into terrestrial modeling, we need to better quantify their presence and determine how soil properties and other environmental factors, such as water availability, temperature, and vegetation, regulate their occurrence (Groffman et al. 2009; Chen et al. 2021; Zhao et al. 2021).

There are a number of contributing factors that have been identified to create HSs. Hot Spots may occur in localized areas where vegetation concentrates nutrient resources due to enhanced atmospheric dry deposition (via impaction and washout), throughfall and stemflow, litterfall, and decreased erosion (Halvorson et al. 1994; Schlesinger et al. 1996; Weathers et al. 2001; Darrouzet-Nardi and Bowman, 2011). Additionally, depressional topographical positions may form HSs relative to upland landscape positions due to the accumulation of organic and inorganic nutrients, and enhanced soil water content promoting microbial activity (Harms and Grimm 2008; Barcellos et al. 2018). On a smaller spatial scale, surface microtopography, such as rocks and preferential flowpaths, can funnel soluble nutrients into the mineral soil to form nutrient HSs (Bundt et al. 2001; Johnson et al. 2011; Frei et al. 2012; Göransson et al. 2014).

In contrast, HMs are regulated by abiotic or biotic controls that are transient at a given location to elevate nutrient pools, processing rates, or fluxes. For example, pulses of nutrients or microbial activity may only be activated immediately following re-wetting events of dry soil (McClain et al. 2003; Harms and Grimm 2008; Barcellos et al. 2018; Barrat et al. 2021), or immediately after catastrophic disturbances such as wildfire (Zhao et al. 2021). In addition, dead and decaying roots or macrofauna can create biopores that accrue nutrients and serve as HMs of microbial activity (Kuzyakov and Blagodatskaya 2015; Hoang et al. 2016). The timeframe and interactions of controls regulate how long the area is “hot,” with HM formation essential for HS occurrence (Bernhardt et al. 2017; Barcellos et al. 2018; Wagner-Riddle et al. 2020), and dictate the respective impact HSs and HMs have on the biogeochemical behavior of the ecosystem. Indeed, the designation of HSs and HMs are defined by individual studies based on the specific temporal and spatial scales considered.

Forests that experience Mediterranean-type climates, like the Sierra Nevada of California, are excellent model systems to explore HS phenomena. Nutrient fluxes in Sierran forests are driven largely by activated hydrologic flowpaths that occur primarily following rains in the fall after a prolonged summer dry-period, and following relatively rapid snowmelt in the spring (Homyak et al. 2014; Williams et al. 1995; Hunsaker and Johnson 2017). Annual precipitation in these ecosystems is highly variable and is becoming more inconsistent with climate change (Zamora-Reyes et al. 2021). In a mixed-conifer forest in the western Sierra, Johnson et al. (2014) reported persistent and elevated magnesium (Mg²⁺) fluxes over fall and spring at specific soil locations, suggesting HSs; however, other nutrient HSs (ortho-phosphate: PO₄³⁻; ammonium: NH₄⁺; nitrate: NO₃⁻; calcium: Ca²⁺; and sodium: Na⁺) exhibited more HM phenomena because their fluxes varied spatially with time. Although this and associated studies (Woodward et al. 2013; Johnson et al. 2011; Woodward et al. 2013) improved our understanding of HS controls in Mediterranean forests, nutrient flux collections were only done in the fall and spring of a single year; hence, it is uncertain if the observed patterns persist across multiple years with contrasting precipitation inputs. Understanding the longer-term controls on HS is likely to become increasingly important as the climate continues to warm (Ma et al. 2017).

Using the same plots and sampling design, and building upon the data collected over a one-year period reported by Johnson et al. (2014), we investigated long-term spatial and temporal patterns of nutrient fluxes in a Sierran mixed-conifer forest using ion exchange resin (IER) capsules. Specifically, over a four-year period, we characterized the magnitude and spatial distribution of nutrient fluxes (PO₄³⁻, NH₄⁺, NO₃⁻, Ca²⁺, Mg²⁺, and Na⁺) after the first significant fall rains following the dry summer growing season and after snowmelt in the spring. We sampled the soil solution at the same spatial locations (horizontally and vertically) repeatedly over time to achieve the following research objectives: (1) characterize the spatiotemporal trends in HS occurrence within and across sampling years, seasons, and soil depths; (2) determine if HSs of different nutrients co-vary with each other over time; and (3) quantify the relative contribution of nutrient HSs to overall biogeochemical fluxes.

Site descriptions

The study area is located on the western slope of the southern Sierra Nevada, California, USA. This region experiences a Mediterranean-type climate where mean annual air temperature is 9.8°C and mean annual precipitation is 1325 mm y^− 1, with 35–60% falling as winter snow (Yang et al. 2021). The vegetation is a mixed-conifer forest with canopy-dominant tree species of sugar pine (Pinus lambertiana Douglas), white fir (Abies concolor (Gordon & Glend.) Lindl. ex Hildebr.), ponderosa pine (Pinus ponderosa P. Lawson & C. Lawson), and incense cedar (Calocedrus deucrrens (Torr.) Florin). Soils are derived from granitic parent material and classified in the Shaver soil series, a coarse-loamy, mixed, superactive, mesic Humic Dystroxerept (Johnson et al. 2014).

Sampling design

Johnson et al. (2014) established a 6- x 6-m grid of resin capsules (UNIBEST PST-1, https://www.unibestinc.com/technology; UNIBEST Corporation, Washington, USA; Yang and Skogley 1992;Dobermann et al. 1994) at two representative plots within a Sierran mixed-conifer forest (Fig. 1). These locations were near the plots where previous field investigations on HS-HM nutrient phenomena took place(Johnson et al. 2011; Woodward et al. 2013). Plot 1 (Lat. 37.061°, Long. -119.183°) is at an elevation of 1,980 m, has an eastern aspect, and is on a 20% slope. Plot 2 (Lat. 37.065°, Long. -119.203°) is at an elevation of 1,910 m, has an eastern aspect, and is on a 5% slope. Plot 1 has less dense vegetation than Plot 2, with 85% and 95% average overstory canopy cover and 45 and 75 m² ha^− 1 tree basal area, respectively (Woodward et al. 2013). At each plot, 16 resin capsules were placed carefully at 2-m intervals horizontally within this grid immediately below the O horizon (i.e., 0-cm mineral depth) using a hand trowel to minimize disturbance. Adjacent to these resin capsules, WECSA® Access System units (WECSA® LLC; Montana, USA) were installed at 20-, 40-, and 60-cm depths at an angle that varied according to the microtopography, such that the deeper capsules would all be vertically coincident with their respective 0-cm capsule. The WECSA® system consisted of a thinned-walled PVC casing with a removable cap on one end and a removable resin capsule carrier that fits inside the casing; this allows the delivery and retrieval of a resin capsule so that the same soil location can be sampled repeatedly over time. The WECSA® systems were not installed in mineral soil locations when large roots or rocks prevented vertical placement below the 0-cm resin. This sampling design resulted in a horizontal and vertical matrix consisting of 60 resin capsules in Plot 1 and 63 in Plot 2 that allowed repeated sampling of the same locations with minimal soil disturbance. Only the 0-cm capsules required any subsequent soil disturbance, and this was kept at a minimum by carefully peeling back the O horizon during capsule retrieval and replacement. When deployed, the resin capsules adsorb bioavailable elements (i.e., inorganic nutrient ions) transported to the resins via diffusion and mass flow in the soil solution (Binkley 1984; Hart and Binkley 1984; Dobermann et al. 1994; Skogley and Dobermann 1996).

Resin capsules were installed at both plots in August 2011 and retrieved in October 2011 after the first major precipitation event in the fall (defined as ≥ 30 mm of daily precipitation). Then, new resin capsules were installed and left in situ until the completion of snowmelt in June 2012 (hereafter referred to as the Water Year 2012 collection; Johnson et al. 2014). This procedure was repeated in Plot 1 for water years (WY) 2013 and 2014 and for Plot 2 for WY 2013, 2014, and 2016 (resin capsules were not installed during WY 2015 in either plot; Table S1). This resulted in a final population of n = 360 at Plot 1 and n = 504 at Plot 2 when considering all spatial locations and sampling periods.

Precipitation and soil volumetric water content data were acquired over the entire study period to evaluate the hydrologic environment experienced by the resin capsule during incubation. Volumetric water content (m³ m^− 3) was measured in soil with a northern aspect at the Upper Providence Creek meteorological station (Long. 37.075°, Lat. -119.182°; 1920 m elevation) located ~ 2 km from the resin plots (Critical Zone Observatory; https://doi.org/10.6071/Z7WC73). Values were recorded hourly using Decagon Devices ECHO-TM at 10-, 30-, 60-, and 90-cm depths, and then averaged by day using the dplyr package in R (R Core Team 2008; Hadley et al. 2018). Daily precipitation (mm d^− 1) was also collected at the Upper Providence meteorological station using a Belfort-TM 50780 shielded weighing rain gauge (Kings River Experimental Watershed; Hunsaker and Safeeq 2018). Soil penetration resistance (kPa; i.e., soil strength) was measured at each plot to identify bulk density differences that may impact water fluxes using a Field Scout SC900 Soil Compaction Meter (Spectrum Technologies, Inc., Aurora, IL, USA) with a 1.27-cm cone size. Soil penetration resistance measurements were made after the last set of resin was collected and taken throughout both plots in undisturbed soil between the locations where WECSA® access systems were installed. Penetration resistance was logged at 106 and 87 points in Plot 1 and Plot 2, respectively, from 0- to 45-cm (maximum instrument depth); these values were then averaged across all recorded points within a plot for each 2.5-cm depth increment.

Laboratory analyses

After the capsules were collected and upon returning to the laboratory at University of California, Merced, USA, adhering soil was removed with deionized water. The resin capsules were processed following the procedure described in Johnson et al. (2014). Briefly, the resin capsules were extracted with three sequential 20-mL solutions of 2 M KCl on an orbital shaker at low speed (≈ 60 rpm) for 20 min each. The three extracts were then combined (60 mL). This procedure was used to ensure complete exchange of adsorbed ions from the resin (Hart and Binkley 1984). For the WY 2012 samples (data reported in Johnson et al. 2014), extracts were sent to the Oklahoma State University Soil, Water and Forage Laboratory (SWAFL) and analyzed for NH₄⁺-N, NO₃⁻-N, and PO₄³⁻-P on a Lachat 8000 Flow-Injection Analyzer (Hach Co., Loveland, CO, USA), and Ca²⁺, Mg²⁺, and Na⁺ were analyzed using a Jarrell Ash Inductively Coupled Plasma Optical Emission Spectrometer (ICP-OES; Thermo Scientific, Waltham, MA, USA). For extracts generated from resin capsules collected after the first water year (i.e., WY 2013, 2014, and 2016), extracts were analyzed for NH₄⁺-N, NO₃⁻-N, and PO₄³⁻-P on a Lachat 8000 flow-injection analyzer, and Ca²⁺, Mg²⁺, and Na⁺ on a Perkin-Elmer Optima 5300 DV ICP-OES (Waltham, MA, USA) at the Environmental Analytical Laboratory, University of California, Merced.

Nutrient fluxes were calculated by converting nutrient concentrations in the extracts to a nmol cm^− 2 d^− 1 basis; this was achieved by transforming mass to moles for that particular nutrient, multiplying the concentrations by the total extractant volume (60 mL), and then dividing those values by the cross-sectional area of the capsule exposed to the soil (5.7 cm² for WECSA® deployed capsules and 11.4 cm² for 0-cm depth capsules; Johnson et al. 2014) and the incubation period (Table S1). Although the number of days the resin was deployed varied by year (Table S1), high nutrient fluxes generally occur in these ecosystems only during brief periods immediately following the first significant rainfall after the summer dry period in the fall, and in the spring during snowmelt (Homyak et al. 2014; Williams et al. 1995; Holloway and Dahlgren 2001; Hunsaker and Johnson 2017).

HS identification

Hot spots were identified as positive moderate outliers using the following equation (Johnson et al. 2014):

$$X>Q3+1.5\left(IQR\right),$$

where X is the HS value (i.e., outlier), Q3 is the third quartile value (75th percentile) and IQR is the interquartile range (25th -75th percentiles). The 25th percentile, 75th percentile, and IQR for a given nutrient flux were determined using all grid locations, depths, seasons, and years, but data from each plot were analyzed separately (Table S1). Plots were not treated as field replicates due to physical and chemical differences that may mask individual plot outliers, and therefore HS patterns. No negative outliers were observed (Johnson et al. 2014).

Statistical analyses

Table 1 Nutrient flux (nmol cm^-2 d^-1) descriptive information including mean (standard deviation), median, minimum, maximum, interquartile range (25^th-75^th quartiles), skew, and control point influence (CPI) for PO₄^3-, NH₄⁺, NO₃^-, Ca²⁺, Mg²⁺, and Na⁺, calculated for Plot 1 and Plot 2 separately.

	Mean (SD)		Median	Min	Max	IQR	Skew
		nmol cm^− 2 d^− 1
Plot 1
PO₄⁺	2.25 (6.17)		0.46	0	60.12	1.27	5.38
NH₄⁺	1.27 (3.22)		0.15	0	29.67	1.12	5.13
NO₃⁻	3.76 (16.2)		0.14	0	186.42	1.53	8.47
Ca²⁺	16.0 (28.0)		4.69	0	187.51	13.6	3.18
Mg²⁺	2.74 (5.26)		0.93	0	46.77	2.43	4.39
Na⁺	1.92 (6.92)		0.06	0	56.34	0.20	5.81
Plot 2
PO₄⁺	0.86 (2.51)		0.23	0	24.51	0.57	6.28
NH₄⁺	4.37 (30.3)		0.18	0	614.73	1.17	17.6
NO₃⁻	7.51 (25.1)		0.14	0	280.25	3.38	6.40
Ca²⁺	18.67 (56.8)		3.48	0	648.65	12.0	7.43
Mg²⁺	4.08 (16.5)		0.65	0	202.98	1.88	8.52
Na⁺	1.38 (4.65)		0.08	0	51.61	0.28	6.43

Mixed-effect models were conducted to understand the likelihood of HS formation (binary response) with water year, season, and depth (i.e., the factors hypothesized to co-vary with HS formation). Models were run separately for each nutrient and plot. Water year (categorical; WY 2012 coded as 0), depth (continuous), season (categorical; fall coded as 0, spring as 1), and the interaction between depth and season were used as fixed effects. Random intercepts were used to account for repeated measures on the grid locations (n = 60 and 63 for Plots 1 and 2, respectively) and for the spatial dependency of the vertically associated grid points (n = 16). Models were performed with the lme4 package in R (Bates et al. 2015) and were fit by maximum likelihood with the Laplace Approximation. Simplified models were used for Ca²⁺ in Plots 1 and 2 and Mg²⁺ in Plot 1 because the full models would not converge due to the lack of response variables (i.e., no HSs) for WY 2013. In the simplified models, WY 2013 was removed from the dataset, but all other model characteristics remained the same. Multi-collinearity of the fixed variables was inspected by calculating the Variance Inflation Factor (VIF) using the VIF function in the car R package (Fox and Weisberg 2019). All VIF values were less than 4 (except for the simplified HS presence models, which were less than 9) for all independent variables (including the interaction term). Marginal (only fixed effects considered) and conditional (both fixed and random effects considered) R² values were calculated following Nakagawa et al. (2013) using the R MuMIn package (Barton 2009). When explanatory variables (WY, season, depth) were statistically significant (p < 0.05), Tukey’s Post Hoc tests were performed to evaluate whether means were significantly different from each other using the lsmeans function within the emmeans R package (Lenth et al. 2018). Models were also run with precipitation, season, and depth as explanatory variables; however, large VIF values resulted between precipitation and season due to multicollinearity and therefore these model results are not reported. Correlations between total precipitation and HS count are provided in the Supplemental Information (Fig. S1). We also evaluated the relationship between soil volumetric water content and HS formation; however, because soil moisture measurements were not co-located with resin capsules and moisture is known to be highly heterogenous within the soil profile (Franklin et al. 2021), we use these data to better understand water dynamics of the entire soil system rather than directly linking them to HS formation.

Spearman Rank correlation coefficients were calculated to identify if multiple nutrient HSs commonly occurred together in space using the rcorr function within the R corrplot package (Wei and Simko 2017). Hot spot presence was coded as a binary response variable (no HS was coded as 0; HS as 1) and data from each plot were again analyzed separately.

The contribution of HSs or bulk matrix soil (i.e., not a HS) to the total nutrient flux across all depths was also evaluated for a given nutrient in each sampling period and plot separately, using Eq. 2:

$$\frac{{\sum Sampling Period Nutrient Flux}_{HS or Matrix Soil}}{{\sum Samping Period Nutrient Fluxes}_{HS and Matrix Soil}} \times 100\%$$

A one-way analysis of variance was run to identify if the flux from HSs was significantly different than the matrix soil. All statistical analyses were carried out using R (version 4.2.3; R Development Core Team 2008).

Meteorological data

Precipitation greatly varied greatly across sampling periods. During the WY 2012 and 2016, resin capsules were exposed to 121 and 113 mm of precipitation from the first major event in the fall, and 874 and 1104 mm of precipitation after the spring snowmelt, respectively (Fig. S2). This contrasts with water years 2013 and 2014 that had 95 and 172 mm of precipitation in the fall, but only 807 and 529 mm in the spring. Overall, WYs 2013 and 2014 were approximately 1.3 and 1.7 times lower in annual precipitation than 2016, respectively.

Changing precipitation patterns over the sampling periods corresponded to distinct patterns in soil volumetric water contents over time and with depth (Fig. S2). Volumetric water content followed precipitation events over the course of the year, and was always higher in the spring compared to the fall across all depths. For example, the 60-cm depth was approximately 0.04 m³ m^− 3 and 0.03 m³ m^− 3 drier in the fall and spring, respectively, in WY 2013 and 2014 compared to WY 2012 and 2016. Soil water content tended to be similar across depths with notable differences in the shallow 10-cm depth which had the lowest moisture content over the summer.

Nutrient flux distributions

All nutrient distributions were right skewed, causing the mean flux values to be greater than the median (Table 1; Fig. 2). Sodium exhibited the lowest median fluxes (Plot 1: 0.06 nmol cm^-2 d^-1; Plot 2: 0.08 nmol cm^-2 d^-1), but had a similar skew as the other nutrients (Table 1). Ammonium (Plot 1: 0.15 nmol cm^-2 d^-1; Plot 2: 0.18 nmol cm^-2 d^-1) and NO₃^- (Plot 1: 0.14 nmol cm^-2 d^-1; Plot 2: 0.14 nmol cm^-2 d^-1) exhibited similar median fluxes, but NH₄⁺ exhibited a greater skew in Plot 2 (17.6) than Plot 1 (5.13). Phosphate (Plot 1: 0.46 nmol cm^-2 d^-1; Plot 2: 0.23 nmol cm^-2 d^-1) and Mg²⁺ (Plot 1: 0.93 nmol cm^-2 d^-1; Plot 2: 0.65 nmol cm^-2 d^-1) had higher median fluxes than Na⁺, NH₄⁺, and NO₃^- with moderate skew (4.39 to 8.52). Calcium had the highest median nutrient flux (Plot 1: 4.69 nmol cm^-2 d^-1; Plot 2: 3.48 nmol cm^-2 d^-1) and tended to have a relatively lower skew than other nutrients within the same plot (Plot 1: 3.18; Plot 2: 7.43). Regardless of the median flux, all nutrients were positively skewed, with the extent of skew variable between plots and among nutrients.

Nutrient HS occurrence

HSs formed most often for NO₃⁻ (Plot 1 = 53; Plot 2 = 104) and Na⁺ (Plot 1 = 54; Plot 2 = 90) across all depths and sampling periods (Plot 1: n = 360, Plot 2: n = 504; Fig. 3). All other nutrients had a similar number of total HS, with 36 to 45 forming in Plot 1 and 56 to 66 in Plot 2.

All nutrients formed HSs repeatedly at the same spatial location in more than one sampling period, but HSs did not occur in all locations (Fig. 4). Nitrate had the greatest number of reoccurring HSs in Plot 1 (15 resin locations), while HS reoccurrence for Na⁺ was greatest in Plot 2 (23 resin locations). Although HSs did reoccur for Ca²⁺, Mg², and Na⁺, they tended to only happen twice across all sampling periods, whereas some resin locations formed HSs for more than two sampling periods for PO₄³⁻, NO₃⁻, and NH₄⁺. For all nutrients, HS reoccurrence was more common at the 0-cm depth than at deeper mineral soil locations. Although HSs often occurred in the same spatial location repeatedly for all nutrients, HSs occasionally persisted over more than two sampling periods for PO₄³⁻, NO₃⁻, and NH₄⁺, and were more transient for Na⁺, Ca²⁺, and Mg²⁺.

Table 2 Results of mixed-effect models for Plot 1 and Plot 2 on hot spot likelihood (binary response) of PO₄^3-, NH₄⁺, NO₃^-, Ca²⁺, Mg²⁺, and Na⁺as effected by the fixed effects of water year (categorical, 2012 coded as 0), season (categorical, fall coded as 0), depth (continuous), and season by depth interaction. Model coefficients were determined for each fixed effect using maximum likelihood (Laplace approximation), with standard errors in parentheses. Significant (p < 0.05) results are shown in bold. Individual (repeated sampling from same location; n = 61 for Plot 1, n = 63 for Plot 2) and vertically coincident (spatial dependency of 0-, 20-, 40-, and 60-cm; n = 16) sampling locations were included as random effects. The variance is reported for each random effect with standard deviation in parentheses. Data for WY2016 were not collected for Plot 1; therefore, this factor was not included in the model and is denoted within the table as no data (ND). Water Year 2013 was not included in models for Ca²⁺ and Mg²⁺ in Plot 1 and Ca²⁺ in Plot 2 due to singularity and large eigenvalue ratios created by the lack of response variables (i.e., no hot spots), and are marked as “Not Included.” Model R² values are reported for the fixed effects only (i.e., marginal) and for the fixed and random effects together (i.e., conditional)

Relationship of water year, season, and depth with HS formation

Water year was a significant predictor of HS occurrence, particularly for Plot 2 (Tables 2 and 3; Fig. 3). Hot spots were more likely to occur for Ca²⁺, Mg²⁺, and Na⁺ in WY 2012 and WY 2016 than WY 2013 and WY 2014 at both plots (WY 2013 was not included in the model for Ca²⁺ and Mg²⁺ at Plot 1 or Ca²⁺ at Plot 1 and 2 due to singularity; no HSs occurred for these nutrients in WY 2013; Table 3). In contrast, HS occurrence for PO₄³⁻ and NO₃⁻ was not significantly different in 2012, 2013, or 2014, and NH₄⁺ HSs formed more often in 2013 than 2012. Hot spot formation was more likely in 2016 than the other water years for PO₄³⁻, NO₃⁻, and NH₄⁺. Overall, Ca²⁺, Mg²⁺, and Na⁺ HSs exhibited more variation among water years than the other nutrients.

Season was occasionally a significant predictor of HS occurrence (Table 2). In Plot 1, PO₄³⁻, NO₃⁻, NH₄⁺, and Na⁺ HSs were greater in the fall after the first major precipitation event compared to following spring snowmelt. Season was only significant in Plot 2 for NH₄⁺; in contrast, HS formation in Plot 1 was more likely in the fall. Calcium and Mg²⁺ HS formation was not impacted by season.

The likelihood of the occurrence of HSs decreased with depth into the soil profile for all nutrients and in both plots (Table 2). The interaction of season and depth was significant for predicting HS occurrence for Ca²⁺ in both plots and Na⁺ at Plot 2 (Tables 2 and 3). Calcium HSs were more likely to occur deeper in the soil profile after snowmelt in the spring (Table 3). In contrast, Na⁺ HSs persisted deeper into the mineral soil after the first major precipitation event in the fall (Table 3).

Table 3 Post-hoc results for hot spot mixed-effect models of PO₄^3-, NH₄⁺, NO₃^-, Ca²⁺, Mg²⁺, and Na⁺. Coefficients (standard error) are presented for the fixed effects of water year and season by depth interaction. The season by depth interaction compares the respective depth (0-, 20-, 40-, and 60-cm) in the fall compared to the spring season. P values are adjusted with Tukey’s method and are shown in bold when significant at p < 0.05. Water Year 2016 was not collected for Plot 1 (ND = no data). Water Year 2013 was not included in models for Ca²⁺ and Mg²⁺ in Plot 1 and Ca²⁺ in Plot 2 due to singularity and large eigenvalue ratios created by the lack of response variables (i.e., no hot spots) and are marked as “Not Included.” Results are only presented for significant independent variables from the original mixed-effects model presented in Table 2 (i.e., blank cells indicate p > 0.05)

Nutrient HS covariance

Many spatial locations exhibited a HS for more than one nutrient within a given location and sampling period (Fig. 5). Indeed, all bi-nutrient correlation coefficients were positive and statistically significant. A HS occurred for two nutrients simultaneously 61 times in Plot 1 and 97 times in Plot 2. It was rare for a HS to form for all six nutrients within the same sampling period (i.e., 7 locations in Plot 1 and 13 in Plot 2). Calcium and Mg²⁺ had the highest correlation coefficient (Plot 1: ρ = 0.71; Plot 2: ρ = 0.72). Phosphate, NH₄⁺, and NO₃⁻ had the next highest correlation coefficients between each other (ρ = 0.59 to 0.68) in both plots.

Contribution of HSs to total nutrient flux

Generally, HSs represented a small volume of the soil. Hot spots represented 8 to 17% of the total sampling volume when considering all sampling events. Sodium HSs occurred in a larger volume compared to the other nutrients (Plot 1: 14%, Plot 2: 17%). Nitrate HSs only occupied 12% of the sampled volume in Plot 1 but was 16% of the sampling volume in Plot 2. Phosphate HSs occurred in the smallest volume, occupying only 8% of the sampling volume in Plot 1 and 9% in Plot 2.

The percent of the total nutrient flux attributed to HSs varied by nutrient when averaged across all sampling periods (Fig. 6). The contribution of NO₃⁻ HSs to the total NO₃⁻ flux was significantly greater than the contribution from the matrix soil, with HSs contributing 77.9 ± 17.9% (standard error) and 87.9 ± 11.1% of the total NO₃⁻ fluxes in Plot 1 and 2, respectively. Ammonium (Plot 1: 62.2 ± 24.7%; Plot 2: 73.5 ± 21.3%) and PO₄³⁻ (Plot 1: 61.4 ± 26.8; Plot 2: 56.4 ± 26.1) HS contributions to their respective total nutrient fluxes followed a similar pattern as NO₃⁻, with HSs representing most of the nutrient flux (although this was only statistically significantly in Plot 2 for NH₄⁺). Total Na⁺ fluxes were composed of approximately half from HSs (Plot 1: 46.7 ± 43.8; Plot 2: 56.6 ± 45.7) and half from the soil matrix. In contrast, Ca²⁺ HSs (Plot 1: 26.8 ± 36.8; Plot 2: 33.0 ± 40.0) and Mg²⁺ HSs (Plot 1: 28.7 ± 34.9; Plot 2: 39.6 ± 40.2) accounted for less than half of the total nutrient flux, with HS contribution for Ca²⁺ in Plot 1 being significantly less. Overall, NO₃⁻, NH₄⁺, PO₄³⁻, and Na⁺ HSs contributed to a relatively larger portion of the total nutrient flux compared to Ca²⁺ and Mg²⁺.

The median isn’t the message

In his personal essay titled, “The Median Isn’t the Message,” Stephen Jay Gould notes we often think about means and medians as “hard realities” and the variation in values around them “a set of transient and imperfect measurements of this hidden essence” (Gould 1991). In Gould’s case, the median life expectancy of his mesothelioma diagnosis was 8 months, although he lived for another 20 years: the skewness of the distribution of life expectancies for individuals with that type of cancer mattered to his expected longevity. A similar principle applies to nutrient fluxes; when we distill datasets down to their mean or median, we discount the extreme values, the nutrient “cold spots” or HSs. Indeed, in many biogeochemical and other scientific studies, efforts are made to remove these outliers from the dataset. In our study, we found spatially and temporally heterogeneous fluxes for all nutrients examined, resulted in right-skewed distributions with maximum values that greatly exceeded the mean or median (Fig. 2, Table 1). Just like Stephen J. Gould’s prognosis, the median did not represent well the population of nutrient fluxes in soil.

Understanding the temporal and spatial controls that produce locations of elevated nutrients is critical to advance our understanding of HS phenomena. Although HSs are now well-recognized by the biogeochemical community, incorporating HS phenomena into predictive biogeochemical models remains challenging (Landsberg et al. 1991; Groffman et al. 2009; Arora et al. 2022). Most ecological models use nutrient averages for spatial and temporal scaling (Strayer et al. 2003), discounting outliers found in soil microsites (Schimel and Bennett 2004), and are often not spatially explicit (Landsberg 2003) or do not consider multiple nutrient HSs with varying concentrations (i.e., Ryel and Caldwell 1998). Although there has been some recent progress in this area (Arora et al. 2022), we need to use sampling designs that allow for their explicit characterization to advance our mechanistic understanding of HS formation and subsequent inclusion into models. Our study advances knowledge of nutrient HS phenomena by using an experimental design that allowed us to elucidate multi-nutrient HS occurrence at different spatial locations and temporal scales.

Influence of yearly drought on HS formation

Hot spot formation was impacted by water year, especially for Ca²⁺, Mg²⁺, and Na⁺, likely because annual precipitation exhibited large variation across sampling periods (ranged from 701 to 1216 mm). A multi-year drought occurred during our study (Bales et al. 2018), resulting in total precipitation to be higher in WYs 2012 (996 mm y^− 1) and 2016 (1216 mm y^− 1) compared to 2013 (902 mm y^− 1) and 2014 (701 mm y^− 1; Fig. S1). Calcium, Mg²⁺, and Na⁺ fluxes and HS formation were greatest during WY 2012 and 2016 and substantially lower in the extreme drought years (Fig. 3 and S3, Tables 2, 3, S2 and S3). Variable water fluxes across drought and non-drought years likely impacted not only nutrient transport through the soil, but also altered the abundance of these nutrients by changing the rates of chemical weathering and cation exchange reactions (Johnson et al. 1968; Gislason et al. 2009). In contrast, PO₄³, NH₄⁺, and NO₃⁻ exhibited less variability in HS formation (Fig. 3) from WYs 2012 through 2014 (Tables 2 and 3), even though these nutrient fluxes were also altered by water year (Fig. S3, Table S2 and S3). This suggests that labile C, water, or other limiting resources were accessible to microorganisms, even during drought years, to promote nutrient mineralization and nitrification, and water fluxes were high enough to transport these ions to the resin capsules across all sampling years. There was a notable increase in HSs for all nutrients in WY 2016, which followed the extreme drought WYs (2013–2015). Nutrients likely accumulated during the extreme drought years and were subsequently mobilized to the ion exchange resin capsules during the following wet year (WY 2016). Water fluxes were likely too inconsistent during our study period to generate HSs repeatedly in the same location over time, making HSs mostly transient rather than persisting in the same location over time.

Influence of seasonal drought on HS formation

Although PO₄³⁻, NH₄⁺, and NO₃⁻ HSs were relatively insensitive to drought on an annual basis, they were responsive to seasonal drought conditions. We found nutrient fluxes and HSs for PO₄³⁻, NH₄⁺, NO₃⁻, and Na⁺ to be higher after the first fall rains compared to spring snowmelt (Tables 2 and S2, Fig. 3 and S3). Nutrients are known to accumulate during the hot, dry summers that characterize Mediterranean climates (Davidson et al. 1990; Lewis et al. 2006). Previous research in the Sierra Nevada found O horizon interflow had higher nutrient concentrations compared to underlying mineral soil solution and stream water, although these differences were highly variable (Woodward 2012; Miller et al. 2005; Johnson et al. 2011). When the first major precipitation event in the fall occurred, O horizon interflow likely became nutrient-rich (Miller et al. 2005, 2006; Johnson et al. 2011) and infiltrated mineral soil via preferential flow paths (Burcar et al. 1994; Bundt et al. 2001) to form HSs.

HS structure with depth

Previous studies have largely focused on characterizing HS occurrence in surficial soil rather than evaluating their formation with depth into the soil. We found HS spot formation was depth-dependent within the soil profile. The occurrence of HSs for all nutrients significantly decreased with depth into the soil (Fig. 3; Table 2) likely because microbial activity, nutrient substrate availability, and water fluxes were highest in the upper soil profile (Jobbágy and Jackson 2001; Sun et al. 2021). For Ca²⁺, this was driven mostly by HSs that formed deeper in the mineral soil after snowmelt in the spring (Marsh and Pomeroy 1999). Sodium differs from the other nutrients studied because it is not an essential element for plant and microbial growth, and therefore it does not accumulate within their biomass to a significant degree, resulting in Na⁺ acting like a semi-conservative tracer of water flow (Vitousek and Reiners 1975). This characteristic likely contributed to Na⁺ having the greatest number of HSs below 0-cm depth out of all nutrients (Fig. 3).

In the previous one-year study at this site using the same sampling devices and locations, PO₄³⁻, NH₄⁺, and NO₃⁻ HSs were also concentrated in the upper sampling depths, whereas Ca²⁺, Mg²⁺, and Na⁺ HSs were identified at greater depths (Johnson et al. 2014). However, all nutrient HSs tended to persist more at depth in the Johnson et al. (2014) study, possibly because statistical analyses were run without the inclusion of the 0-cm depth resin capsule data. When we ran the same mixed-effect models without the data from the 0-cm depth, all nutrient fluxes except Na⁺ still decreased with depth (data not shown). Hot Spot likelihood also decreased with depth at Plot 2 for PO₄³⁻, NH₄⁺, NO₃⁻, and Mg²⁺, although Plot 1 nutrient HSs were no longer significantly impacted by depth. This result highlights that the observed depth trends are not strictly regulated by elevated HS formation at the interface of organic and mineral soil horizons. The dissipation of nutrient HSs with depth (especially the growth-limiting nutrients PO₄³⁻, NH₄⁺, and NO₃⁻; Vitousek and Reiners 1975) may be due to plant root uptake because HS occurrence has a negative relationship with rooting patterns in this Mediterranean forest (Hart and Firestone 1991). Globally, nutrients required by plants in greater amounts are generally found in the upper profile, while others such as Na⁺ persist deeper in the profile (Jobbágy and Jackson 2001). Therefore, it is not surprising that nutrient HSs follow these broad-scale nutrient patterns.

Influence of HSs on biogeochemical structure

By definition, HSs are found in select spatial compartments, therefore they necessarily occupy a relatively small percentage of the total soil volume. We found that, although PO₄³⁻, NH₄⁺, and NO₃⁻ HSs only occurred in 8–17% of the total area sampled, HSs were responsible for 56–88% of the total nutrient flux on average (Fig. 6). This finding is similar to that observed previously in surficial soil (3-cm depth) in an alpine ecosystem where HSs only accounted for 14% of the study area but provided more than 50% of the bioavailable inorganic N (Darrouzet-Nardi and Bowman 2011). Additionally, a meta-analysis found rhizosphere HSs accounted for 10–33% of C and N net mineralization while only occupying 8–26% of the soil volume to a 10-cm depth (Finzi et al. 2015). In our study, Ca²⁺ and Mg²⁺ HSs accounted for a similar amount of the sampling area as PO₄³⁻, NH₄⁺, and NO₃⁻, however the proportion of the total nutrient flux attributed to HSs was relatively smaller (an average of 27–40 %). Therefore, groth-limiting nutrient HSs, especially PO₄³⁻, NH₄⁺, and NO₃⁻, may play a disproportionate role in regulating ecosystem processes and function compared to the volume of soil they represent.

Hot Spot behavior tended to cluster into groups of nutrients that have similar biogeochemical cycles. This is likely because nutrients primarily under biological control (PO₄³⁻, NH₄⁺, and NO₃⁻) require high substrate availability and high microbial activity to co-occur with active hydrological flowpaths to form HSs (McClain et al. 2003). In contrast, soil nutrient fluxes that are primarily under geochemical control (i.e., chemical weathering, ion exchange reactions; Ca²⁺, Mg²⁺, and Na⁺) largely require only the co-occurrence of substrate abundance and water flow, and are less dependent on high rates of microbial activity. These drivers likely contributed to biologically versus geochemically controlled nutrient HSs to have: dissimilar responses to yearly and seasonal drought (Tables 2 and 3), contrasting patterns in reoccurrence (Fig. 4), differences in the tendency for nutrient HSs to co-occur within respective groups at the same spatial locations (Fig. 5), and biologically controlled nutrient HSs to contribute to a larger portion of the total flux compared to geochemically controlled nutrients (Fig. 6).

Plot characteristics

Hot spot patterns demonstrated variability between plots that is likely due to a variety of environmental differences between them, including soil bulk density (related to soil strength), slope degree, and vegetation density. Generally, Plot 2 had higher soil strength than Plot 1, suggesting that pore connectivity was higher in Plot 2 (Lipiec and Hatano 2003). Additionally, Plot 2 was on a shallower slope than Plot 1 (Plot 1: 20%, Plot 2: 5%), which may have resulted in greater water infiltration in Plot 2 by reducing overland flow. Plot 2 had greater vegetation density than Plot 1, which likely resulted in more homogeneous substrate distribution. Taken together, these factors likely resulted in HS formation to be less seasonally dependent (Table 2) and created more co-occurring HS across all nutrients in Plot 2 compared to Plot 1 (Fig. 5). Given the inter-plot variability, identifying easily measured variables that co-vary with HS formation is needed for model incorporation.

HS ecological significance

The transient nature of HSs likely plays an important role in nutrient competition and utilization between plants and microbes as well as among individual plants. Although HSs can be areas of intense competition for available nutrients between vegetation and soil microorganisms (Jingguo and Bakken 1997), especially N (Kaye and Hart 1997), the extremely dry summers in Mediterranean-type climates limits root development in the surficial O horizon (Hart and Firestone 1991); therefore, soil microorganism likely obtain nutrients from the organic layer with minimal plant competition until leaching occurs into the mineral soil (Johnson et al. 2011). We found that reoccurrence, or duration, of nutrient HSs in soil tended to be more frequent in surficial soil than deeper in the soil profile (Fig. 4). Fungal species with vast mycelial networks (i.e., filamentous fungi) can likely exploit these reoccurring HSs by actively foraging for the nutrients contained in these locations; in contrast, more stationary microorganisms (i.e., bacteria and archaea) attached primarily to surfaces or held within soil aggregates are less likely to utilize nutrients contained within these locations as they are more dependent on diffusion and mass flow processes for nutrient uptake (Stark and Hart 1999).

The transient nature of HSs benefits some vegetation species over others. Reoccurring HSs of PO₄³⁻, NH₄⁺, or NO₃⁻ may stimulate a morphological response of the belowground plant biomass (Robinson 1994; Hutchings et al. 2003; Wang et al. 2018). However, HSs of all nutrients were more commonly transient in nature and thus are more likely to elicit a physiological response in vegetation (Fransen et al. 1999). We hypothesize that plants with low morphological plasticity, but an extensive rooting system that can exhibit physiological plasticity, would likely benefit most from transient HSs (Crick and Grime 1987; Cui and Caldwell 1997; Wang et al. 2018), while reoccurring HSs would be more advantageous for species with high morphological plasticity to exploit these persistent elevated nutrient patches (Jackson and Caldwell 1989; Hutchings and Kroon 1994). Although microorganisms may outcompete vegetation in the short-term for transient nutrient HSs, plants that access reoccurring HS with their roots and mycorrhizal symbionts can acquire more nutrients over the long-term (Cui and Caldwell 1997; Hodge et al. 2000; Schimel and Bennett 2004).

The response of plants to nutrient heterogeneity is impacted by the concentration and chemical composition of the nutrient patch, the nutrient distribution within the soil profile, the duration of the nutrient pulse, and the frequency it occurs (Jackson and Caldwell, 1989; Bilbrough and Caldwell 1997; Jingguo and Bakken 1997; Fransen et al. 1999; Wijesinghe et al. 2001). Because HSs have extremely high fluxes compared to the soil matrix, are composed of co-varying macronutrients, are found at soil depths relevant to roots, persist over time in some locations, and are responsible for a large portion of the total nutrient flux, they likely play a disproportionate role in supporting ecosystem structure and function.

Funding: This research was supported by the National Science Foundation through the Southern Sierra Critical Zone Observatory (SSCZO; EAR-0725097, 1239521, and 1331939), Environmental Systems Graduate Group at the University of California, Merced, and by the U.S. Department of Energy, Office of Biological and Environmental Research, Environmental System Science program through the River Corridor Scientific Focus Area project at the Pacific Northwest National Laboratory.

Competing Interests: The authors have no relevant financial or non-financial interests to disclose.

Author Contributions: DWJ conceived and designed the study; SCH, MEB, and DWJ performed the research; and MEB and SCH analyzed the data. MEB was the lead writer for the manuscript, with notable contributions from SCH and DWJ. All authors provided critical manuscript revisions.

Data Availability: The datasets generated during this study are available in the Dryad repository, https://doi.org/10.6071/M3QQ3Z

Arora B, Briggs MA, Zarnetske JP et al (2022) Biogeochemistry of the Critical Zone. Adv Crit Zone Sci 9–47. https://doi.org/10.1007/978-3-030-95921-0_2
Bales RC, Goulden ML, Hunsaker CT et al (2018) Mechanisms controlling the impact of multi-year drought on mountain hydrology. Sci Rep 8:701. https://doi.org/10.1038/s41598-017-19007-0
Barcellos D, O’Connell C, Silver W et al (2018) Hot spots and hot moments of soil moisture explain fluctuations in iron and carbon cycling in a humid tropical forest soil. Soil Syst 2:59. https://doi.org/10.3390/soilsystems2040059
Barrat HA, Evans J, Chadwick DR et al (2021) The impact of drought and rewetting on N2O emissions from soil in temperate and Mediterranean climates. Eur J Soil Sci 72:2504–2516. https://doi.org/10.1111/ejss.13015
Barton K (2009) Package ‘MuMIn’. R package version 1.40.4. https://CRAN.R-project.org/package=MuMIn
Bates D, Maechler M, Bolker B, Walker S (2015) Fitting Linear Mixed-Effects Models Using lme4. J Stat Softw 67:1–48. https://doi.org/10.18637/jss.v067.i01
Bernhardt ES, Blaszczak JR, Ficken CD et al (2017) Control Points in Ecosystems: Moving Beyond the Hot Spot Hot Moment Concept. Ecosystems 20:665–682. https://doi.org/10.1007/s10021-016-0103-y
Bilbrough CJ, Caldwell MM (1997) Exploitation of springtime ephemeral N pulses by six Great Basin plant species. Ecology 78:231–243. https://doi.org/10.1890/0012-9658(1997)078[0231:eosenp]2.0.co;2
Bilyera N, Kuzyakova I, Guber A et al (2020) How “hot” are hotspots: Statistically localizing the high-activity areas on soil and rhizosphere images. Rhizosphere 16:100259. https://doi.org/10.1016/j.rhisph.2020.100259
Binkley D (1984) Ion Exchange Resin Bags: Factors Affecting Estimates of Nitrogen Availability 1. Soil Sci Soc Am J 48:1181–1184. https://doi.org/10.2136/sssaj1984.03615995004800050046x
Bruckner A, Kandeler E, Kampichler C (1999) Plot-scale spatial patterns of soil water content, pH, substrate-induced respiration and N mineralization in a temperate coniferous forest. Geoderma 93:207–223. https://doi.org/10.1016/S0016-7061(99)00059-2
Bundt M, Widmer F, Pesaro M et al (2001) Preferential flow paths: biological “hot spots” in soils. Soil Biol Biochem 33. https://doi.org/10.1016/s0038-0717(00)00218-2
Burcar S, Miller WW, Johnson DW, Tyler SW (1994) Seasonal Preferential Flow in Two Sierra Nevada Soils under Forested and Meadow Cover. Soil Sci Soc Am J 58:1555–1561. https://doi.org/10.2136/sssaj1994.03615995005800050040x
Campo J, Maass JM, Jaramillo VJ, Yrízar AM (2000) Calcium, potassium, and magnesium cycling in a Mexican tropical dry forest ecosystem. Biogeochemistry 49:21–36. https://doi.org/10.1023/a:1006207319622
Charley JL, West NE (1977) Micro-patterns of nitrogen mineralization activity in soils of some shrub-dominated semi-desert ecosystems of Utah. Soil Biol Biochem 9:357–365. https://doi.org/10.1016/0038-0717(77)90010-4
Chen J, Arora B, Bellin A, Rubin Y (2021) Statistical characterization of environmental hot spots and hot moments and applications in groundwater hydrology. Hydrol Earth Syst Sc 25:4127–4146. https://doi.org/10.5194/hess-25-4127-2021
Crick JC, Grime JP (1987) Morphological plasticity and mineral nutrient capture in two herbaceous species of contrasted ecology. New Phytol 107. https://doi.org/10.1111/j.1469-8137.1987.tb00192.x. :403 – 14
Cui M, Caldwell MM (1997) A large ephemeral release of nitrogen upon wetting of dry soil and corresponding root responses in the field. Plant Soil 191:291–299. https://doi.org/10.1023/a:1004290705961
Darrouzet-Nardi A, Bowman WD (2011) Hot spots of inorganic nitrogen availability in an alpine-subalpine ecosystem. Colo Front Range Ecosyst 14:848–863. https://doi.org/10.1007/s10021-011-9450-x
Davidson EA, Stark JM, Firestone MK (1990) Microbial production and consumption of nitrate in an annual grassland. Ecology 71:1968–1975. https://doi.org/10.2307/1937605
Day KJ, Hutchings MJ, John EA (2003) The effects of spatial pattern of nutrient supply on yield, structure and mortality in plant populations. J Ecol 91:541–553. https://doi.org/10.1046/j.1365-2745.2003.00799.x
Dobermann A, Langner H, Mutscher H et al (1994) Nutrient adsorption kinetics of ion exchange resin capsules: A study with soils of international origin. Commun Soil Sci Plan 25:1329–1353. https://doi.org/10.1080/00103629409369119
Farley RA, Fitter AH (2001) Temporal and spatial variation in soil resources in a deciduous woodland. J Ecol 87:688–696. https://doi.org/10.1046/j.1365-2745.1999.00390.x
Finzi AC, Abramoff RZ, Spiller KS et al (2015) Rhizosphere processes are quantitatively important components of terrestrial carbon and nutrient cycles. Glob Change Biol 21:2082–2094. https://doi.org/10.1111/gcb.12816
Fox J, Weisberg S (2019) An R Companion to Applied Regression. Third edition. Sage, USA
Franklin SM, Kravchenko AN, Vargas R et al (2021) The unexplored role of preferential flow in soil carbon dynamics. Soil Biology Biochem 161:108398. https://doi.org/10.1016/j.soilbio.2021.108398
Fransen B, Blijjenberg J, de Kroon H (1999) Root morphological and physiological plasticity of perennial grass species and the exploitation of spatial and temporal heterogeneous nutrient patches. Plant Soil 211:179–189. https://doi.org/10.1023/a:1004684701993
Frei S, Knorr KH, Peiffer S, Fleckenstein JH (2012) Surface micro-topography causes hot spots of biogeochemical activity in wetland systems: A virtual modeling experiment. J Geophys Research: Biogeosciences 117:1–18. https://doi.org/10.1029/2012JG002012
Gibson DJ (1986) Spatial and temporal heterogeneity in soil nutrient supply measured using in situ ion-exchange resin bags. Plant Soil 96:445–450. https://doi.org/10.1007/bf02375151
Gislason SR, Oelkers EH, Eiriksdottie ES et al (2009) Direct evidence of the feedback between climate and weathering. Earth Planet Sci Lett 277:213–222. https://doi.org/10.1016/j.epsl.2008.10.018
Göransson H, Edwards PJ, Perreijn K et al (2014) Rocks create nitrogen hotspots and N:P heterogeneity by funnelling rain. Biogeochemistry 121:329–338. https://doi.org/10.1007/s10533-014-0031-x
Gould J (1991) Bully for Brontosaurus: Reflections in Natural History. W.W. Norton & Company, Inc., New York
Groffman PM, Butterbach-Bahl K, Fulweiler RW et al (2009) Challenges to incorporating spatially and temporally explicit phenomena (hotspots and hot moments) in denitrification models. Biogeochemistry 93:49–77. https://doi.org/10.1007/s10533-008-9277-5
Hadley W, Francois R, Henry L, Muller K (2018) dplyr: A Grammar of Data Manipulation. R package version 1.0.0. https://CRAN.R-project.org/package=dplyr
Halvorson JJ, Bolton H, Smith JL, Rossi RE (1994) Geostatistical analysis of resource islands under Artemisia tridentata in the shrub-steppe. The Great Basin Naturalist 54:313–328
Harms TK, Grimm NB (2008) Hot spots and hot moments of carbon and nitrogen dynamics in a semiarid riparian zone. J Geophys Research: Atmos 113:G1. https://doi.org/10.1029/2007jg000588
Hart SC, Binkley D (1984) Colorimetric interference and recovery of adsorbed ions from ion exchange resins. Commun Soil Sci Plant Anal 15:893–902. https://doi.org/10.1080/00103628409367528
Hart SC, Firestone MK (1991) Forest floor-mineral soil interactions in the internal nitrogen cycle of an old-growth forest. Biogeochemistry 12:103–127. https://doi.org/10.1007/bf00001809
Hoang DTT, Pausch J, Razavi BS et al (2016) Hotspots of microbial activity induced by earthworm burrows, old root channels, and their combination in subsoil. Biol Fertil Soils 52:1105–1119. https://doi.org/10.1007/s00374-016-1148-y
Hodge A, Robinson D, Griffiths BS, Fitter AH (1999) Why plants bother: root proliferation results in increased nitrogen capture from an organic patch when two grasses compete.Plant, Cell & Environment 22:811–820. https://doi.org/10.1046/j.1365-3040.1999.00454.x
Hodge A, Stewart J, Robinson D et al (2000) Competition between roots and soil micro-organisms for nutrients from nitrogen‐rich patches of varying complexity. J Ecol 88:150–164. https://doi.org/10.1046/j.1365-2745.2000.00434.x
Holloway JM, Dahlgren RA (2001) Seasonal and event-scale variations in solute chemistry for four Sierra Nevada catchments. J Hydrol 250:106–121. https://doi.org/10.1016/s0022-1694(01)00424-3
Homyak PM, Sickman JO, Melack JM (2014) Pools, transformations, and sources of P in high-elevation soils: Implications for nutrient transfer to Sierra Nevada lakes. Geoderma 217:65–73. https://reader.elsevier.com/reader/sd/pii/S0016706113004023?token=04252640CDD8AB860C35595A566889031F645375B8A401D787410531098EBDFE4998C64A76D1821996FE236181B158D1
Hunsaker CT, Johnson DW (2017) Concentration-discharge relationships in headwater streams of the Sierra Nevada, California. Water Resour Res 53:7869–7884. https://doi.org/10.1002/2016wr019693
Hunsaker CT, Safeeq M (2018) Kings River Experimental Watersheds meteorology data. Fort Collins, CO: Forest Service Research Data Archive. https//doi.org/10.2737/RDS-2018-0028
Hutchings MJ, John EA, Wijesinghe DK (2003) Toward understanding the consequences of soil heterogeneity for plant populations and communities. Ecology 84:2322–2334. https://doi.org/10.1890/02-0290
Hutchings MJ, Kroon HD (1994) Foraging in Plants: the Role of Morphological Plasticity in Resource Acquisition. Adv Ecol Res 25:159–238. https://doi.org/10.1016/s0065-2504(08)60215-9
Jackson RB, Caldwell MM (1989) The timing and degree of root proliferation in fertile-soil microsites for three cold-desert perennials. Oecologia 81:149–153. https://doi.org/10.1007/BF00379798
Jingguo W, Bakken LR (1997) Competition for nitrogen during decomposition of plant residues in soil: Effect of spatial placement of N-rich and N-poor plant residues. Soil Biol Biochem 29:153–162. https://doi.org/10.1016/s0038-0717(96)00291-x
Jobbágy EG, Jackson RB (2001) The distribution of soil nutrients with depth: Global patterns and the imprint of plants. Biogeochemistry 53:51–77. https://doi.org/10.1023/a:1010760720215
Johnson DW, Glass DW, Murphy JD, Stein CM, Miller WW (2010) Nutrient hot spots in some Sierra Nevada forest soils. Biogeochemistry 101:93–103. https://doi.org/10.1007/s10533-010-9423-8
Johnson DW, Miller WW, Rau BM, Meadows MW (2011) The Nature and Potential Causes of Nutrient Hotspots in a Sierra Nevada Forest Soil. Soil Sci 176:596–610. https://doi.org/10.1097/ss.0b013e31823120a2
Johnson DW, Woodward C, Meadows MW (2014) A Three-Dimensional View of Nutrient Hotspots in a Sierra Nevada Forest Soil. Soil Sci Soc Am J 78:S225–S236. https://doi.org/10.2136/sssaj2013.08.0348nafsc
Johnson NM, Likens GE, Bormann FH, Pierce RS (1968) Rate of chemical weathering of silicate minerals in New Hampshire. Geochim Cosmochim Acta 32:531–545. https://doi.org/10.1016/0016-7037(68)90044-6
Kaye JP, Hart SC (1997) Competition for nitrogen between plants and soil microorganisms. Trends Ecol Evol 12:139–143. https://doi.org/10.1016/s0169-5347(97)01001-x
Kleb HR, Wilson SD (1997) Vegetation Effects on Soil Resource Heterogeneity in Prairie And Forest. Am Nat 150:283–298. https://doi.org/10.1086/286066
Kratz AM, Maier S, Weber J et al (2022) Reactive Nitrogen Hotspots Related to Microscale Heterogeneity in Biological Soil Crusts. Environ Sci Technol 56:11865–11877. https://doi.org/10.1021/acs.est.2c02207
Kurunc A, Ersahin S, Uz BY et al (2011) Identification of nitrate leaching hot spots in a large area with contrasting soil texture and management. Agric Water Manage 98:1013–1019. https://doi.org/10.1016/j.agwat.2011.01.010
Kuzyakov Y, Blagodatskaya E (2015) Microbial hotspots and hot moments in soil: Concept & review. Soil Biol Biochem 83:184–199. https://doi.org/10.1016/j.soilbio.2015.01.025
Landsberg J (2003) Modelling forest ecosystems: state of the art, challenges, and future directions. Can J For Res 33:385–397. https://doi.org/10.1139/x02-129
Landsberg JJ, Kaufmann MR, Binkley D et al (1991) Evaluating progress toward closed forest models based on fluxes of carbon, water and nutrients. Tree Physiol 9:1–15. https://doi.org/10.1093/treephys/9.1-2.1
Lenth R, Singmann H, Love J (2018) Emmeans: Estimated marginal means, aka least-squares means. R package version 1.1.3. https://cran.r-project.org/web/packages/emmeans
Lewis DJ, Singer MJ, Dahlgren RA, Tate KW (2006) Nitrate and sediment fluxes from a California rangeland watershed. J Environ Qual 35:2202–2211. https://doi.org/10.2134/jeq2006.0042
Lipiec J, Hatank R (2003) Quantification of compaction effects on soil physical properties and crop growth. Geoderma 116:107–136. https://doi.org/10.1016/S0016-7061(03)00097-1
Lovett GM, Jones CG, Turner MG, Weathers KC (2005) Ecosystem Function in Heterogeneous Landscapes. Springer, New York, NY. https://doi.org/10.1007/0-387-24091-8_1
Ma X, Razavi BS, Holz M et al (2017) Warming increases hotspot areas of enzyme activity and shortens the duration of hot moments in the root-detritusphere. Soil Biol Biochem 107:226–233. https://doi.org/10.1016/j.soilbio.2017.01.009
Maestre FT, Bradford MA, Reynolds JF (2005) Soil nutrient heterogeneity interacts with elevated CO2 and nutrient availability to determine species and assemblage responses in a model grassland community. New Phytol 168:637–650. https://doi.org/10.1111/j.1469-8137.2005.01547.x
Maestre FT, Reynolds JF (2007) Amount or pattern? Grassland responses to the heterogeneity and availability of two key resources. Ecology 88:501–511. https://doi.org/10.1890/06-0421
Marsh P, Pomeroy JW (1999) Spatial and temporal variations in snowmelt runoff chemistry, Northwest Territories, Canada. Water Resour Res 35:1559–1567. https://doi.org/10.1029/1998WR900109
McClain ME, Boyer EW, Dent CL et al (2003) Biogeochemical Hot Spots and Hot Moments at the Interface of Terrestrial and Aquatic Ecosystems. Ecosystems 6:301–312. https://doi.org/10.1007/s10021-003-0161-9
Miller WW, Johnson DW, Denton C et al (2005) Inconspicuous Nutrient Laden Surface Runoff from Mature Forest Sierran Watersheds. Water Air Soil Pollut 163:3–17. https://doi.org/10.1007/s11270-005-7473-7
Miller WW, Johnson DW, Loupe TM, Sedinger JS, Carroll EM, Murphy JD, Walker RF, Glass DW (2006) Nutrients flow from runoff at burned forest site in Lake Tahoe Basin. Calif Agric 60:65–71. https://doi.org/10.3733/ca.v060n02p65
Mueller EN, Wainwright J, Parsons AJ (2008) Spatial variability of soil and nutrient characteristics of semi-arid grasslands and shrublands, Jornada Basin, New Mexico. Ecohydrology 1:3–12. https://doi.org/10.1002/eco.1
Nakagawa S, Schielzeth H (2013) A general and simple method for obtaining R² from generalized linear mixed-effects models. Methods Ecol Evol 4:133–142. https://doi.org/10.1111/j.2041-210x.2012.00261.x
Palacios PG, Maestre FT, Gallardo A (2011) Soil nutrient heterogeneity modulates ecosystem responses to changes in the identity and richness of plant functional groups. J Ecol 99:551–562. https://doi.org/10.1111/j.1365-2745.2010.01765.x
Parkin TB (1987) Soil Microsites as a Source of Denitrification Variability 1. Soil Sci Soc Am J 51:1194–1199. https://doi.org/10.2136/sssaj1987.03615995005100050019x
R Development Core Team (2008) R: A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria. http://www.R-project.org/
Robinson D (1994) The responses of plants to non-uniform supplies of nutrients. New Phytol 127:635–674. https://doi.org/10.1111/j.1469-8137.1994.tb02969.x
Robinson D (1994) The responses of plants to non-uniform supplies of nutrients. http://onlinelibrary.wiley.com/doi/10.1111/j.1469-8137.tb02969.x/epdf
Robinson D, Hodge A, Griffiths BS, Fitter AH (1999) Plant root proliferation in nitrogen rich patches confers competitive advantage. Proceedings of the Royal Society of London Series B: Biological Sciences 266:431–435. https://doi.org/10.1098/rspb.1999.0656
Ryel RJ, Caldwell MM (1998) Nutrient acquisition from soils with patchy nutrient distributions as assessed with simulation models. Ecology 79:2735–2744. https://doi.org/10.1890/0012-9658(1998)079[2735:NAFSWP]2.0.CO;2
Schimel JP, Bennett J (2004) Nitrogen mineralization: challenges of a changing paradigm. Ecology 85:591–602. https://doi.org/10.1890/03-8002
Schlesinger WH, Raikes JA, Hartley AE, Cross AF (1996) On the Spatial Pattern of Soil Nutrients in Desert Ecosystems. Ecology 77:364–374. https://doi.org/10.2307/2265615
Skogley E, Dobermann A (1996) Synthetic Ion-Exchange Resins: Soiil and Environmental Studies. J Environ Qual 25:13–24. https://doi.org/10.2134/jeq1996.00472425002500010004x
Stark HM, Hart SC (1999) Effects of disturbance on microbial activity and N-cycling in forest and shrubland ecosystems. United States Department of Agriculture Forest Service General Technical Report PNW 101–105
Strayer DL, Ewing HA, Bigelow S (2003) What kind of spatial and temporal details are required in models of heterogeneous systems? Oikos 102:654–662. https://doi.org/10.1034/j.1600-0706.2003.12184.x
Sun T, Wang Y, Lucas-Borja ME, Jing X, Feng W (2021) Divergent vertical distributions of microbial biomass with soil depth among groups and land uses. J Environ Manage 292:112755. https://doi.org/10.1016/j.jenvman.2021.112755
Turner MG, Gardner RH (2015) Ecosystem Processes in Heterogeneous Landscapes. Landscape Ecology in Theory and Practice. Springer, New York, pp 287–332. DOI: 10.1007/978-1-4939-2794-4_8
Vidon P, Allan C, Burns D et al (2010) Hot Spots and Hot Moments in Riparian Zones: Potential for Improved Water Quality Management1. JAWRA J Am Water Resour Association 46:278–298. https://doi.org/10.1111/j.1752-1688.2010.00420.x
Vitousek P, Reiners W (1975) Ecosystem succession and nutrient retention: a hypothesis. Bioscience 25:376–381. https://doi.org/10.2307/1297148
Wagner-Riddle C, Baggs EM, Clough TJ et al (2020) Mitigation of nitrous oxide emissions in the context of nitrogen loss reduction from agroecosystems: managing hot spots and hot moments. Curr Opin Env Sust 47:46–53. https://doi.org/10.1016/j.cosust.2020.08.002
Wang P, Shu M, Mou P, Weiner J (2018) Fine root responses to temporal nutrient heterogeneity and competition in seedlings of two tree species with different rooting strategies. Ecol Evol 8:3367–3375. https://doi.org/10.1002/ece3.3794
Washburn CS, Arthur MA (2003) Spatial variability in soil nutrient availability in an oak-pine forest: potential effects of tree species. Can J For Res 33:2321–2330. https://doi.org/10.1139/x03-157
Weathers KC, Cadenasso ML, Pickett STA (2001) Forest Edges as Nutrient and Pollutant Concentrators: Potential Synergisms between Fragmentation, Forest Canopies, and the Atmosphere. Conserv Biol 15:1506–1514. https://doi.org/10.1046/j.1523-1739.2001.01090.x
Wei T, Simko V (2017) R package "corrplot": Visualization of a Correlation Matrix. Version 0.84. https://github.com/taiyun/corrplot
Wijesinghe DK, John EA, Beurskens S, Hutchings MJ (2001) Root system size and precision in nutrient foraging: responses to spatial pattern of nutrient supply in six herbaceous species. J Ecol 89:972–983. https://doi.org/10.1111/j.1365-2745.2001.00618.x
Wijesinghe DK, John EA, Hutchings MJ (2005) Does pattern of soil resource heterogeneity determine plant community structure? An experimental investigation. J Ecol 93:99–112. https://doi.org/10.1111/j.0022-0477.2004.00934.x
Williams MW, Bales RC, Brown AD, Melack JM (1995) Fluxes and transformations of nitrogen in a high-elevation catchment. Sierra Nev Biogeochemistry 28:1–31. https://doi.org/10.1007/bf02178059
Woodward C (2012) Nutrient hot spots in a Sierra Nevada soil: physical assessments and contributing factors. Dissertation, University of Nevada, Reno
Woodward C, Johnson DW, Meadows MW, Miller WW, Hynes MM, Robertson CM (2013) Nutrient hot spots in a Sierra Nevada forest soil: temporal characteristics and relations to microbial communities. Soil Sci 178:585–595. 10.1097/SS.0000000000000023
Yang JE, Skogley EO (1992) Diffusion kinetics of multinutrient accumulation by mixed-bed ion‐exchange resin. Soil Sci Soc Am J 56:408–414. https://doi.org/10.2136/sssaj1992.03615995005600020011x
Yang Y, Hart SC, McCorkle EP et al (2021) Stream Water Chemistry in Mixed-Conifer Headwater Basins: Role of Water Sources, Seasonality, Watershed Characteristics, and Disturbances. Ecosystems 24:1853–1874. https://doi.org/10.1007/s10021-021-00620-0
Zamora-Reyes D, Black B, Trouet V (2021) Enhanced winter, spring, and summer hydroclimate variability across California from 1940 to 2019. Int J Climatol 42:4491–4988. https://doi.org/10.1002/joc.7513
Zhao S, Zhang B, Sun X, Yang L (2021) Hot spots and hot moments of nitrogen removal from hyporheic and riparian zones: A review. Sci Total Environ 762:144168. https://doi.org/10.1016/j.scitotenv.2020.144168

Download PDF

Journal Publication

published 13 Jan, 2024

Read the published version in Biogeochemistry →

Reviewers agreed at journal
10 Jul, 2023
Reviewers invited by journal
10 Jul, 2023
Editor invited by journal
03 Jul, 2023
Editor assigned by journal
03 Jul, 2023
First submitted to journal
29 Jun, 2023

You are reading this latest preprint version

The Median Isn’t the Message: Elucidating Soil Nutrient Hot Spots

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Materials and methods

Results

Discussion

Declarations

References

Supplementary Files

Status:

Journal Publication

Version 1