Next Article in Journal
Effect of Perchlorate on Combustion Properties of Directly-Written Al/PVDF Composites
Previous Article in Journal
A Quantitative Analysis of Fuel Break Effectiveness Drivers in Southern California National Forests
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Patch Burning Improves Nutritional Quality of Two Gulf Coast Grasses—And Winter Burning Is Better than Summer Burning

1
Powderhorn WMA Biologist, Texas Parks and Wildlife Department, 418 South 16th Street, Port O’Connor, TX 77982, USA
2
Department of Rangeland and Wildlife Sciences, Caesar Kleberg Wildlife Research Institute, Texas A&M University-Kingsville, 700 University Blvd., MSC 218, Kingsville, TX 78363, USA
3
Carson City District Bureau of Land Management, 5665 Morgan Mill Road, Carson City, NV 89701, USA
4
East Foundation, 200 Concord Plaza Drive, Suite 410, San Antonio, TX 78216, USA
5
La Paloma Ranch, P.O. Box 447, Charlotte, TX 78011, USA
*
Author to whom correspondence should be addressed.
Fire 2023, 6(3), 105; https://doi.org/10.3390/fire6030105
Submission received: 8 February 2023 / Revised: 3 March 2023 / Accepted: 6 March 2023 / Published: 7 March 2023

Abstract

:
Gulf cordgrass (Spartina spartinae [Trin.] Merr. ex Hitchc.) and seacoast bluestem (Schizachyrium scoparium [Michx.] Nash var. littorale [Nash] Gould) are dominant native warm season grasses in the Texas Coastal Prairies and Marshes ecoregion. Mature Gulf cordgrass nutritive value is considered poor for grazing animals, while seacoast bluestem nutritive value is considered fair to good in spring and early summer. We compared season of burn effects on crude protein (CP) and neutral detergent fiber (NDF) of these grasses. Our study employed four patches (200–305 ha) burned in a patch-burn grazing system (two winter, two summer 2016). Forage samples were collected before burning and every 3 d for 40 d after burning, then weekly for 50 d. Regardless of season of burning, CP was adequate for maintaining a lactating beef cow 90 d post-burning. Although peak CP in Gulf cordgrass following burning did not differ between seasons, CP increased more quickly following winter burning, when it rose abruptly for ~6.4 d and then increased more slowly to its peak at ~30 d post burn, after which it declined. After summer burning, CP of Gulf cordgrass did not increase significantly, and then declined after reaching its peak at ~29 d post burn. Gulf cordgrass NDF declined more rapidly following summer burning than winter burning but remained low for a longer period after winter burning. In seacoast bluestem, CP increased following winter burning, peaking at ~11 d after burning, but there was little change following summer burning, which did not peak until 31 d post burn. NDF of seacoast bluestem declined similarly in both seasons. This study indicates that prescribed fire during winter is more beneficial for improving the nutritive value of Gulf cordgrass and seacoast bluestem than summer burning.

1. Introduction

The Gulf Coast Prairies and Marshes ecoregion in Texas consists of productive rangelands that provide cover and forage to wildlife and livestock [1,2]. This ecoregion produces a nearly continuous supply of forage because winters are relatively mild (average low and high temperatures, December through February, are 9–22 °C) [3], and growing seasons are 289 to 300 days long [4]. Gulf cordgrass (Spartina spartinae [Trin.] Hitchc.) and seacoast bluestem (Schizachyrium scoparium [Michx.] Nash var. littorale [Nash] Gould) are native, perennial bunchgrasses that are principal components of this ecoregion [5].
Gulf cordgrass is well adapted to soils high in salinity; it also thrives in a variety of soil textures ranging from sandy loams to clays [6,7]. The greatest herbage yield for Gulf cordgrass occurs during the rainy months of spring and during September and October, with its greatest standing crop yield occurring in fertile clay loams and clays because these soil textures have higher moisture-holding capabilities [8,9]. Although Gulf cordgrass is highly productive and can maintain green tissue year-round, livestock do not graze the mature foliage to an appreciable extent if other forages are available [10]. Mature growth produces stiff and spine-like leaf blades of low nutritional quality [8,11,12] (crude protein generally < 5%).
Seacoast bluestem nutritive value is fair to good in spring and early summer [13,14,15] and provides forage for livestock throughout its wide geographic range [15] (and references therein). With long growing seasons (plants can remain green throughout the winter), seacoast bluestem also provides forage throughout much of the year [16,17] and is a preferred forage in the Coastal Prairies [2,18,19].
Management practices to improve forage quality of mid- and tall grasses such as Gulf cordgrass and seacoast bluestem are based on basic principles of plant growth and morphology of grasses [20,21,22,23]: the removal of fibrous mature material during strategic growth stages can promote nutritious regrowth, increase its abundance, and prolong its period of availability. Although prescribed burning often is used to implement these improvements, more information is needed on responses of these grasses to different seasons of burning. In particular, most landowners in this ecoregion who incorporate livestock grazing into their management enterprises graze cattle year-round: generally, there is no need to rest pastures in good condition to accumulate fuel loads prior to burning because of the productivity of these grasslands. Additionally, because units typically are large (>150 ha), livestock remain in burn units not only during prescribed burning but also post-burning. Thus, whereas plants experience fire and grazing simultaneously in these settings, managers have an option of when to burn—and this is the focus of our work. In particular, our objective is to document changes in forage quality [crude protein (CP) and neutral detergent fiber (NDF)] after prescribed fire in grazed pastures that were burned in winter (early February) or late summer (late July). Our goal was to develop prescribed fire recommendations regarding season of burning for improving nutritional value and utilization of Gulf cordgrass and seacoast bluestem rangelands using a patch-burn grazing design [24,25] where “spatially distinct patches [are] burned within larger areas” [24] (p. 132). We hypothesized that nutritive value would improve following burning in both seasons, and that this response would differ between seasons because plants would be burned in different growth stages.

2. Materials and Methods

2.1. Study Area

The East Foundation’s El Sauz Ranch is located primarily in Willacy County, Texas, with a small portion in Kenedy County, Texas, USA. El Sauz Ranch (26.5577° N/97.4263° W) is 11,082 ha, primarily in the Gulf Coast Prairies and Marshes Ecoregion. There are numerous ecological site types in the study area including sandy flats, active sandhills, low coastal sands, and coastal sands, with coastal sands being the primary site type on the property [26]. Soil series within the study area are: Arrada sandy clay loam, Dune land, Falfurrias fine sand, Galveston fine sand, (gently undulating), Galveston-Mustang complex, Incell clay, Lalinda sandy clay loam, Mustang fine sand, Sauz-Saucel sandy loam, and Sauz loamy fine sand [26]. The site has a humid subtropical climate with 658 mm mean annual rainfall and an average temperature fluctuation between 18.9 and 26.7 °C [27].
Vegetation within the study area consisted of native grasses such as Gulf cordgrass, seacoast bluestem, switchgrass (Panicum virgatum L.), hooded windmill grass (Chloris cucullata Bisch.), red lovegrass (Eragrostis secundiflora J. Presl.), tumble lovegrass (E. sessilispica Buckley), brownseed paspalum (Paspalum plicatulum Michx.), and coastal sandbur (Cenchrus spinifex Cav.). Forbs include partridge pea (Chamaecrista fasciculata (Michx.) Greene), Gulf croton (Croton punctatus Jacq.), littleleaf sensitive-briar (Mimosa microphylla Dryand.), lavender thrift (Limonium carolinianum [Walter] Britton), and American snoutbean (Rhynchosia americana [Houst. ex Mill.] M.C. Metz) [26]. Mottes of native woody species include primarily honey mesquite (Prosopis glandulosa Torr.) and live oak (Quercus virginiana Mill.) [26].
El Sauz was using a moderate stocking density of cattle close to 1 AU per 14 ha that was managed as a continuous grazing system. The cow-calf operation had roughly 660 breeding cows and 27 bulls that were not provided any supplemental feed. Breeding cows were a cross between Santa Gertrudis and Beefmasters (Bos taurus) breeds.

2.2. Field Methods

In June 2015 as part of a larger patch-burn and grazing study, we delineated 10 study patches (size range: 200–305 ha) in areas dominated by Gulf cordgrass and seacoast bluestem using existing roads as firebreaks wherever possible. We used a completely randomized design, with three treatments, (1) non-burn control, (2) winter burn, and (3) summer burn, randomly assigned to patches with 2, 4, and 4 replications, respectively. We defined winter treatments as any burn conducted in January or February, whereas summer treatments were defined as any burn conducted in July or August. Control treatments were not used for this aspect of our study; we compared effects of winter burning to those of summer burning. For nutritional analyses we collected data from the two patches burned in winter 2016 and two patches burned in summer 2016. Each patch had two 60 m transects that traversed a Gulf cordgrass community and an adjacent “other” vegetative community where seacoast bluestem dominated with roughly half of each transect in each vegetative community.
We estimated fuel load (kg ha−1) and fuel moisture (%) by gathering all aboveground biomass (standing live, standing dead, and litter) to the ground level in two, 0.25 m2 frames randomly placed on the ground within each vegetation community near the sampling transects. We placed fuel load samples in paper bags, weighed them while wet, and transported them to a drying trailer where they dried at 40 °C until all moisture was lost. Samples were then re-weighed. Fuel moisture was calculated using the following formula:
%   Fuel   Moisture =   wet   weight - dry   weight dry   weight   × 100
We placed a HOBO U30/RX3000 (Onset® Computer Corporation, Bourne, MA, USA) weather station at the center of the study area to acquire daily rainfall (mm) data following burning. On burn days, we recorded weather variables of air temperature (°C), wind speed (mps), wind direction, and relative humidity (%) near the burn patch approximately every 30 min during the burn (Kestrel® 4500 Weather Meter, Nielsen-Kellerman, Boothwyn, PA, USA). We measured fine fuel load (kg ha−1), fuel moisture (%), and fire temperature (°C) in each burn patch. We recorded fire temperatures using HOBO Type J, K, T, E, R, S, B, N Thermocouple Data Loggers (Onset® Computer Corporation, Bourne, MA) with High Temperature Inconel Overbraided Ceramic Fiber Insulated Thermocouples (XCIB-K-1-2-10, OMEGA® Engineering, Norwalk, CT). We placed each data logger inside a PVC pipe section capped at both ends with the thermocouple extended through a slot in the pipe. We buried the PVC with the data logger in the ground for protection from extreme heat immediately prior to burning. We used two data loggers near each vegetation transect in each burn patch, one near the Gulf cordgrass section and one near the seacoast bluestem section of the transect. The thermocouple was placed at the ground level near the base of a Gulf cordgrass plant on that end of each transect and near the base of a seacoast bluestem plant on the other end.
Fuel load samples were kept for pre-burn nutritional analyses after dry weights were obtained. To evaluate the influence of days since burning on nutritional value after winter and summer burning treatments, Gulf cordgrass and seacoast bluestem forage samples were collected in recently burned patches every third d for 36 d, and then once weekly for another 49 d. Forage samples were obtained by randomly placing four 1 m2 quadrats in both the Gulf cordgrass and seacoast bluestem communities near each transect, and hand-clipping all Gulf cordgrass or seacoast bluestem in each quadrat to a stubble height of ~2.5 cm. All forage clipped from a quadrat was placed into a labeled paper bag. Forage samples were kept in a drying trailer at 40 °C until no further weight loss occurred and were weighed for utilization estimation prior to nutritional analyses. After weighing, the four samples of Gulf cordgrass and seacoast bluestem from each transect were combined by species to create a composite sample. Composite samples were then ground in a THOMAS® Wiley mill (Thomas Scientific, Swedesboro, NJ, USA). Ground material was mixed and placed into labeled 113 g WHIRL-PAK® bags (Nasco, Fort Atkinson, WI, USA) for nutritional analyses.
Winter burning treatments were conducted on 5 and 10 February and summer treatments on 3 and 22 July 2016. Burning began as soon as fuels were dry enough to carry a fire, typically around 1000 (GMT -6). Blacklining was typically completed by 1400, and the burning the interior of the plot by 1800. Burn patches were pasture size and irregularly shaped because existing roads were used as firelines wherever possible. Fuels were unprotected and most closely resembled fuel model GR 8 [28] in Gulf cordgrass-dominated areas, and GR 6 or GR 7 in areas dominated by seacoast bluestem depending on recent rainfall and soil moisture. With a predominant southeast wind off the Gulf of Mexico, the lighting strategy of each patch began with blacklining occurring on the downwind west and north sides inside double mineral lines that were created roughly 30.5 m apart. With four to five torch carriers spread across the width of the area, the blacklines were ignited by walking and lighting into the wind in areas where no Gulf cordgrass occurred. In areas with thick Gulf cordgrass blacklining was completed using a backfire. Blacklining was conducted from 2 days to approximately 1 month before burning the interior of the patch because fire behavior is highly dependent on weather conditions, thus waiting for a suitable burning day is necessary. Flanking fires were used to burn the interior of the patch. Two lines of fire were simultaneously ignited using drip torches, starting in the southwest and northeast corners of the patch, lighting toward the southeast corner until they met. This created two flanking fires that pulled together in the middle of the patch.

2.2.1. Forage Chemical Analyses

Forage samples were analyzed at Texas A&M University-Kingsville for crude protein (CP) and sequential detergent fiber analysis. Van Soest et al. [29] detergent fiber analysis of neutral detergent fiber (NDF) was determined with an ANKOM 200 Fiber Analyzer (ANKOM® Technology Corp., Macedon, NY, USA). Total nitrogen (N) for the CP analysis (N × 6.25) was calculated using a vario MACRO analyzer in the carbon-nitrogen (CN) mode (Elementar Analysensysteme GmBH, Hanau, Germany).

2.2.2. Statistical Analyses

We used graphs [30,31,32] of CP and NDF for a 90-day period following fire and locally weighted regression (loess; R Core Team, 2020, v. 4.0.30) as an “exploratory graphical tool” to gain “insight into the behavior of the data [to] help us choose parametric models” [33] (p. 596). Preliminary analyses suggested that, for the most part, CP and NDF changed approximately linearly over time but not monotonically, with periods of rapid change until thresholds were reached, followed by subsequent period(s) of change in different directions and at different rates (Supplemental Equations).
Based on these results, we chose as a class of parametric models piecewise linear regression [34,35,36] with days after burning as an explanatory variable to describe changes in CP and NDF following fire and to test hypotheses related to rates of change within a season as well as between seasons of burning. Regression models with 0, 1, or 2 thresholds were fit for each season of burning. Our set of candidate models, which regressed CP or NDF as the response variable, Y, on days after burning, D, included:
    Y i = β 0 + β 1 D + e i
  Y i = β 0 + γ 1 D + γ 2 D T X 2 + e i
where X 2 =   1   if   D > T   0   if   D T ; so that the slope of the first segment is β 1 = γ 1   , and the slope of the second segment is β 2 = γ 1 + γ 2 ;
Y i = β 0 + γ 1 D + γ 2 D T 1 X 2 + γ 3 D T 2 X 3 + e i
where X 2 =   1   if   D > T 1   0   if   D T 1 and X 3 =   1   if   D > T 2   0   if   D T 2 ; so that the slopes of the first, second and third segments are β 1 = γ 1 , β 2 = γ 1 + γ 2 , and β 2 = γ 1 + γ 2 + γ 3 , respectively (notation follows) [34,37]. Equations 1, 2, and 3 are for simple linear regression (no thresholds), a continuous function with one threshold, and a continuous function with two thresholds, respectively. Regressions were analyzed with the segmented package [38] (R Core Team 2020, v. 4.0.3). We used BIC and AICc criteria for model selection [39]. Slopes of models were compared between seasons following [40] with the lm package and esticon statements using the doBy package [41] in R Core Team (2020, v. 4.0.3). Influence of outliers was assessed with dffit statistics [34] (base package, R Core Team, 2020, v. 4.0.3). For model validation we calculated the prediction coefficient of determination ( r p 2 ) [34]. Initial (pre-burn) forage values are presented graphically but were not included in the regression analyses. For data presentation, estimated slopes for a given species are denoted as, for example, β ^ W 2 where the subscript “W” refers to season of burn (“winter” in this example) and the sub-subscript “2” refers to the line segment (the second segment in this example).
Nutritional values estimated at 90 days after burning were compared to samples collected prior to each burn using a one-sample t test when n = 1 for pre-burn data (cordgrass NDF and seacoast bluestem NDF in summer) and a two-sample t-test [42] when n > 1.

3. Results

3.1. Burn Conditions

Ambient air temperature range was approximately 10 °C cooler during winter burning than summer burning (Table 1). Wind speeds were similar during both seasons of burning. Relative humidity was lower and had a greater range during winter burns than during summer burns. Mean fuel loads and mean high fire temperature were greater in the Gulf cordgrass community than the seacoast bluestem community in both seasons.

3.2. Precipitation

Rainfall throughout the 90 d following the winter burns was 50.3 mm, and 116.3 mm following summer burn treatments (Figure 1). Precipitation totaled 48.3 mm in the month prior to the winter burn treatments, and only 2.5 mm during the month prior to summer burn treatments.

3.3. Gulf Cordgrass CP

Pre-burn Gulf cordgrass mean CP levels were 4.55% (±0.29%) and 4.97% (±0.61%) for winter- and summer-burned patches, respectively (Figure 2). Mean CP was 12.68% (±0.30%) throughout the 90-day period following winter burns (range: 8.1–21.8%); mean CP was 13.43% (±0.34%) throughout the 90-day period after summer burning (range: 6.6–18.9%; Figure 2a).
A single datum for the winter burn with CP > 20% (for which the dffit statistic was an order of magnitude larger than for any other observation) was removed from the data set for analysis (but is shown in Figure 2a). The top model (Table S1) was a two-threshold model that explained 66% ( F 3 , 57 = 37.2 , P < 0.0001 ; r p 2 = 0.61 ) of the variation in CP. For this model, CP increased rapidly ( β ^ W 1 = 2.035 ± 0.668 ,   t 57 = 3, P = 0.0035) until day 6.4 (±0.07) when estimated CP reached 11.8% (±0.38). After 6.4 days, CP increased more slowly ( β ^ W 2 = 0.115 ± 0.026 ,   t 57 = 5.9, P < 0.0001) until day 29.7 (±2.9) when it peaked at 15.4% (±0.36), after which CP declined ( β ^ W 3 = 0.098 ± 0.010 ,   t 57 = −9.4, P < 0.0001) to 9.5% (±0.39) at 90 days post-burning. Estimated CP at 5 DAB (8.9% ± 0.82) did not differ ( t 57 = 0.64 ,   P = 0.5268 ) from estimated CP at 90 DAB. A one-threshold model failed to describe the initial rapid increase in CP but modeled the decline after reaching peak CP very similarly to the two-threshold model (Table S1).
A piecewise regression model with one threshold fit CP patterns in Gulf cordgrass following summer burns (Figure 2a). Days after burning explained 45% of the variation in CP ( F 2 , 43 = 17.8 , P < 0.0001 ; r p 2 = 0.45 ; Table S1). Although peak CP was estimated at 15.3% (±0.43%) at 28 (±6.9) DAB, CP did not change significantly immediately following burning up to the threshold   β ^ S 1 (=0.058 ± 0.041, t43 = 1.4, P > 0.1621). After 28 days, however, CP decreased   ( β ^ S 2 = −0.086 ± 0.015, t43 = −5.7, P < 0.0001; Figure 2a), reaching 10% (±0.66) at 90 DAB, which was lower ( t 43 = 4.4 ,   P < 0.0001 ) than estimated CP at 6 DAB (14.05 ± 0.69).
Top models for winter (two thresholds) and summer (one threshold) burning differed in form (F3,100 = 29.7, P < 0.0001; Figure 2a). The primary differences between seasons involved responses immediately after burning. In particular, CP increased more rapidly whether the first or the second segment of the winter response was compared to the first segment of the summer response: that is, β ^ W 1 β ^ S 1 = 1.992 ± 0.774 t100 = 2.57, P = 0.0115, and β ^ W 2 β ^ S 1 = 0.098 ± 0.047 , t100 = 2.10, P = 0.0379. However, peak CP following winter burning (15.38% at 29.7 DAB) and peak CP following summer burning (15.32% at 28 DAB) did not differ ( C P ^ W , 29.7   D A B C P ^ S , 28   D A B = 0.06 ± 0.55 ,   t 100 = 0.10 ,   P = 0.9176 ).
Importantly, however, responses of CP following these two seasons of burning were similar in four respects. First, CP peaked at similar levels (15.4% and 15.3% following winter and summer burning, respectively). Second, peak CP was reached at similar times after burning (~30 and ~28 DAB for winter and summer, respectively). Third, after reaching peak levels, CP declined at similar rates ( β ^ W 3 β ^ S 2 = 0.012 ± 0.018 , t100 = −0.67, P = 0.4994) until 90 DAB. Fourth, estimated CP at 90 DAB did not differ ( C P ^ W , 90   D A B C P ^ S , 90   D A B = 0.52 ± 0.73 ,   t 100 = 0.72 ,   P = 0.4759 ) between winter (9.47% ± 0.44) and summer (9.99% ± 0.57) burns.

3.4. Seacoast Bluestem CP

Pre-burn seacoast bluestem mean CP levels were 8.18% (±0.51%) and 5.21% (±0.13%) in winter- and summer-burned patches, respectively. Mean CP was 11.66% (±0.23%) throughout the 90-day period following winter burns (range: 8.22–21.63%); after summer burns mean CP was 11.23% (±0.31%) throughout the 90-day period after burning (range: 1.72–32.88%; Figure 2b).
A single datum for the winter burn with CP > 19% (for which the dffit statistic was an order of magnitude larger than for any other observation) was removed from the data set for analysis (but is shown in Figure 2b). The top model (Table S1) was a two-threshold model that explained 82% ( F 3 , 23 = 35.1 , P < 0.0001 ;   r p 2 = 0.67 ) of the variation in CP. For this model, CP increased rapidly ( β ^ W 1 = 0.848 ± 0.161 ,   t 23 = 5.3, P < 0.0001) until day 11.3 (±1.2) when estimated CP reached 16.2% (±0.56), followed by a decline ( β ^ W 2 = 0.308 ± 0.054 ,   t 23 = −5.7, P < 0.0001) until 27.1 (±4.7) DAB when CP was estimated at 11.36% (±0.5). Thereafter, CP declined gradually ( β ^ W 3 = 0.035 ± 0.014 ,   t 23 = −2.6, P = 0.0171) until it reached 9.12% (±0.507) at 90 DAB, a value only marginally lower ( t 23 = 1.72 ,   P = 0.0769 ) than estimated CP at 5 DAB (10.86% ± 0.77).
A one-threshold model for CP following summer burning had a slightly smaller AICc value than a two-threshold model, but a two-threshold model explained more variability (59%) in CP ( F 3 , 19 = 9.1 , P = 0.0006 ; r p 2 = 0.39 ). The two-threshold model described an initial gradual increase ( β ^ S 1 = 0.075 ± 0.036 ,   t 19 = 2.6, P = 0.0529) in CP until 31 DAB (±8.7) when CP reached 11.8% (±0.40), followed by a long period of only slightly-declining CP ( β ^ S 2 = 0.024 ± 0.015 ,   t 19 = −1.59, P = 0.1293) until 78 (±5) DAB when CP reached 10.7% (±0.47). CP abruptly declined ( β ^ S 3 = 0.241 ± 0.072 ,   t 19 = −3.3, P = 0.0034) and was estimated at 7.77% (±0.68) at 90 days post-burning.
The two-threshold models differed (F4,42 = 17.9, P < 0.0001) between seasons of burning. The initial increase following burning was not only shorter but also more rapid ( β ^ W 1 β ^ S 1 = 0.773 ± 0.159 , t42 = 4.9, P < 0.0001) for plants burned in the winter than for plants burned in the summer. Furthermore, CP peaked at higher levels ( C P ^ W , 11.3   D A B C P ^ S , 27.1   D A B = 4.39 ± 0.68 ,   t 42 = 6.5 ,   P < 0.0001 ) following winter burns (11.8%) than summer burns (4.39%). Winter-burned plants also experienced a shorter and more rapid ( β ^ W 2 β ^ S 2 = 0.284 ± 0.054 , t42 = −5.26, P < 0.0001) period of decline following peak seasonal CP than plants burned in the summer. In fact, the decline in CP during the 3rd period for winter burned plants (from 27 to 90 DAB) was similar ( β ^ W 3 β ^ S 2 = 0.011 ± 0.021 , t42 = −0.54, P = 0.5918) to the decline during the second period for summer burned plants (from 27 to 78 DAB). CP was only slightly higher ( C P ^ W , 78   D A B C P ^ S , 78 D A B = 1.15 ± 0.62 ± 0.73 ,   t 42 = 1.87 ,   P = 0.0684 ) in summer-burned plants (10.7%) than in winter-burned plants (9.5%) 78 DAB and did not differ ( C P ^ W , 78   D A B C P ^ S , 78 D A B = 1.32   ± 0.87 ,   t 42 = 1.51 ,   P = 0.1389 ) at 90 DAB between winter (9.1%) and summer (7.8%) burns.

3.5. Gulf Cordgrass NDF

Gulf cordgrass mean NDF was 75.7% and 74.5% prior to winter and summer burning, respectively. Gulf cordgrass mean NDF was 70.1% (±0.57%) after winter burning (range: 63.8–76.0%); after summer burning mean NDF was 69.6% (±0.51%, range: 64.5–74.3%) (Figure 2c).
A two-threshold model (Table S1) that explained 75% (F3,32 = 32.5, P < 0.0001, r p 2 = 0.69 ) of the variation in Gulf cordgrass NDF following winter burns estimated thresholds at 19 (±3.1) and 47 (±6.1) DAB (Figure 2c). NDF decreased ( β ^ W 1   = −0.559 ± 0.097, t32 = −5.7, P < 0.0001) immediately after burning until 19 DAB, reaching an estimated 68.1% (±0.67) after which it stabilized ( β ^ W 2   = −0.035 ± 0.04, t32 = −0.85, P = 0.4014) until 47 DAB at an estimated 67.1% (±0.72); NDF then increased ( β ^ W 3   = 0.218 ± 0.032, t32 = 6.7, P < 0.0001) until 90 DAB with an estimated 76.5% (±0.96).
Gulf cordgrass NDF also decreased ( β ^ S 1   = −0.204 ± 0.09, t24 = −2.3, P = 0.0334) following summer burning until 23.6 ± 6.0 DAB when it was estimated at 67.4% (±0.75). In contrast to winter burning, however, NDF after summer burning increased thereafter ( β ^ S 2   = 0.096 ± 0.027, t24 = 3.6, P = 0.0016) until 90 DAB, reaching an estimated 73.8% (±1.31).
Models describing NDF responses differed (F3,56 = 11.8, P < 0.0001) between seasons of burning. First, the decrease in NDF after burning was more rapid ( β ^ W 1 β ^ S 1 = 0.355 ± 0.135 , t42 = −2.63, P = 0.0110) following winter burning than summer burning. However, despite a sharper decline in NDF following winter burning, minimum NDF values did not differ ( N D F ^ W , 47   D A B N D F ^ S , 23.6   D A B = 0.30 ± 1.05 ,   t 56 = 0.10 ,   P = 0.7758 ) between winter (67.13% at 47 DAB) and summer (67.43% at 23.6 DAB) burning. Additionally, despite the fact that NDF increased following the final threshold identified for each season of burning, this threshold occurred later following winter burning (47 DAB) than following summer burning (23.6 DAB). Finally, although the increase in NDF following minimum levels was more rapid following winter burning ( β ^ W 3 β ^ S 2 = 0.122 ± 0.043 ,  t56 = 2.83, P = 0.0065), NDF levels did not differ ( N D F ^ W , 90   D A B N D F ^ S , 90   D A B = 2.71 ± 1.58 ,   t 56 = 1.71 ,   P = 0.0930 ) at 90 DAB between winter (76.5% ± 1.06) and summer (73.8% ± 1.18) burning.

3.6. Seacoast Bluestem NDF

Seacoast bluestem mean NDF was 74.4% (±0.91%) and 76.8% (±0.75%) before winter and summer burning, respectively. Mean NDF was 71.4% (±0.32%) after winter burning (range: 45.3–85.7%), and 69.7% (±0.33%) after summer burning (range: 58.3–79.4%; Figure 2d).
A two-threshold model (Table S1) best described changes in seacoast bluestem NDF following winter burns (F3,32 = 12.9, P < 0.0001, r 2 = 0.55 ,   r p 2 = 0.41 ). NDF decreased from an initial estimated 72% (±1.48) 5 DAB to 65% (±1.23) at 11 ± 2.7 DAB, after which it increased relatively rapidly to 71.8% (±0.61) at 21.2 ± 2.9 DAB; thereafter, NDF increased more slowly, peaking at 74% (±0.89) at 90 DAB.
A simple linear regression best described changes in seacoast bluestem NDF following summer burns (F1,27 = 5.5, P = 0.0265, r 2 = 0.17 ,   r p 2 = 0.07 ). This model predicted an increase in NDF from an estimated 67.5% (±0.9) at 6 DAB to 72.2% (±1.34) 90 DAB but explained < 20% of the variation in NDF. Although models differed (F2,59 = 4.54, P = 0.0147) between seasons of burning, the slow increase in NDF following summer burning did not differ from the slow increase in NDF following winter burning observed between days 21 and 90 after burning ( β ^ w i n t e r β ^ s u m m e r = 0.023 ± 0.030 , t59 = −0.79, P = 0.4336). Further, minimum NDF values did not differ ( N D F ^ W , 11 D A B N D F ^ S , 6   D A B = 2.46 ± 1.60 ,   t 59 = 1.51 ,   P = 0.1370 ) between winter (65% ± 1.43) and summer seasons (67.5% ± 0.79); and NDF did not differ ( N D F ^ W , 90 D A B N D F ^ S , 90   D A B = 1.8 ± 1.56 ,   t 59 = 1.16 ,   P = 0.2526 ) at 90 DAB following winter (74% ± 1.02) and summer (72.2% ± 1.18) burning.

3.7. Duration of Burning Effects: Pre-Burn vs. 90 DAB Nutritional Values

Cordgrass CP was higher (t3.2 = 32, P < 0.0001) at 90 DAB (9.5% ± 0.39) than prior to winter burning (9.5% ± 0.39); a similar response (t3.4 = 16, P = 0.0002) was observed at 90 days after summer burning (10% ± 0.66) compared to pre-burn values (4.97% ± 0.61). Seacoast bluestem CP also was higher (t1 = 16, P = 0.0396) at 90 days after summer burning (7.77% ± 0.68) compared to pre-burn values (5.21% ± 0.13). In contrast to cordgrass, however, we detected no difference (t1 = 2.5, P = 0.2142) 90 days after winter burning (9.12% ± 0.51) compared to pre-burn values (8.18% ± 0.51). Cordgrass NDF 90 days after winter burning (76.5% ± 0.96) was higher (t36 = 6, P < 0.0001) than prior to burning (75.7%) but lower (t26 = 2.8, P = 0.0101) 90 days after summer burning (73.8% ± 1.31) than prior to burning (74.5%). Seacoast bluestem NDF at 90 days after summer burning (72.2% ± 1.18) was lower (t28 = 21, P < 0.0001) than observed prior to burning (76.8%), but not different (t1 = 0.6, P = 0.6443) 90 days after winter burning (74 ± 0.89) compared to pre-burn values (74.4% ± 0.91).

4. Discussion

This study was not designed to test hypotheses about the interaction between fire and grazing on nutritive value: burned patches were continuously grazed by the cooperating ranch both prior to and following burning. Grazing animals [44], fire [45], and weather [46] affect rangeland vegetation composition and structure as well as nutritive value e.g., [47,48,49]. “Pyric herbivory”—the interaction between fire and grazing—has been documented in grasslands worldwide e.g., [24,50], and it is widely recognized that “As fire and herbivory both remove above-ground biomass, they clearly compete—but interactions of herbivores and fire can be more complex than this” [49] (p. 2). When aboveground biomass of perennial grasses is removed in a single event—whether by fire or by grazing—plant regrowth from reserves stored primarily in stem bases quickly replaces aboveground tissues [22]. We observed regrowth of Gulf cordgrass and seacoast bluestem within days after prescribed burning, a response that was apparent whether plants were burned in winter or summer and supported by McAtee et al. [51]. It has been widely documented that this regrowth is high in nutritive value for grazing animals, and its underlying mechanisms are well understood: herbaceous regrowth is high in protein but low in structural components (fiber); as forage matures, however, fiber increases and protein decreases [49,52,53,54]. Species-specific responses arise because of differences in season of growth (e.g., cool vs warm season species) and plant morphology (e.g., stoloniferous/rhizomatous vs caespitose growth form).
We analyzed the nutritional content of Gulf cordgrass and seacoast bluestem following the winter or summer 2016 prescribed burning treatments. We recognized that seasonal changes in nutritional content will occur without burning, and we did not examine that, thus we cannot compare the changes we recorded after burning to non-burning. Although both Gulf cordgrass and seacoast bluestem are dominant native, warm-season bunchgrasses in southern Texas coastal prairies, they differ in growth form. Gulf cordgrass plant basal diameters average 29 cm [51]; plants can reach 1.5 m tall [5] and are strictly caespitose. Although basal diameters of seacoast bluestem can be 25–30 cm [55], plants generally are smaller but also have the ability to produce rhizomes that promote lateral spread [1].
Gulf cordgrass CP increased after burning, and although peak CP following burning did not differ between seasons, the rate at which the peak was reached was faster following winter burning. This response might be attributable to warmer-than-normal soil surface temperatures in spring (because of the removal of vegetation by fire), enhanced soil bacterial growth, and subsequent plant nutrient uptake [56,57]. The decline in CP following peak post-burning levels did not differ between seasons of burning, nor did CP levels differ 90 DAB between seasons of burning. Gulf cordgrass NDF showed the opposite pattern following burning, declining more rapidly following summer burning than winter burning but reaching minimum NDF values that did not differ between seasons of burning. Although NDF remained low for a longer period following winter burning (i.e., NDF began to increase nearly 20 days sooner following summer burning than winter burning), the rate of increase in NDF was faster following winter burning. Despite these differences in rates of response, 90-day NDF values did not differ between seasons. During the 90-day period following both summer and winter burning of Gulf cordgrass, crude protein content was greater than 9%, which is the minimum threshold for the maintenance of a lactating beef cow [58]. Comparing only beginning (soon after burning) and ending (90 DAB) nutritional values between seasons of burning fails to take into account important changes—thresholds (one or two? or none?) and interval length between them—that take place throughout the 90-day monitoring period e.g., [59]. Taken together, our results suggest that winter burning improved the forage quality of Gulf cordgrass more than summer burning: i.e., following winter burning, CP increased more rapidly, and NDF remained depressed for a longer period of time compared to summer burning. Although scientific literature is replete with general statements about the importance of season of burning on its effects on vegetation e.g., [60] (p. 51), there has been relatively little research (<1% of 338 papers reviewed by Limb et al. [61]) conducted on effects of fire on nutritive value. For example, season of burning has been shown to affect the nutritive value of purple threeawn (Aristida purpurea; [62]) but had no effect on CP in grasses in South Africa [63].
Although CP in seacoast bluestem also increased following winter burning, there was relatively little change following summer burning, a result that is difficult to explain given that (1) a common response to defoliation (whether through grazing or burning) of grasses is stimulation of new leaf growth that is higher in CP than older leaf material, and (2) summer-burned plants experienced several ~1-cm rainfall events within ~45 DAB. Buttery and Ehreneich [64], however, cited and confirmed results of Campbell et al. [65] that Andropogon divergens crude protein increased when plants were burned in a young leaf stage but had no effect when plants were burned in full or mature leaf stages. As with Gulf cordgrass, the CP response of seacoast bluestem following winter burning may be attributable to increased soil surface temperatures following burning and enhanced soil microbial activity. The rapid and substantial increase in CP following winter burning, coupled with concomitant low NDF, suggest that, like Gulf cordgrass, seacoast bluestem responded more favorably to winter than to summer burning. Similar to Gulf cordgrass, the differences between seasons of burning on seacoast bluestem nutritive value were most apparent soon after burning; after respective thresholds were reached, declines in CP and increases in NDF were similar between seasons of burning.
Duration of burning effects have important management implications. In our study, Gulf cordgrass CP was ~5% higher at 90 days after burning compared to pre-burn values for both winter- and summer-burned plants. Seacoast bluestem CP was ~2.5% higher 90 days after summer burning but neither higher nor lower 90 days after winter burning. NDF of both Gulf cordgrass and seacoast bluestem was lower 90 days after summer burning compared to pre-burn values; NDF values following winter burning were higher in cordgrass and not different in seacoast bluestem. Although statistical differences in duration of burning effects on CP were large enough to have management implications, it is not likely that duration of burning effects on NDF, even when statistically significant, were large enough to have practical management implications.
Prescribed fire can be used to accomplish a number of different management objectives in coastal grasslands of southern Texas. For example, late summer burns might be more effective than early winter burns in providing wintering habitat and resting stops for migratory birds in the Gulf Coast region of Texas [66]. However, human health hazards associated with summer prescribed burning in this region (heat exhaustion and heat stroke: [67]) are also important considerations. These can be mitigated by burning during more tolerable winter conditions.

5. Conclusions

Our goal was to assess changes in CP and NDF in Gulf cordgrass and seacoast bluestem following winter or summer burning in a patch-burn grazing system. Our results indicated that nutritive value responses of two dominant grasses in the Gulf Coast Prairies and Marshes Ecoregion to winter or summer prescribed fire generally were similar, to the extent that CP and NDF of both grasses responded more favorably to winter burning. Although Gulf cordgrass CP was no different 90 days after winter burning than immediately after winter burning, CP 90 days after summer burning was lower than immediately after summer burning, a difference that is likely attributable to a season-of-year effect.
Our results show that although the nutritive values of Gulf cordgrass and seacoast bluestem were enhanced by prescribed burning, whether applied in late summer or in early winter, winter burning was more beneficial. Understanding how rangeland is affected by prescribed fire in different seasons can help managers achieve specific goals of their ranching enterprise.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/fire6030105/s1, Table S1: Fit statistics, test statistics, estimated thresholds, and estimated slopes of line segments for models with zero, one or two thresholds to describe changes in CP and NDF in Gulf cordgrass and seacoast bluestem after winter or summer burning on the East Foundation’s El Sauz Ranch in Willacy and Kenedy Counties, Texas, in 2016. Estimated equations provided at the bottom of the table.

Author Contributions

Conceptualization, J.A.O.-S., T.A.C., A.O.-S.J., S.R.-H. and D.B.W.; methodology, S.R.-H., D.B.W. and J.A.O.-S.; formal analysis, D.B.W.; investigation, V.L.H., J.S.A.-S., A.E.T., S.R.-H. and D.B.W.; resources, T.A.C., A.O.-S.J. and S.R.-H.; data curation, S.R.-H. and D.B.W.; writing—original draft preparation, V.L.H.; writing—review and editing, S.R.-H. and D.B.W.; validation, S.R.-H. and D.B.W.; visualization, S.R.-H. and D.B.W.; supervision, S.R.-H.; project administration, S.R.-H.; funding acquisition, S.R.-H., D.B.W. and J.A.O.-S. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the East Foundation (Master Research Agreement 2012-00001). This research received additional financial support from Rene Barriéntos Scholarship Fund, Houston Safari Club, South Texas Quail Coalition, Lon and Leigh Cartwright Graduate Scholarship Fund, Caesar Kleberg Wildlife Research Institute, and Texas A&M University-Kingsville.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data used in this study are available on request from the corresponding author.

Acknowledgments

We thank the TAMUK Prescribed Burning Crew and the following undergraduate and graduate students for help conducting prescribed fires and for collecting field data: K. Allred, R. Combs, J. Cortez, A. DiMaggio, D. Hall, M. Ogden, T. Wilkins, and H. Winters. F. Hernandez and R. Stanko provided helpful comments. This manuscript is Caesar Kleberg Wildlife Research Institute Publication Number 20-138, and East Foundation Publication Number 058.

Conflicts of Interest

East Foundation personnel had involvement in the conceptualization of the study and in providing resources.

References

  1. Hatch, S.; Gandhi, K.; Brown, L. Checklist of the Vascular Plants of Texas; Texas Agricultural Experiment Station: College Station, TX, USA, 1990; p. MP-1655. [Google Scholar]
  2. Pitman, W.D. Pastures of the U.S. Western Gulf Coast Region. In Pastures, Dynamics, Economics and Management; Prochazka, N.T., Ed.; Nova Science Publ., Inc.: Hauppauge, NY, USA, 2011; pp. 85–107. [Google Scholar]
  3. U.S. Climate Data—Raymondville, TX. Available online: https://www.usclimatedata.com/climate/raymondville/texas/united-states/ustx1120 (accessed on 5 November 2015).
  4. Yang, Y.; Wilson, L.T.; Jifon, J.; Landivar, J.A.; da Silva, J.; Maeda, M.M.; Want, J.; Christensen, E. Energycane growth dynamics and potential early harvest penalties along the Texas Gulf Coast. Biomass Bioenerg. 2018, 113, 1–14. [Google Scholar] [CrossRef]
  5. Everitt, J.H.; Drawe, D.L.; Little, C.R.; Lonard, R.I. Grasses of South Texas; Texas Tech University Press: Lubbock, TX, USA, 2011; p. 336. [Google Scholar]
  6. Scifres, C.J.; McAtee, J.W.; Drawe, D.L. Botanical, edaphic, and water relationships of Gulf cordgrass (Spartina spartinae [Trin.] Hitchc.) and associated communities. Southwest. Nat. 1980, 25, 397–409. [Google Scholar] [CrossRef]
  7. Cooper, D.J.; Wagner, J.I. Tropical storm driven hydrologic regimes support Spartina spartinae dominated prairies in Texas. Wetlands 2013, 33, 1019–1024. [Google Scholar] [CrossRef]
  8. Scifres, C.J.; Drawe, D.L. Gulf cordgrass: Distribution, ecology and responses to prescribed burning. In Prescribed Range Burning in the Coastal Prairie and Eastern Rio Grande Plains of Texas; Hanselka, C.W., Ed.; Texas Agricultural Experiment Service Bulletin: College Station, TX, USA, 1980; pp. 83–92. [Google Scholar]
  9. Garza, A.J.; McLendon, T.; Drawe, D.L. Herbage yield, protein content, and carbohydrate reserves in gulf cordgrass (Spartina spartinae). J. Range Manag. 1994, 47, 16–21. [Google Scholar] [CrossRef]
  10. Gulf Cordgrass, Spartina Spartinae Fact Sheet. Available online: https://plants.usda.gov (accessed on 11 November 2015).
  11. McAtee, J.W.; Scifres, C.J.; Drawe, D.L. Improvement of gulf cordgrass range with burning or shredding. J. Range Manag. 1979, 32, 372–375. [Google Scholar] [CrossRef]
  12. Haynes, V.L.; Avila-Sanchez, J.S.; Rideout-Hanzak, S.; Wester, D.B.; Ortega, J.A. Effects of prescribed burning on gulf cordgrass, Spartina spartinae (Trin.) Merr. ex Hitchc. MOJ Ecol. Environ. Sci. 2018, 3, 110–115. [Google Scholar]
  13. Savage, D.A.; Heller, V.G. Nutritional Qualities of Range Forage Plants in Relation to Grazing with Beef Cattle on the Southern Plains Experimental Range; Tech. Bull. No. 943; USDA: Washington, DC, USA, 1947; p. 61. [Google Scholar]
  14. Grelen, H.E.; Hughes, R.H. Common Herbaceous Plants of Southern Forest Range; Res. Pap. SO-210; USDA Forest Service: New Orleans, LA, USA, 1984; p. 147. [Google Scholar]
  15. Fire Effects Information System. Available online: https://www.feis-crs.org/feis/ (accessed on 5 January 2023).
  16. Dyksterhuis, E.J. The vegetation of the western Cross Timbers. Ecol. Monogr. 1948, 18, 326–376. [Google Scholar] [CrossRef]
  17. Frasier, J.R. The Effect of Three Grazing Management Systems on Cattle Diets on the Welder Wildlife Refuge. Master’s Thesis, Texas Tech University, Lubbock, TX, USA, 1979. [Google Scholar]
  18. Drawe, D.L.; Box, T.W. Forage ratings for deer and cattle on the Welder Wildlife Refuge. J. Range Manag. 1968, 21, 225–228. [Google Scholar] [CrossRef]
  19. Lonard, R.I.; Judd, F.W. The biological flora of coastal dunes and wetlands: Schizachyrium littorale (G. Nash) E. Bicknell. J. Coast. Res. 2010, 26, 654–662. [Google Scholar]
  20. Dahl, B.E. Developmental morphology of plants. In Wildland Plants: Physiological Ecology and Developmental Morphology; Bedunah, D.J., Sosebee, R.E., Eds.; Society for Range Management: Denver, CO, USA, 1995; pp. 22–58. [Google Scholar]
  21. Lemaire, G. Ecophysiology of grasslands: Dynamic aspects of forage plant populations in grazed swards. In Proceedings of the International Grassland Congress Proceedings, XIX International Grasslands Congress, São Pedro, SP, Brazil, 11–21 February 2001; FEALQ, UKnowledge: Lexington, KY, USA, 2001. [Google Scholar]
  22. Sosebee, R.E.; Wester, D.B.; Villalobos, J.C.; Britton, C.M.; Wan, C.; Nofal, H. How grasses grow—How plant growth relates to grazing management. In Proceedings of the Second National Conference on Grazing Lands, a Symposium Sponsored by the Grazing Lands Conservation Initiative, Nashville, TN, USA, 7–10 December 2003. Also published in Cao Ye Xue Bao 2005, 14, 117–125. (In Chinese). [Google Scholar]
  23. Blair, J.; Nippert, J.; Briggs, J. Grassland ecology. In Ecology and the Environment; Monson, R.K., Ed.; Springer: Dordrecht, Switzerland, 2014; pp. 389–424. [Google Scholar]
  24. Allred, B.W.; Fuhlendorf, S.D.; Engle, D.M.; Elmore, R.D. Ungulate preference for burned patches reveals strength of fire-grazing interaction. Ecol. Evol. 2011, 1, 132–144. [Google Scholar] [CrossRef]
  25. Spiess, J.W.; McGranahan, D.A.; Geaumont, B.; Sedivec, K.; Lakey, M.; Berti, M.; Jovick, R.J.; Limb, R.F. Patch-burning buffers re-sources and livestock performance to mitigate drought in the northern Great Plains. Rangel. Ecol. Manag. 2020, 73, 473–481. [Google Scholar] [CrossRef]
  26. Web Soil Survey. Available online: http://websoilsurvey.nrcs.usda.gov/ (accessed on 2 October 2015).
  27. Normal Monthly Station Details for Port Mansfield, National Weather Service. Available online: https://www.ncdc.noaa.gov/cdoweb/datasets/NORMAL_MLY/stations/ (accessed on 7 October 2015).
  28. Scott, J.H.; Burgan, R.E. Standard Fire Behavior Fuel Models: A Comprehensive Set for Use with Rothermel’s Surface Fire Spread Model; Gen. Tech. Rep. RMRS-GTR-153; USDA Forest Service: Fort Collins, CO, USA, 2005; p. 72. [Google Scholar]
  29. Van Soest, J.P.; Robertson, J.B.; Lewis, B.A. Methods of dietary fiber, neutral detergent fiber and non-starch polysaccharides in relation to animal nutrition. J. Dairy Sci. 1991, 74, 3583–3597. [Google Scholar] [CrossRef] [PubMed]
  30. Toms, J.D.; Lesperance, M.L. Piecewise regression: A tool for identifying ecological thresholds. Ecology 2003, 84, 2034–2041. [Google Scholar] [CrossRef]
  31. Ryan, S.E.; Porth, L.S. A Tutorial on the Piecewise Regression Approach Applied to Bedload Transport Data; Gen. Tech. Rep. RMRS-GTR-198; USDA Forest Service: Fort Collins, CO, USA, 2007; p. 41. [Google Scholar]
  32. Kelly, M.G.; Phillips, G.; Teixeira, H.; Varbiro, G.; Herrero, F.S.; Willby, N.J.; Poikane, S. Establishing ecologically-relevant nutrient thresholds: A tool-kit with guidance on its use. Sci. Total Environ. 2022, 807, 150977. [Google Scholar] [CrossRef]
  33. Cleveland, W.S.; Devlin, S.H. Locally weighted regression: An approach to regression analysis by local fitting. J. Amer. Stat. Assoc. 1988, 86, 596–640. [Google Scholar] [CrossRef]
  34. Montgomery, D.C.; Peck, E.A.; Vining, G.G. Introduction to Linear Regression Analysis, 5th ed.; John Wiley & Sons: Now York, NY, USA, 2012; p. 672. [Google Scholar]
  35. Wester, D.B. Regression: Linear and non-linear, parametric and non-parametric. In Quantitative Analyses in Wildlife Sciences; Brennan, L.A., Tri, A.N., Marcot, B.G., Eds.; Johns Hopkins University Press: Baltimore, MD, USA, 2019; pp. 9–31. [Google Scholar]
  36. Matthews, T.J.; Rigal, F. Thresholds and the species-area relationship: A set of function for fitting, evaluating and plotting a range of commonly used piecewise models in R. Front. Biogeogr. 2021, 13, e49404. [Google Scholar] [CrossRef]
  37. Neter, J.; Kutner, M.H.; Nachtsheim, C.J.; Wasserman, W. Applied Linear Regression Models, 3rd ed.; McGraw-Hill: New York, NY, USA, 1996; p. 720. [Google Scholar]
  38. Muggeo, V.M.R. segmented: An R package to fit regression models with broken-line relationships. R News 2008, 8, 20–25. [Google Scholar]
  39. Gao, D.; Cao, Z.; Xu, P.; Perry, G. On piecewise models and species-area patterns. Ecol. Evol. 2019, 9, 8351–8361. [Google Scholar] [CrossRef]
  40. Graybill, F.A. Theory and Application of the Linear Model; Duxberry Press: North Scituate, MA, USA, 1976; p. 704. [Google Scholar]
  41. Højsgaard, S.; Halekoh, U. doBY: Groupwise Statistics, Lsmeans, Linear Contrasts, Utilities. R Package v 4.6.11. 2021. Available online: https://CRAN.R-project.org/package=doBy (accessed on 5 March 2023).
  42. Ruxton, G.D. The unequal variance t-test is an underused alternative to Student’s t-test and the Mann-Whitney U test. Behav. Ecol. 2006, 17, 688–690. [Google Scholar] [CrossRef]
  43. Hanselka, C.W. Improving Gulf Cordgrass Range; Texas Agricultural Extension Service Bulletin: College Station, TX, USA, 1981. [Google Scholar]
  44. Rouquette, F.M. The roles of forage management, forage quality, and forage allowance in grazing research. Prof. Anim. Sci. 2016, 32, 10–18. [Google Scholar] [CrossRef]
  45. Proffitt, K.M.; DeVoe, J.; Barker, K.; Durham, R.; Hayes, T.; Hebblewhite, M.; Jourdonnais, C.; Ramsey, P.; Shamhart, J. A century of changing fire management alters ungulate forage in a wildfire-dominated landscape. Forestry 2019, 92, 523–537. [Google Scholar] [CrossRef]
  46. McCuistion, K.; Grigar, M.; Wester, D.B.; Rhoades, R.; Mathis, C.; Tedeschi, L. Can we predict forage nutritive value with weather parameters? Rangelands 2014, 36, 2–9. [Google Scholar] [CrossRef] [Green Version]
  47. Stuth, J.; Maraschin, G.E. Sustainable management of pastures and rangelands. In Grassland Ecophysiology and Grazing Ecology; Lemaire, G., Hodgson, J., Moraes, A.D., Nabinger, C., Carvalho, P.D.F., Eds.; CABI Publ.: Wallingford, UK, 2000; pp. 339–354. [Google Scholar]
  48. Raynor, E.J.; Joern, A.; Briggs, J.M. Bison foraging responds to fire frequency in nutritionally heterogeneous grassland. Ecology 2015, 96, 1586–1597. [Google Scholar] [CrossRef] [Green Version]
  49. Archibald, S.; Hempson, G.P. Competing consumers: Contrasting the patterns and impacts of fire and mammalian herbivory in Africa. Philos. Trans. R. Soc. B Biol. Sci. 2016, 371, 20150309. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  50. Thapa, S.K.; de Jong, J.F.; Hof, A.R.; Subedi, N.; Joshi, L.R.; Prins, J.J.T. Fire and forage quality: Postfire regrowth quality and pyric herbivory in subtropical grasslands in Nepal. Ecol. Evol. 2022, 12, e8794. [Google Scholar] [CrossRef] [PubMed]
  51. McAtee, J.W.; Oefinger, R.D.; Scifres, C.J. Predicting topgrowth yield of Gulf cordgrass from basal diameter and volume measurements. Southwest. Nat. 1978, 23, 197–204. [Google Scholar] [CrossRef]
  52. Van de Vijver, C.A.D.M.; Poot, P.; Prins, J.J.T. Causes of increased nutrient concentrations in post-fire regrowth in an east African savanna. Plant Soil 1999, 214, 173–185. [Google Scholar] [CrossRef]
  53. Mbatha, K.R.; Ward, D. The effects of grazing, fire, nitrogen and water availability on nutritional quality of grass in semi-arid savanna, South Africa. J. Arid Environ. 2010, 74, 1294–1301. [Google Scholar] [CrossRef]
  54. Pavlu, K.; Kassahun, T.; Pavlu, V.V.; Pavlu, L.; Blazek, P.; Homolka, P. The effects of first defoliation and previous management intensity on forage quality of a semi-natural species-rich grassland. PLoS ONE 2021, 16, e0248804. [Google Scholar] [CrossRef]
  55. Valverde, T.; Pisanty, I. Growth and vegetative spread of Schizachyrium scoparium var. littoralis (Poaceae) in sand dune microhabitats along a successional gradient. Canad. J. Bot. 1999, 77, 219–229. [Google Scholar] [CrossRef]
  56. Sharrow, S.H.; Wright, S.H. Effects of fire, ash, and litter on soil nitrate, temperature, moisture and tobosagrass production in the rolling plains. J. Range Manag. 1977, 30, 266–270. [Google Scholar] [CrossRef]
  57. Fultz, L.M.; Moore-Kucera, J.; Dathe, J.; Davinic, M.; Perry, G.; Wester, D.; Schwilk, D.; Rideout-Hanzak, S. Forest wildfire and grassland prescribed fire effects on soil biogeochemical processes and microbial communities: Two case studies in the semi-arid Southwest. Appl. Soil Ecol. 2016, 99, 118–128. [Google Scholar] [CrossRef]
  58. Holechek, J.L.; Pieper, R.D.; Herbel, C.H. (Eds.) Range animal nutrition. In Range Management Principles and Practices, 2nd ed.; Prentice-Hall, Inc.: Englewood Cliffs, NJ, USA, 1995; pp. 325–370. [Google Scholar]
  59. Foley, M.M.; Martone, R.G.; Fox, M.D.; Kappel, C.V.; Mease, L.A.; Erickson, A.L.; Halpern, B.S.; Selkeo, K.A.; Taylor, P.; Carborough, C. Using ecological thresholds to inform resource management: Current options and future possibilities. Front. Mar. Sci. 2015, 2, 95. [Google Scholar] [CrossRef] [Green Version]
  60. Lyon, L.J.; Hooper, R.G.; Telfer, E.S.; Schreiner, D.S. Fire effects on wildlife foods. In Wildland Fire in Ecosystems: Effects of Fire on Fauna; RMRS-GTR-42; Smith, J.K., Ed.; USDA Forest Service, Rocky Mountain Research Station: Ogden, UT, USA, 2000; Volume 1, pp. 51–58. [Google Scholar]
  61. Limb, R.F.; Fuhlendorf, S.D.; Engle, D.M.; Miller, R.F. Synthesis paper: Assessment of research of rangeland fire as a management practice. Rangel. Ecol. Manag. 2016, 69, 415–422. [Google Scholar] [CrossRef] [Green Version]
  62. Dufek, N.A.; Vermeire, L.T.; Waterman, R.C.; Ganguli, A.C. Fire and nitrogen addition increase forage quality of Aristida purpurea. Rangel. Ecol. Manag. 2014, 67, 298–306. [Google Scholar] [CrossRef]
  63. Everson, C.S.; Everson, T. The long-term effects of fire regime on primary production of montane grasslands in South Africa. Afr. J. Range. For. Sci. 2016, 33, 33–41. [Google Scholar] [CrossRef]
  64. Buttery, R.F.; Ehrenreich, J.H. Nutritive Quality of Little Bluestem in the Missouri Ozarks; Technical Paper 179; USDA, Forest Service, Central States Forest Experiment Station: Columbus, OH, USA, 1961; p. 9. [Google Scholar]
  65. Campbell, R.S.; Epps, E.A., Jr.; Moreland, C.C.; Farr, J.L.; Bonner, F. Nutritive Values of Native Plants on Forest Range in Central Louisiana; Report 277; Louisiana State University Agricultural Experiment Station: Baton Rouge, LA, USA, 1954; p. 18, (originally published as Louisiana Agricultural Experiment Station Bulletin No. 488). [Google Scholar]
  66. Britton, C.M.; Rideout-Hanzak, S.; Brown, S.D. Effects of burns conducted in summer and winter on vegetation of Matagorda Island, Texas. Southwest. Nat. 2010, 55, 193–202. [Google Scholar] [CrossRef]
  67. Wade, D.D.; Lunsford, J.D. A Guide for Prescribed Fire in Southern Forests; Technical Publication R8-TP-11; USDA, Forest Service, Southern Region: Atlanta, GA, USA, 1989; p. 56. [Google Scholar]
Figure 1. Rainfall following winter and summer burning on the East Foundation’s El Sauz Ranch in Willacy and Kenedy Counties, Texas, in 2016. Black represents winter and red represents summer.
Figure 1. Rainfall following winter and summer burning on the East Foundation’s El Sauz Ranch in Willacy and Kenedy Counties, Texas, in 2016. Black represents winter and red represents summer.
Fire 06 00105 g001
Figure 2. Crude protein (%) and neutral detergent fiber (%) content of Gulf cordgrass and seacoast bluestem following winter and summer burning on the East Foundation’s El Sauz Ranch in Willacy and Kenedy Counties, Texas, in 2016. Gray shaded area in (a,b) indicates the maintenance levels for lactating cows [43]. Black represents winter, and red represents summer. Hollow circles are pre-burn values, shown for illustrative purposes but not included in regression analyses. For winter and summer burns, selected models were: (a) 2-threshold and 1-threshold; (b) 2-threshold and 1-threshold, (c) 2-threshold and 2-threshold, and (d) 2-threshold and 0-threshold, respectively.
Figure 2. Crude protein (%) and neutral detergent fiber (%) content of Gulf cordgrass and seacoast bluestem following winter and summer burning on the East Foundation’s El Sauz Ranch in Willacy and Kenedy Counties, Texas, in 2016. Gray shaded area in (a,b) indicates the maintenance levels for lactating cows [43]. Black represents winter, and red represents summer. Hollow circles are pre-burn values, shown for illustrative purposes but not included in regression analyses. For winter and summer burns, selected models were: (a) 2-threshold and 1-threshold; (b) 2-threshold and 1-threshold, (c) 2-threshold and 2-threshold, and (d) 2-threshold and 0-threshold, respectively.
Fire 06 00105 g002
Table 1. Burning weather, fuel conditions, and fire temperatures averaged by season of treatment on the East Foundation’s El Sauz Ranch in Willacy and Kenedy Counties, Texas, in 2016.
Table 1. Burning weather, fuel conditions, and fire temperatures averaged by season of treatment on the East Foundation’s El Sauz Ranch in Willacy and Kenedy Counties, Texas, in 2016.
Burn ConditionsWinter 2016Summer 2016
Temperature range °C 20–2734–38
Relative humidity range %28–5051–61
Wind speed range m·s−11.8–5.41.3–5.5
Gulf cordgrass fuel load kg·ha−114,54412,775
Seacoast bluestem fuel load kg·ha−1 88989437
Gulf cordgrass fuel moisture % 35.731.9
Gulf cordgrass mean high fire temperature °C726838
Seacoast bluestem mean high fire temperature °C512532
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Haynes, V.L.; Avila-Sanchez, J.S.; Toomey, A.E.; Ortega-S., J.A.; Campbell, T.A.; Ortega-S., A., Jr.; Rideout-Hanzak, S.; Wester, D.B. Patch Burning Improves Nutritional Quality of Two Gulf Coast Grasses—And Winter Burning Is Better than Summer Burning. Fire 2023, 6, 105. https://doi.org/10.3390/fire6030105

AMA Style

Haynes VL, Avila-Sanchez JS, Toomey AE, Ortega-S. JA, Campbell TA, Ortega-S. A Jr., Rideout-Hanzak S, Wester DB. Patch Burning Improves Nutritional Quality of Two Gulf Coast Grasses—And Winter Burning Is Better than Summer Burning. Fire. 2023; 6(3):105. https://doi.org/10.3390/fire6030105

Chicago/Turabian Style

Haynes, Victoria L., J. Silverio Avila-Sanchez, Adam E. Toomey, J. Alfonso Ortega-S., Tyler A. Campbell, Alfonso Ortega-S., Jr., Sandra Rideout-Hanzak, and David B. Wester. 2023. "Patch Burning Improves Nutritional Quality of Two Gulf Coast Grasses—And Winter Burning Is Better than Summer Burning" Fire 6, no. 3: 105. https://doi.org/10.3390/fire6030105

Article Metrics

Back to TopTop