Supplement 1. Data files to calculate nutrient removal and daily metabolism, summary of autotroph tissue stoichiometry and proxy estimation of diel calcium variation.

File List Calcite_Proxy_Data.csv (MD5: 0eb0a030d878b5020e39b62e1d83c593) Tissue_Stoichiometry_Data.csv (MD5: 90b5e91f413cd35e2a6f12d2c2bfcc44) RawData_Sept2009.csv (MD5: 090f9dd058ddc30e4e385f08bc43ca76) RawData_April2010.csv (MD5: 9bf8f69e49fecf8f4ddffb659a93c580) RawData_May2010.csv (MD5: e512d00f440adec8dc1158803c425896) RawData_Oct2010.csv (MD5: 701839ffa9863abac5442e5c27e8be2d) RawData_Dec2010.csv (MD5: 5b46fbb51fe83f785be5e5ee360d8c7c) RawData_Feb2011.csv (MD5: 144535aa4175e234ea0884f8e40516f3) RawData_Mar2011.csv (MD5: aacbaad8e6bc8cf82684392afc3c4a7a) RawData_May2011.csv (MD5: 93db5bfa46d5b173270f1cd8205a5f9e ) Description Calcite_Proxy_Data.csv – This file provides the raw data from three different 24–48 hours deployments of the relationship between specific conductance (measured in situ) and caclium concentration (measured from grab samples). In each table, the first column is the date and time, the second column is measured Ca concentration (in mM) and the third column is measured specific conductance (in μS/cm). Missing values are samples that were delineated as outliers or were missing. Tissue_Stoichiometry_Data.csv – This table presents measurements of tissue stoichiometry from the springs of the Ichetucknee River complex. The first column provides information about the source of the sample (BLH = Blue Hole Spring, CED = Cedear Head Spring, ICH = Ichetucknee Head Springs, MIS = Mission Spring). The second column is the species name (L = Lynbya wollei, SAG = Saggitaria Kurziana, V = Vaucheria spp. and V = Valisneria americana). The third and fourth columns provide the latitude and longitude, respectively, of the collected sample in decimal degrees; all samples were collected on February 2nd, 2011. The fifth column indicates whether the sample was obtained from a sunny (Sun) are shady (Shade) site. Columns 6–8 present the mass fraction of each element (C, N and P, respectively) within the sampled tissues. Column 9–11 show the molar ratio of the element abundances (C:N, C:P and N:P, respectively). RawData_Sept2009 – This file presents the raw data and derived values for the sensor deployment between September 6th and September 16th 2009. The first column indicates the date and time. The second column indicates the discharge in L/s. Columns 3 through 5 are the measure SRP concentration in μM, the same concentration in μg SRP/L, and the 3-hour smoothed concentrations (also in μg SRP/L). Column 6 is the diel maxima (μg SRP/L), obtained for each day from the period near midday, without any calcite correction. Column 7 is the hourly P retention (g P/h), estimated from the concentration deviation between the smoothed hourly measurement (column 5) and the daily maxima, multiplied by discharge. The next four columns (8–11) are estimates of daily P retention, first in g P/d (column 8), then on a per unit area basis (g P·m^-2·d^-1; column 9); column 10 is the diel SRP range (μg SRP/L) and column 11 is the estimated C:P ratio given the value in column 9 and the net primary production estimate from daily GPP (column 43) and assuming autotrophic respiration is 0.5 of GPP and a photosynthetic quotient of 1. Columns 12–20 are the same as columns 3–11, but where the SRP concentrations have been corrected for calcite co-precipitation fluxes (based on calculations in columns 21–29). Shown are the SRP concentration in μM (column 12) and μg SRP/L (column 13), the 3-hour smoothing (μg SRP/L; column 14), the diel SRP maximum (μg SRP/L; column 15), and the instantenous P uptake (g P/h; column 16), estimated from the smoothed, calcite co-precpitation corrected SRP signal (column 14). Columns 17–20 are the daily P retention (g P/d; column 17), P retention per area (g P·m^-2·d^-1; column 18), the diel SRP range and the C:P molar ratio for the calcite co-precipitation corrected signal. Columns 20–29 document how calcite co-precipitation was estimated, based on water temperature (column 20) and specific conductance (column 21). The estimated calcium concentration (mM) is shown in column 22, based on parameters from the Calcite_Proxy_Data.csv file. The diel [Ca] maxima (mM) is in column 23, estimate for each day based on variation in column 22. The mass removal of Ca (mM/h; column 24) is based on the difference between the hourly measurement and the diel maxima. The associated mass flux of P onto the precipitated calcite is computed in columns 25–28, using the equations reported in the manuscript text. Column 28 is the molar correction used in column 12 to adjust the measured P concentration to correct for calcite co-precipitation. Column 29 reports the daily flux of P onto calcite (mM P/d). Columns 30–43 show the computation of ecosystem metabolism from diel variation in dissolved oxygen and water temperature. Columns 30–32 show the raw data (pH, dissolved oxygen as % saturation, and dissolved oxygen in mg DO/L). A diel DO baseline is estimated each 24-hour period (mg DO/L; column 33), and an hourly change in DO (column 34, mg DO/L) was calculated. The saturation DO level (mg DO/L; column 35) was compared with the observed value to estimate a saturation deficit (column 36, mg DO/L). The reaeration constant k (hr^-1) was estimated using the nighttime regression technique, and the mean value across each deployment was used; this value is reported in column 37. Column 38 reports the gas exchange per volume (g DO·m^-3·hr^-1) as the product of the saturation deficit and the reaeration constant. The observed hourly DO change (mg DO/L; column 34) was adjusted based on this reaeration flux to yield column 39. The mean nighttime flux measured in column 39 was used to estimate ecosystem respiration on an hourly basis (g DO·m^-3·hr^-1) in column 40, and scale to a daily respiration flux per unit area (g DO·m^-2·d^-1) in column 41. Hourly primary production (g DO·m^-3·hr^-1) was estimated based on deviation from the daily baseline, adjusted for variation in reaeration with changing DO concentration (column 42), and was converted to daily gross primary production (GPP; g DO·m^-2·d^-1) based on flow and benthic area (column 43). Columns 44–46 report raw weather data for the period of the deployment, including hourly temperature (column 44, degrees C), hourly rainfall (column 45, cm) and hourly solar radiation (column 46, w/m²). Columns 47–52 report how N retention was quantified. Column 47 is the raw hourly nitrate concentration (mg N/L). The diel maxima sets the baseline (column 48; mg N/L) from which hourly removal is estimated using discharge (column 49, g N/h). This flux is adjusted to a benthic area flux (mg N·m^-2·d^-1) based on the stream bed area, summing over 24 hours (column 50). Column 51 is the dissimilatory flux (mg N·m^-2·d^-1) based on the remaining difference between the measured (column 47) and the expected concentration based on flow-weighted springs inputs (column 52; mg N/L). RawData_April2010 – This file is organized identically to the RawData_Sept2009 file. Note that there are no SUNA data for this deployment. RawData_May2010 – This file is organized identically to the RawData_Sept2009 file, with the exception that the nitrate concentrations are reported for 15-intervals; this necessitates inserting a new data-time column (column 47), and moving the remaining columns over. Note also that there are many more rows of nitrate data than other solute measurements, where hourly resolution was retained. RawData_Oct2010 – This file is organized identically to the RawData_Sept2009 file. RawData_Dec2010 – This file is organized identically to the RawData_Sept2009 file, with the exception that the nitrate concentrations are reported for 15-intervals; this necessitates inserting a new data-time column (column 47), and moving the remaining columns over. Note also that there are many more rows of nitrate data than other solute measurements, where hourly resolution was retained. RawData_Feb2011 – This file is organized identically to the RawData_Sept2009 file, with the exception that the nitrate concentrations are reported for 15-intervals; this necessitates inserting a new data-time column (column 47), and moving the remaining columns over. Note also that there are many more rows of nitrate data than other solute measurements, where hourly resolution was retained. ...