Post-fire SOC and delta-13C at different types of microsite at a grassland-shrubland ecotone


Woody plant encroachment of grassland ecosystems is a geographically extensive phenomenon that can lead to rapid land degradation and significantly alter global biogeochemical cycles, and this ecosystem change has been particularly well documented in the desert grassland of the southwestern United States. Fires are known to decrease vegetation cover and increase soil erodibility, and the shifts in wildfire regimes are currently occurring in Chihuahuan Desert. It is generally recognized that the invasion of woody vegetation into grasslands and savannas will increase the carbon stored in arid ecosystems. However, carbon storage may be complicated by disturbance such as wildfire, which alters the distribution and amount of C pools in the drylands. The relative distribution of each vegetation type to the soil C pool and its variability after fires are not well-understood in this ecosystem. This research will investigate the variations of SOC and its vegetation source partition at microsite scale in the woody shrub encroached grassland after the occurrence of fire, which will provide further information on wildfire’s influence on soil C pool dynamics in arid and semiarid lands. The post-fire changes of the spatial pattern of SOC and vegetation contributions in the shrub encroached grassland will be analyzed using a geostatistical method outlined in Guan et al. (2018). Overall, understanding the post-fire redistribution and sources of SOC may provide insights on the important role played by fire, aeolian processes and vegetation in the dynamics of desert grassland ecosystems.

Data set ID: 


Core Areas: 

Additional Project roles: 




A 100 m × 100 m monitoring area was established in March 2016. A prescribed fire was set to burn the monitoring area on March 10, 2016, the beginning of the windy season, to create the burned treatment. Within each (5 m × 5 m) sampling plot, 50 randomly distributed soil samples were collected from the top 5 cm of the soil profile. The coordinates of the sampling locations were randomly generated and a different set of sampling locations was used for each sampling period.

Data were collected during the following time periods:
2016/03/11 - 2016/03/13
2016/06/14 - 2016/06/16
2017/03/17 - 2017/03/19
2017/06/26 - 2017/06/28


Instrument Names: Meter stick, shovel, sampling bags, ball mill, element analyzer.

Manufacturer: ball mill (PBM-04 Planetary Ball Mill, RETSCH, Germany), element analyzer (FLASH 2000 OEA, Thermal Fisher Scientific, USA).

Model Number: PBM-04 Planetary Ball Mill, FLASH 2000 OEA

Quality Assurance: 

Each sample was analyzed three times and the results were averaged.

Additional information: 

We greatly acknowledge the contributions of Jon Erz, Eric Krueger and Andy Lopez (FWS, SNWR), Scott Collins and Amaris Swan (Sevilleta LTER, New Mexico, USA), Julie McDonald and Bethany Theiling (The University of Tulsa) for their assistance in field work and laboratory analysis.

The site is a black grama (Bouteloua eriopoda) dominated grassland with creosote bush shrubs (Larrea tridentata), and the soil is primarily sandy loam. This sampling area is large enough to capture the heterogeneity in the landscape comprising of shrub, grass and bare soil microsites.

The soil TC and TN results of the same field site can be found in Guan et al., 2018. Ecosystems (Post-fire Redistribution of Soil Carbon and Nitrogen at a Grassland– Shrubland Ecotone).

15N Tracer Study in Grassland and Shrubland at the Sevilleta National Wildlife Refuge, New Mexico (1989-2002)


Because grasses and shrubs may induce different spatial distributions of nutrients in desert soils, this study was initiated to examine the redistribution of nitrogen in grassland and shrubland soils over a long time period. The stable isotope N15 was applied to plots in grassland and shrubland, and the plots were measured annually from 1989-1993 and again in 1999, 2001, and 2002.

Core Areas: 

Data set ID: 



Data sources: 



Plot establishment - We established four 10 X 10-m plots in grassland and shrubland habitats near the Five Points area of the Sevilleta Long Term Ecological Research Site (LTER) in the northern Chihuahuan Desert, New Mexico, USA.

In the grassland, paired plots (two plots separated by 50-m) were located at sites where Bouteloua eriopoda (Torr.) Torr. (Black grama) and B. gracilis (H.B.K.) Lag. ex Steud. (Blue grama) dominate. In the shrubland, paired plots were located in areas dominated by Larrea tridentata D.C. (Cov.)(creosotebush).

Tracer application - 15NH4Cl tracer was applied to ten 15.25 cm diameter points arranged in a stratified random design in each of the four 10 X 10 m plots in the grassland and shrubland. 0.33 g of 15NH4Cl were dissolved in 500 ml of deionized water and applied to ten sites per plot in 50 mL aliquots in July, 1989.

Sampling - The site of each spike application was sampled annually during the summer in 1989, 1990, 1991, 1992, and 1993. Two samples with a volume of 28.5 cubic centimeters were removed from the site of spike application, which had a total volume of 1815 cubic centimeters to 10-cm depth. Each sample contained a small percents of the total soil volume, about 1.5 percent. All samples were air-dried, sieved through a 2 mm sieve, and shipped to Duke University. Ground soil samples were analyzed for 15N.

Field collections were made every even-numbered month as close to the 15th as possible.

The samples taken in 1989-1993 and 1999 were analyzed by Larry Giles on a mass spectrometer at the Duke University

Precipitation Chemistry Data For the Sevilleta National Wildlife Refuge, New Mexico (1989-present)


This data set includes precipitation chemisty from 20 funnel collectors on the Sevilleta National Wildlife Refuge (NWR). Variables measured include volume, NO3-N, NH4-N, SO4, Cl, Na, K, Ca, Mg, and PO4. The sample interval depends on the frequency of significant precipitations events. Field collection of precipitation chemistry samples occurred as soon after a significant precipitation event as possible (usually within a week after the event). In some cases collection of precipitation chemistry samples occurred more often than meteorlogical station samples (summer, fall), and sometimes less often (winter, spring). Precipitation chemistry samples were collected only when there was enough sample to allow for complete chemical analysis, (usually greater than 100mls). After 1995 all samples continued to be collected for volume but only an a partial set of the samples continued to be analyzed for inorganic nutrients.

Core Areas: 

Data set ID: 


Additional Project roles: 




To quantify spatial and temporal patterns of inputs of precipitation and dissolved inorganic nutrients in this precipitation across the Sevilleta


How the Samples were Collected

The 26.67cm (10.50in) polyethylene funnels are acid washed prior to collection with a 50 deionized water rinse. For each collection, a one gallon acid washed replacement bottle is prepared containing 1ml of a 1000ppm solution of phenyl mercuric acetate (PMA). The PMA functions as a preservative. The collected sample is routed through tygon tubing between the funnel and the bottle and forms a vapor trap to prevent evaporation of the sample. A tube running from the sample bottle to a 4oz. bottle containing water, provides a second vapor trap but allows the sample bottle to breathe during precipitation collection.

Funnel Precipitation

Collection Samples should be collected with care being taken to not contaminate the precipitation samples. Hands are a primary source of Na contamination so do not touch any surfaces that precipitation samples have or will come in contact with. Inspect funnel setup for damage of contamination. Look inside the funnel to see if it has been contaminated by bird droppings. If so, the funnel will have to be replaced with a clean one. Unscrew the lid of bottle; put lid from replacement bottle on the sample bottle. Put new bottle on stand and replace cap. With the two tubes coming out of the lids it is necessary to slightly twist these tubes counter clockwise before starting to twist the cap on so that the tubes do not end up too twisted after screwing on the lid. One can also simply screw the bottle on to the lid. Make sure that the seal is tight so that precipitation will not evaporate out of the bottle.

Wet/dry collectors

A wet/dry collector has been installed at Met40 (Deep Well). The Wet/Dry collector sample will be collected after significant storm events. The bulk chemistry collector at this site (PE3) should always be collected at the same time. As with all precipitation sample collection, care must be taken to not touch any surfaces that have or will come in contact with precipitation, therefore, all buckets should be handled by the sides, bottoms or handles. Care should also be taken to avoid stirring up dust in the vicinity of the collector which may bias the dry sample collection. The procedure for collecting the wet and dry portions of this collector, is as follows: Take two acid washed buckets with loosely fitting lids (that should always be transported in plastic garbage bags), out to the collector. Also, take 1000ml wide mouth bottle and 1000ml graduated cylinder. Place lid from one of the clean buckets on the dry collection bucket. Remove the tie-down strap and remove the dry bucket from the stand. Place the clean bucket in the holder and secure it with the tie-down. Actuate the moisture sensor so that the cover moves over the dry bucket (spit works). If cover does not move, check the battery voltage and connections and replace battery if necessary. Open 1000ml bottle, remove tie-down from wet bucket, remove bucket from holder, swirl sample in bucket well and transfer up to 1000ml of it to the wide mouth bottle. Place wet and dry buckets back in plastic garbage bags for return to laboratory.

Laboratory Processing

When bulk field samples arrive in the laboratory; individual samples are weighed, volumes are recorded (1gram = 1ml of water) and a 60ml subsample is poured off and stored for the wet chemistry analyses. Generally, samples with less than 100ml are not collected until the 100ml volume is exceeded. In some cases, bulk field samples containing less than 100ml, must be collected (as may be the case at the end of a quarter). For these samples, 100ml of deionized water is added to the original volume, so all analyses can be completed. Identified contaminated bulk samples, are noted as such, volumes are recorded, then the samples are often discarded. The primary precipitation chemical analyses are done using the 60ml subsample. In an attempt to isolate contaminated samples that were impossible to identify as such in the field; samples are initially analyzed for nitrate, ammonium, and major cations including; sodium, potassium, calcium and magnesium. Data collected from these analyses, are then examined and compared with individual sample volumes. Ratios of specific cation concentrations within a sample are compared, then correlated with expected ratios. Samples with unusually high ratios, are determined to be contaminated. Contaminated samples are recorded as such and receive no further analyses. The remaining samples are then analyzed for chloride, and sulfate.

1st quarter: January 1 - March 30

2nd quarter: April 1 - June 30

3rd quarter: July 1 - September 30

4th quarter: October 1 - December 31

Specific analytical procedures

The following text contains specific analytical descriptions and procedures used to analyze precipitation chemistry for the Sevilleta LTER. The Table below lists each analysis indicated with the appropriate analytical procedure.

Analysis Procedure/Methods
NO3-N    Technicon Auto-analyzer II; Method No. 100-70W

NH4-N    Technicon Auto-analyzer II; Method No. 98-70W
            (modification; phenol reagent adjusted to pH=12).

SO4       Technicon Auto-analyzer II; Method No. 226-72W

Cl          Technicon Auto-analyzer II; Method No. 99-70W
             (modification; final concentrations adjusted to
             account for PMA interference).

After May 1992 analyses for NO3, SO4, Cl and PO4 were performed using
a Dionex D-100 Ion Chromatograph

Base cations: All Major Cation analyses are completed with a Perkin Elmer 306 Atomic Absorption Spectrophotometer. The flame method is used.



Ca (modifications; [La] added, (1ml [La]/10ml sample)
      to suppress [Ca] and [Mg] ionization interference. Sample receives(2X) dilution).

Mg (modifications; [La] added, (1ml [La]/10ml sample)
       to suppress [Ca] and [Mg] ionization
       interference. Samples receive (2X) dilution).

Data sources: 



log file created 12/15/1998 - D.M. July 5th, 1991. File recieved from Todd Haagenstad Data set was reformatted for "rdb," and new variables were added including year, quarter, and error (see database variables) by Michelle L. Murillo. Additional information and some editing was done on the abstract, further editing is still required. December 11, 1992-Edited for 1991 and 1992 data files. Feb 11, 2000-Updated for period since 1991. Mar 8, 2005 - File updated with data from 2002-2004 D.M. doc. July 31, 2009 - Redid all 1989-2004 data and added variable mm. This is a derived value based on volume and the cross sectional area of the collector for current funnels mm=Volume/55.9 for buckets mm=volume/64. D.M. July 31, 2009 - changed all date variables to format YYYYMMDD D.M. July 31, 2009 - added new data from 2005-2008 D.M.

Quality Assurance: 

A measure of the quality of the data can be derived from the flag variable.  These letter codes are derived from field recordings as well as results from lab analysis. Below are are list of the flags and what
they denote. Capitals and small letters are equivalent

Flag Meaning

a No obvious problems denoted in the sample
c Contaminated - either observed in the field or found from lab results
G Bottle overflowed so volume came from associated weather station gauge
L Sample lost - either during collection or processing
M Sample bottle missing in the fileld - often due to strong winds
O Overflowed - bottled overflowed
R Sample obtained from rinsing down funnel- for samples when no rain occurred for a long time
P Funnel plugged - volume may be incorrect due to evaporation
v Volume only - chemical analysis not done on samples - this was true for many of the samples after 1995

Additional information: 

Additional Information on the Data Collection Period

Data collected four times per year (usually January, April, July and October).Sampled one week per season, four seasons per year. Samples were generally collected monthly and usually on the last day of the month or on the first day of the subsequent month. The exceptions are 1) when there is little or no precipitation during that month in which case the collection is merely postponed until the following month or 2) the opposite extreme when there is heavy rain during the month so that it bcomes necessay to collect the samples before the collection bottles overflow.

Where the Data were Collected

A network of 20 funnel collectors and 2 wet/dry chemistry collectors was set out to monitor precipitation chemistry. There are 10 funnel collectors located on the west side and 10 on the east side of the Sevilleta NWR. Table 1. gives coordinates and description of the location of these collectors. In Table 1. P = ppt, E = east side, W = west side and the number = collector number. Figure 1. shows the individual collector site locations. A bulk chemistry funnel collector was placed at each of the 6 meteorlogical stations noted in the meteorlogical data archive files, with the other 14 distributed over the entire range of the Sevilleta. After 1995 only 6 of the funnels continued to be processed completely along with the wet and dry samples from each of the wet/dry collectors. In Table 1 an *denotes those samples continuing to be fully analyzed.

New funnels were added to the east side of the network beginning in 1998. The locations and installation dates of these funnels have now been added to the table. In addition PE4 was moved about 2 km west in conjunction with the installation of a new Met station and analyses were begun again in Jan 1999. Analyses were restarted at PE2 in Nov, 2001 as well. Some of the west side funnels have been totally discontinued. These include. PW3, PW5, PW6, PW9 and PW10

The location of all funnels can be seen at:

Table 1. Description and coordinates of precipitation chemistry collectors.

No.        Location             Met.sta.      Lat.            Long.              Elev.(meters)   Date Installed
PE1      NW corner                        34.4063  -106.7874     1446               011789
PE2*    Black Well                        34.3971  -106.6609     1547               011789
PE3*    Deep Well         Met40       34.3601  -106.6891    1571               011789
PE4*    Five Points                       34.3301  -106.7061     1596               011789
PE4     New Five Points   Met49      34.3351  -106.7287     1611               010199
PE5     Gibbs Farm                       34.2699   -106.7516    1544               011789
PE6*   Southern Gate    Met41       34.2186  -106.7950    1515                011789
PE7     Montosa Road                   34.2589   -106.6664    1683               011789
PE8     Nunn Flats                        34.3702   -106.6172    1635               011789
PE9*   Cerro Montosa    Met42      34.3682    -106.5347   1944               011789
PE10   Sepultura Canyon              34.3024    -106.6203   1803               050389
PE13* Savanna            Met48      34.4145    -106.5229    1794               010199
PE14* Blue Grama        Met50      34.3350    -106.6313    1669              113001
PE15* SE McKenzie Flats              34.2800    -106.6723    1696              113001

PW1 NE Corner                           34.4058     -106.8792   1448              020289
PW2 Bronco Well          Met45      34.4068     -106.9328   1524              020289
PW3 222 Well                            34.4184      -107.0316   1800             020289
PW4* Watersheds        Met43      34.3987     -107.0376   1748              020289
PW5 Mid North                           34.3684      -106.9546   1560              020289
PW6 La Cueva                           34.3464      -107.1415   1577             020289
PW7* Headquarters                    34.3512     -106.8815   1440              020289
PW8* Rio Salado         Met44      34.2952     -106.9241   1485              020289
PW9 West Mesa                         34.2617     -107.0650   1734              033089
PW10 South Central                   34.2537      -106.9881   1596              050389

* Funnels being collected as of Jan 1 2009

Wet/Dry Buckets
As with the funnels, the P designates a precipitation sample, the E and W designate the east or west side of the refuge. 11 is the wet fraction while 12 is dry fraction solublized in 250 ml of deionized

No.       Location                Met.sta.    Lat.              Long.          Elev.(meters)   Date Installed

PE11*  Deep Well         Met40      34.3601   -106.6891    1571              061990
PE12*  Deep Well         Met40      34.3601   -106.6891    1571              061990

PW11* Rio Salado        Met44      34.2952   -106.9241    1485              040492
PW12* Rio Salado        Met44      34.2952   -106.9241    1485              040492

In February, 2002, the Wet/Dry collector was moved from Rio Salado to the Cerro Montoso site. The wet and dry samples being collected at this site were given the designation PE19 and PE20 respectively. Coordinates are given below.

PE19* Cerro Montoso    Met42      34.3682   -106.5347    1944              050202
PE20* Cerro Montoso    Met42      34.3682   -106.5347    1944              050202

Subscribe to RSS - biogeochemistry