Eulleim’ 72¢ “ - Relation offoilg Rainfall and ' Grazing Management to Vegetation — — Western Edwards Plateau of T exar lvoaemfim I954 TEXAS AGRICULTURAL EXPERIMENT STATION R. D. LEWIS. DIRECTOR, COLLEGE STATION, TEXAS A TEXAS anus: swarm: A n uuuz ,. B \\ I72 Acres A + ~ LEGEND - < 172 Acres \ § wmDMlLL AND TANK é WAT£R TROUGN --— FENCE 230 Acres = i c A 3Z0 Acres (Form \ ' fi 4| Acres , I h,‘ 364 Acres ‘i621 x H < E 3i K I ‘ 86 Acres I72 Acres . 17g Aq-es , F 86 Acres k/nxwar A M E K 86 Acres L M I72. Acres , I72 ACPQS X H 66 Aeres i < I , x—-———i<-—-— p , 320 Acres < N X O use Acres 3Z0 Acres if’. (qMI T Figure 1. Map of the Texas Range Station showing pasture boundaries, acreages and livestock water facilities. DIGEST Experiments were conducted from 1938-53 at the Texas Range Station near Barnhart to study the ects of climate, soils and grazing on the vegetation. This station comprises approximately 3,160 acres land, is owned by the University of Texas and is operated by the Texas Agricultural Experiment tion. . ' t l . The 16 pastures on the station have been subjected to different rates of stocking with various binations of sheep and cattle since grazing experiments were started in 1938. res of rangeland on the Edwards Plateau of Texas. The most important forage species are tobosa, ffalo and curly mesquitegrass. Associated with these grasses, but in smaller amounts are three-awn ‘g. ses, side-oats grama, vine mesquite and many others. The poisonous bitterweed and annual l I oomweed are the most abundant weeds. Dominant woody species are mesquite, pricklypear and cholla. i t» i.’ Vegetation on the experimental pastures is fairly representative of that supported by several million The nature of the soil has had a pronounced effect on the kind and amount of vegetation on the area. .» “he soil also has influenced the response of the vegetation to grazing and rainfall. , A cover of tobosagrass is superior, from the standpoint of soil and water conservation, to a cover V, buffalo or curly mesquite, or to bare ground, which is subject to annual weed growth during periods favorable rainfall. Tobosagrass favors more rapid rates of water intake into the soil, more stable i _il aggregation and a more favorable soil temperature for plant growth. A close relationship was found between available forage and annual rainfall for the period 1938-53. owever, fluctuations in the amount of forage usually lag from 1 to 3 years behind the fluctuations in F nual rainfall. On areas heavily grazed by livestock, the vegetation does not respond immediately to et or dry years. Certain changes in floristic composition are caused by the rainfall pattern and should ‘i 7 recognized and separated from changes due to grazing. The extreme drouth prevailing since the fall of 1950 has caused serious reductions in ground cover d forage production. Death losses have been extremely high in curly mesquitegrass and almost as (Q _1 eat in buffalo and purple three-awn. Tobosa was the most drouth resistant of the grasses in the area. Studies of the response of the vegetation to grazing management show the following: i (1) Sideoats grama, vine mesquite and cane bluestem are the best indicators of past stocking rate I the area. These grasses are more abundant on the lightly stocked pastures and seldom are found on . (2) Tobosagrass has been affected very little by the grazing practices for the 16-year period. t decreased slightly on the ridgetop soils but increased on other soils in pastures grazed with sheep. "i attle utilized tobosa more uniformly than sheep, but sheep caused damage to this grass by “spot azing.” Livestock utilization of tobosa noticeably increased by mowing to remove the old growth. (3) Buffalograss, although more palatable than curly mesquite, was more resistant to heavy i grazing. Both of these turf grasses produced less forage than tobosagrass during years of low rainfall. (4) Three-awn grasses were the most reliable indicators of the class of livestock grazed on the pastures. The three-awns increased under heavy sheep grazing and decreased under cattle grazing. Annual weeds, particularly annual broomweed, were less abundant on the sheep pastures than on the cattle pastures. if (5) Pricklypear increased under heavy yearlong and seasonal grazing. This cactus also increased i" i the exclosures during the drouth. Mesquite tree populations were more closely related to the type -: ,1’ soil than to the grazing practices used since 1938. con-rams Page Digest . . . . . . . . . . . _ . . . . . . . . . . . . . . . . . . . _ _ _ . _ . . . . . .. 3 Introduction .............................................................................................. ..L ..... .. 5 Experimental Area ................................. ._ i ................ -. 5 Location ..................................................................................................... .. 5 Climate. . .................................. -. 5 Soils _ _ _ _ _ . . . . . . . . . . _ . . . . . . . . . _ . . . . . . .. 6 Vegetation ................................................................................................. .- 6 Methods of Steudy ............................................................................................. .- 6 Location of Plots for the Study of Vegetation ................................ .. 6 Survey Procedure for the Inclined-point-contact Method ............. .. 6 The Belt-transect Method ........................... .. 7 Brush Survey by the Circular-plot Method 7 Methods of Measuring Forage Production and Utilization .......... .- 7 ‘Soil and Plant Relationships ........................................................................ .. 7 Plant Distribution as Related to Soils .............................................. .. 7 Effects of Vegetation and Soils on Water Infiltration .................. .. 8 Effects of Vegetation on Soil Aggregation 12 Effects of Vegetation on Soil Temperatures _____________________________________ __ 12 Effects of Rainfall on Vegetation .............................................................. .. 12 Relation of Rainfall to Available Forage .......................................... _. 12 Effects of Rainfall on Floristic Composition .................................. .. 14 Forage Production ................................................... ............................ .. 15 Effects of the Drouth on Vegetation ________________________________________________ __ 15 Response of Vegetation to Grazing ............................................................ .. 16 History of Grazing Use ........................................................................ .. 16 Effects of Stocking Rate and Season of Use .................................. .. 18 Effects of Class of Livestock on Vegetation .................................... ._ 19 Utilization Studies ................................................................................ 20 Acknowledgments ............................................................................................ __ 22 Literature Cited .............................................................................................. __ 22 ‘Relation of Soils, Rainfall and Grazing Management to Vegetation, Western Eeln/ara’; Plateau 0f T exa; GERALD W. THOMAS and VERNON A. YOUNG* INCE THE BEGINNING 0F RANCHING in North rainfall belt. Most of the precipitation occurs as l v erica, men have recognized that the kind and rainfall. Figure 2 shows the distribution of this E. mint 0f Vegetatign the ranch Supports deter- rainfall by years. There is considerable variabil- ines to a large degree the success of the enter- illy among Years, and 9 Ollt 0f the 16 have had A 'se. This vegetation, in turn, is influenced by less than average rainfall- }; various environmental conditions under which Average monthly rainfall also is shown in grows. dFaggors such 2111s climate and soils have Figure 2_ May has the highest average’ 218 911091199 e eel} on l3 e Vegetatlon- Grazlng inches and February the lowest, 0.57 inches. tensities and kind of livestock also are very i portant but too often they are overlooked or - ANNUAL PRECIPITATION - v‘. r5145 F4071’ J/a/m/L, &'77l‘4f/I‘7é\'44' l Grazing experiments were initiated on the exas Range Station in 1938 to supply information , g the influence of soils, rainfall and grazing on 4° *4” _, vegetation type common to a large area of West exas. Results obtained will be of value to _ Avwmeswuuvmclmmrw - 16.5w ‘nchmen in this area and in other regions where 3° l: 30 bosa, buffalo or curly mesquitegrass are im- [ 3 a I rtant forage plants. w '3‘ . ; . g e . “i h‘ a ~ a . a ., "l W22" ml°§ é - i" ---- --"'~--—*—-~3 - ------- -~'“.--~ " EXPERIMENTAL AREA T a = ~ 2 ., 3 é , H 0:‘ p‘. )0 t?! ,0‘. ‘D. y‘ l ation w P“ "o The Texas Range Station 1S located in Crockett unty, between Ozona and Barnhart. It com- o_ ’~‘ _ - ises approximately 3,160 acres of land owned 38 =9 +0 41 43 44 4s w 4e so 52 sa ° l -' the University of Texas. The area has been _ Y5“ T ‘anaged for experimental purposes since 1938 by " Avfkfifif 410N730’ PREUP/TAT/UIV '- l e. Texas Agricultural Experiment Station under ' Tm" “"7” "*°""”‘/ 5”"”‘¢’*»T"“‘ " cooperative agreement with the University of exas. pl . The station lies on a broad drainage divide of z he Edwards Plateau at an elevation of 2,700 feet AVERAGE -_ 1.59" tween the headwater streams of the Concho iver on the north and the Devil’s River on the uth. The relief of the land is rather smooth l5 * » ith occasional shallow depressions and dry lake- ' ----- r - g ds. A few areas have gentle to moderate slopes. g Figure 1 shows the pasture boundaries, acre- ‘ges and livestock water facilities on the Texas ' iiange Station. a 2.0 l1 .32" 1 .55" ____._._... ‘o W'O'O'C o o 0 ~ 0.0.0. 53 o V O‘ .3} '¢'..Q¢ v ,v.v‘v Q.Q:o.O' ‘.8,’ .0‘ 0.95 " ‘v35?’ o 4'1 Q f0! ‘ It!‘ '3 a Q ‘g 0 8 ~03 5M ‘»:< i‘ v .\‘:'p.. go's 4:329. V O'Q' .0 o l: v “ '23:? “v.79. 0:920?’ -- ave/res — "6 I ‘Q 123:2» 9 ate g_ The average annual precipitation, at the sta- P5 - *3 ‘on headquarters for the past 16 years was 16.68 ches, Figure 2. This amount is probably too w for a longtime average because of the influ- . once of the recent extended drouth. Rainfall ° ;cords from other weather stations in the region 1 dicate that the range station is near the 20-inch vvv .. 92332:‘: '3 o: v - 9‘O.'O;O u. O‘W 202v! 0Y0 y § 4Q QQ O‘; I Q ¢$ 'O\ ~2 ‘U’? y??? e p§tfg O’ O 9,0 O A. I s“‘.¢" A§’A“\'A 4 y .0 'Q v ., . .- ‘z. w: qqzoo; 0 ‘W9 3% "30 22¢!» I 0:0 O Q ‘"4 '0 '0 w?" w» :39 O '¢ L o o ' Yofo .0 1'0 o’ L .0. "'4 Jan F576 v Nmm um 7mm) JuKE _ -" MONTH —~ Figure 2. Annual precipitation and average monthly pre- espectively, assistant professor and head, Department cipit-ation for the Texas Range Station from f Range and Forestry. . 1938 through 1953. 0C F NOV DEK Summer temperatures at the range station are high, with maximums of 100 to 110° F. The nights usually are cool. Frequent southerly winds occur throughout the day and often continue into the night. In general, the winters are mild with occasional periods of below-freezing weather which may last for several days. The mean annual temperature is about 65° F. Soils Soils of the Texas Range Station are typical of the western part of the Edwards Plateau. These soils are clay types or phases of clay types, moderately deep, and underlain with lime- stone or caliche. A survey made in 1938 by Carter, Templin and Mowery showed that there are five soil series on the station: Ozona, Valera, - Tobosa, Irion and Randall. Although the general surface appearance of these soils is similar, there are differences in surface relief and moisture rela- tionships. In a region of light rainfall, this differ- ence in relief is associated with considerable variations in soil development and in native vege- tative growth. The upland or ridgetop soils on the station are the Ozona and Valera soil series. Since the slopes are gentle, little or no water accumulates on the surface of these soils. Depth to the caliche layer ranges from very shallow to about 2 feet. Surface layers often are gravelly or stony. The dry lakebed areas fall into the Irion and Randall soil series. Drainage from these areas is poor, occurring mainly through holes and crevices in the underlying limestone. The deeper lakes of the Randall series may contain water for ex- tended periods following heavy rains. The Tobosa series comprise moderately deep, clayey soils that occur between the high swells and the lakebed areas. The subsoils are heavier and less permeable than those of the Ozona and Valera series. Drainage is slow, but, since these soils are above the lakebed areas, water seldom stands on the surface very long after rains. Vegetation Many species of plants have been found on the Texas Range Station during the past 16-year period, but the bulk of the forage is produced by only a few of these species. The most abundant grasses are sod-forming species: buffalograss (Buchloe dactyloides), cur- ly mequite (Hilario. belangeri) and tobosagrass (Hilario. mutica). Species of lesser abundance are sideoats grama (Bouteloua curtipendula), three-awn grasses of the Purpureae group (Aris- tida spp.), vine mesquite (Panicum obtusum), hairy tridens (Tridens pilosus), fall witchgrass (Leptoloma cognatum), muhly (Muhlenbergia arenacea), tumblegrass (Schedonnardus painicu- latus) and hairy grama (Bouteloua hirsufa). A few scattered plants of cane bluestem (Androp- 6 ogon barbinodis) and green spangletop (Li, chloa dubia) grown in pastures C and N w, have been lightly stocked. f Several annual grasses were noted follo periods of high rainfall. The most importa a these were little barley (Hordeum pusillim) weeks fescue (Festuca octoflora) and an panic (Paniciim fasciculatum). 1 Approximately one-fourth to one-half of . ground area on the station is bare. These =. areas between the bunches of perennial grass q subject to invasion by many species of weeds =1] ing periods of favorable rainfall. The most im ant species are bitterweed (Actinea oclorata) annual broomweed (Gutierrezia texana). Forbs of lesser importance are croton. goatweed (Croton neomexcicanus), buffal. (Solanimi rostratum), caltrop (Kallstro brachystylis), snow-on-the-mountain (Euphx marginata), perennial broomweed (Gutierr microcephala), buckhorn plantain (Plcmtago ‘ ceolata) , hoarhound (M arrubium vulgare), A ican carrot (Daucus pusillus), rose verbena ( . bena canadensis) , prairie thistle (Erynigiume venworthii), Engelmann’s daisy (Engelmw pinnatifida) , Drummond phlox‘ (Phlox drum » dii) and many others. ‘ The dominant species of brush are mes,“ (Prosopis glandulosa) and pricklypear (Op 1' spin). Other species of cacti include c 8 (Ommtia imbricata) and species of Echino w» and M ammillaria. Several other brush speci’ minor importance also are present. METHODS OF STUDY» Location oi Plots for the Study oi Vegetation l, The original 86 plots for vegetational st,‘ on the Texas Range Station were establish 1938. An additional 210 plots were located l‘ at random on soil types which had not ' adequately sampled. ; = All permanent plots were marked byrtwo stakes, spaced approximately 25 feet apart," set in the ground to a depth of about 2 "feet. metal tag with the pasture identification and number was attached to the southstake at location, Figure 3. ' ' ’ a ~ Survey Procedure for the Inclined-point-contact Method i’ The inclined-point-contact method has ' used for measuring relative abundance of ve tion on the Texas Range Station since 1938. _ point-contact frame consisted of 10 pins. It; pin was dropped toward the ground, and a recorded for each species contacted between . lower crossbar and the ground. For exam: pin number 1 contacted species A more than “s only one hit was recorded. If pin number 1; tacted species A and species B two hits a recorded, one for each species. The point-co" e was moved 10 times along the line between yes at each plot location. Readings were made xis manner for 100 pins at each plot. Belt-transect Method uring the period 1950-53 belt transects were lped at each of the permanent plot locations. belts covered approximately the same area ed by the inclined-point-contact method. glfhe belt transect used in this study was 1 g wide and 20 feet long. The zero point on the 1 was maintained at the tagged stake. The Station within this belt was mapped by means standard square-foot quadrat and chain. fFrom these belt transects, it was possible to 'y the effect of grazing and drouth on indi- _ al plants and to follow trends through the s. Only perennial plants were located on the I: of the belt. Annual weeds were counted hin sod and bare areas along the belt. Stubble ht and foliage-density measurements were in- ted for the area of each sod grass mapped. itional notations were made of other factors idered important, such as amount of erosion. Survey by the Circular-plot Method ;-Woody plants were charted on circular plots . each sampling location to supplement the data A 'ved from the belt transects. These plots had ‘i-foot radius and were centered on the tagged ker stake of each permanent charting location. position of the belt transect within the cir- plot and the location of all brush species e noted with the proper orientation in relation true north. Basal diameter, height and canopy perage were recorded for all tree species. fThe positions of pricklypear, cholla and other ti were mapped within the circular plot. Ant s, livestock trails and other factors which may 5» influenced the vegetation also were indicated. Pthods of Measuring Forage Production Temporary exclosures were established during summer of 1951" in pastures O and P for the lsurement of forage production and utilization. elve exclosures were placed in each pasture, 10 ’ Ozona clay soils and 2 in the lakebed areas. f of the exclosures on Ozona clay were located a tobosa sod and half on mixed buffalo and ly mesquite sod. All of the exclosures in the i. beds were located on buffalograss sod. Y, Clippings of forage in the exclosures and in grazed areas were based on standard 9.6 @ lare-foot plots. Plots of this size allowed for lid conversion to pounds per acre by multiply- f: the number of grams per plot by 10. Clip- g: were weighed in an air-dry condition. Sub- 1| ples were ovendried for determinations of isture content. Sod areas of buffalo and curly mesquite were ‘lped to a stubble height of approximately one- a inch for production and utilization measure- ments. This stubble height was very close to that actually remaining after intense sheep grazing. Tobosagrass was clipped to Within 1 inch of the ground. Exclosures were moved twice each year--in August and December. Actual forage production during each period was determined from the clip- pings made inside the exclosures. The amount of forage remaining after livestock grazing was found by clipping the forage from plots outside of the exclosure. Forage utilization by livestock was determined by comparisons of clippings from the exclosures With those of the grazed areas. The amount of utililzation of forage by both sheep and cattle also was obtained by stubble- height measurements of species within the belt transects in each pasture. SOIL AND PLANT RELATIONSHIPS Plant Distribution as Related to Soils Differences in the amount and composition of the vegetation on the Texas Range Station Were closely associated with differences in soil characteristics, particularly surface texture, soil depth and the nature of the soil profile. The greatest differences in the vegetation were due y‘ 13+ w. . _ . ,.>-.-- 2. '» . ‘M621: - '- $4 - Figure 3. A typical plot location in pasture D. Note the tagged stake in the foreground and the gen- eral vegetation condition (summer 1950). 7 Figure 4. Soil boundary in pasture C between Ozona clay (left) and Tobosa clay (right). Note the abundance of mesquite trees and tobosagrass on the Tobosa clay area. to differences among the soil series. Soil types and phases within these soil types also affected plant distribution and abundance but not as much as the soil series. Differences in plant occurrence among soils often were as pronounced as those shown 1n Figure 4. The relationships among soil series and the occurrence of the major plant species are shown in Figures 5 and 6. These graphs present sum- maries of data collected in 1950, 1952 and 1953. The greatest density of tobosagrass was found on the Tobosa and Valera series. Very little tobosagrass grew in the dry lakebeds of the Randall series. Highly significant differences in the amount of tobosagrass were found between the Tobosa clay and Ozona clay types. Both the shallow phase of Ozona clay and the shallow phase of Irion clay showed small amounts of tobosa- grass. Buffalograss occurred in the largest amounts in the dry lakebed areas of the Randall soil series Where it sometimes formed pure stands. The second highest concentration was found on the poorly drained Irion soils. Fairly large stands of this species also were encountered on the ridgetop soils of the Valera series. Highly signi- ficant differences in the amount of buffalograss on different soils were noted by Potts (1946) and Thomas (1951). Curly mesquitegrass grew in close association with buffalograss on all soils except those of the Randall series. Curly mesquite seldom occurred in the large lakebeds or on the Irion soils. Sideoats grama occurred in small amounts on all soils except the depressional areas of the Irion and Randall series. This species was most abun- dant on Ozona clay. All soil types on the range station were found to support three-awn grasses. The dry lakebeds 8 showed the highest densities of purple three-q and the uplands supported more Wright’s t awn. The relationships between soil series and . amount and kind of brush species are show Figure 6. The highest concentration of brush recorded on the Randall and Tobosa soils. _ quite trees were more abundant on Tobosa than on the other soils. Pricklypear plants "A most common in the lakebed areas. The upl areas of the Ozona and Valera series were no p heavily populated with brush species as the 0t soil series. - l‘ The lakebed areas of the Randall and I soil series had the lowest total densities vegetation of all soil series represented. large amount of bare ground on these soils 3 subject to invasion by bitterweed and other an weeds during years of favorable rainfall. Oz clay had the smallest amount of bare gro before the drouth but Tobosa clay had the l’ bare ground in 1953. Extensive areas of t; ground also were noted on the shallow of Ozona clay and Irion clay. These areas been classified as “hazard areas” during i bitterweed season. i . Effects of Vegetation and Soils on é Water Infiltration Infiltration tests were made during the =0 mers of 1950, 1951 and 1952 to determine i» relative effects of the major grass species. rate of water intake into the soil. Infiltra was measured by means of concentric cylin or rings, as described by Leithead (1950). was added periodically to the inner and o cylinders to maintain a depth of 2 inches. j accurate record was kept of the amount a) to the inner ring. The outer, or buffer ring, stricted the lateral flow from the internal a partment. Each test was continued for a 2- f period. Infiltration measurements were taken on v sagrass sod, on buffalo and curly mesquite sod and on bare ground. The three series’? measurements were made within the radius l. 10-foot circle to keep soil variability to a i mum. A typical location for an infiltration is shown in Figure 7. 6- Y Infiltration tests were made in 1950 on different soil types in pasture C. These a i were repeated three times for each soil typ widely separated, randomized locations. Du. the summer of 1951 and of 1952, the tests confined to Tobosa clay soils. Five tests made on this soil type in pasture C. Infiltra tests also were made in 1951 on sideoats 1 sod and on buffalo and curly mesquite sod inj Ozona clay exclosure in pasture D. a Data on infiltration of water are summa g in Table 1. Curves showing accumulated 9;, of water absorbed by the soil are given inFi \ . <2. i§ ./~:~ a ‘i. . .1 I ____- AVERAGE DISTANCE IN FEET -_- —- LEGEND — OTHER SPECIES a 5'3‘ CURLY NIESQUITE GRASS 5M KYO. V sun-mm 6 CURLY MESQLHTE A BUFFALO eRAss SOD q TOBOSA GRA$5 SOD To605A GRAss (Avsusnso oeusnv) 20 — w — — U —- r- -— r— - 19 1a n 1e m w" l~jIf ' / ? 1o % / % 9 4 ? g 6 ¢ ¢ Z / / fi 7 ¢ z ¢ % / ¢ / /» % 6 % ¢ ¢ 5 é z o. é % 4 w. '2‘ Z % O / 3 . g g a~ ¢ Z Z w: é % o. I900 3 I953 7/950 r953 /6 I950 fi VALERA ozoNA ToaosA IRION RANDALL. sou. seems Figure 5. Effects of soil differences 0n the vegetation of the Texas Range Station. represent average measurements from the 1950 and 1953 belt transects. Data B - N fi-j n11 "- C KO O0 AVERAGE NUMBER PER PLOT‘ i ll 6 || f...» I '- .::.' ‘.-_'- 2'- ".:'; 5 5 Y 5 : :1 -2Z- . ":5 ':'h LN‘ \---- '-:I- '::;' 7;I- 1Y1 ,::'. 4 'n'o ' 0 - ‘n. u" w”. ~¢"‘ 10.0 1-14 ' l u... "r- '.. ' uQvn 11.: . . 1 "f," "i. ' o‘ n0'u0 3 ===¢ :.':- . 1' ‘ ,l n,‘ ,-~ 1:0 v." 7,’: N I930 (951 I9 5O I952. OZONA roeosn —-—-son. ssmes -—-- 1990 1992 VA LERH g l .--.....-....-..... .___ .___... 2.'.:-~.-. ~_-..._.. ~. . __ __ QF.-_§_§IQD§§- o‘... n!‘ u .-;~. .. >4‘ 19‘? Pr?‘ PP.“ 2"?!" QOO MOM Mum hCDQI QQQ 01mm e-I an 0-4 on e-n e-I 5) “P? 995° 9'8".“ Pr“? PM" 9F!“ Nubw N007 (D09 I-IQQI QQQ (On-IN ment of these data by analyses of variance indi- cated a significant effect of the grass cover on both initial intake rates and total amount of water intake. These analyses also show a high avari- ability in water intake within soils and no signi- ficant differences between any of the soil series. Most of the data were consistent in showing high rates of absorption at the beginning of the tests, then a rapidly diminishing rate until a fairly constant ultimate rate was obtained. These trends are shown graphically in Figure 8. No effective rainfall occurred in 1951 for 2 to 3 months before the infiltration tests were made. A general drouth was becoming increas- ingly severe in 1952 and there was no available moisture for plant growth at the time of the tests. Bare ground had supported scant annual weed or grass growth since the spring of 1950 and prac- tically no plant litter remained on the ground. Data collected during these last 2 years did not differ markedly from those obtained in 1950. However, in 1952 the bare ground had the highest final intake rate as well as the highest total amount of intake. a With the concentric-ring method, the destruc- tive and puddling effect of raindrop action was not considered. This is an important factor under actual rainfall conditions, as pointed out by Osborn (1950). However, at all times the water was muddy on bare ground and practically. clear on both sod conditions. The relative positions of the curves for tobosa sod and buffalo and curly mesquite sod remained unchanged during all the tests. Analyses of variance indicated that differ- ences from year to year were not significant, OZ 0N0 CL/l Y ‘a 7 ‘*1 <. i w s S V4 P l‘ Q on é 3 g z S1 0Q lad Li!‘ 2:0 i?‘ 3; TIME (nun/arm?) Q, T0608!) 0L!!!’ 3&1 s ' I jod Ki a 5 a, grass v‘ \ 59$ r; ;e 0e 0e I \.¢4 1'0 693,51‘, . w 3 , e0 e o S»? ',..__“:_fl£'/Z¢!9; {£222. fe-s- - Q/ r»—:v'*'*' "i o0 ' 1:0 Z6‘ 2:0 2:6‘ 3ft! TIME div/Mares) Figure 8. Typical curves showing accumulative amount of water absorbed by the soil under three dif- ferent ground-cover conditions. ll although they were close to the 5 percent level. The slowest rates under all conditions occurred ‘when the drouth was most severe. The rapid rate of water intake under both sod types tested in the exclosures indicated that grazing is a factor in affecting infiltration. Sod- forming grasses tended to develop a bunch-like appearance under protection from grazing and there was a heavy accumulation of plant litter. From the measurements of soil height it was found that the soil level under a cover of tobosa- grass was one-half to 2 inches higher than the soil‘ level on areas occupied by buffalograss, curly mesquitegrass or annual weeds. The accumulation of litter and increased soil porosity under tobosa sod undoubtedly contributed to the increased rate of water infiltration as compared with the infil- tration rate of the turf grasses. Also, the extent of soil aggregation and the stability of these soil aggregates have influenced the results obtained. Effects of Vegetation on Soil Aggregation Laboratory studies on the nature and distribu- tion of soil aggregates were made on soil samples of Ozona clay collected on the Texas Range Station in 1950, 1951 and 1952. Determinations were made on samples of the upper 6 inches of soil taken at random in pastures C. and R. Three ground-cover conditions were represented: tobo- sagrass sod, mixed buffalo and curly mesquite sod and bare ground. Comparisons of aggregate size distribution were made for the three ground-cover conditions by wet-sieve analysis, using a modification of the Yoder technique. Results of these analyses are shown in Figure 9. These curves represent “summation percent- ages” plotted according to the procedure outlined by Kroth and Page (1946). Two different dates of sampling are shown. Samples collected during IOO- ~ - PERCENT OF’ SAMPLE — ..~.-:-'~'-"‘ AGGREGATE suzc oasmneunon sv smzucs “i- (WET-SIEVE ANALY5|5 FOR TWO OIFFERENT auras or samPl-ING) | 1 1 n g J,——— 0.1 0.’: 0'4 0.; I-‘o w‘ 1° Pnnrncrs 5m: m Mmls Figure 9. Curves showing the effects of vegetation on aggregation of soil samples collected in the winter and summer of 1950. 12 ' centage of large aggregates occurred in soil f ‘A the winter had a higher percentage of large ag gates in each type of ground cover. The distr tion pattern of aggregates was similar in . samples from the three ground-cover condit at both dates of sampling. The maximum = bare ground but total aggregation was grea under tobosagrass sod. “ Wet-sieve aggregation tests also were m] on large clods of soil. These clods were slak 6 water for 1 hour and were wet-sieved for 1 p instead of the standard 30 minutes. At the of this time, the clods were completely bro down to particles less than 5 millimeters in i meter. The aggregate distribution patterns 1p were very similar to those shown in Figur‘, These results suggest a characteristic distribul pattern of aggregates for the soil type and i cate that the aggregates are very stable. ’ Effects of Vegetation on Soil Temperatures The amount and kind of vegetation havf direct influence on soil temperature. In 1 studies of soil temperatures conducted on 5 station, striking differences in temperatures L observed under tobosagrass sod, buffalograss and bare ground. Typical curves showing " temperatures at 3 and 6 inches under sod and o ground are plotted in Figure 10. * Soil temperatures under a tobosagrass co were lower in the day and higher at night “l temperatures measured at similar depths if bare ground. Thus, tobosagrass reduced . maximum temperatures and increased the n? mum temperatures, causing a more uniform l temperature. ’ At increased soil depth, temperatures la further behind the air temperatures. Figure shows that soil temperature at a 3-inch under bare ground on June 29, 1952 reach maximum 2 hours later than the outsidej temperature (measured in the shade). A depth of 6 inches, the lag behind air tempera , was 6 hours. ‘ Observations of soil temperature were p, in 1950 under favorable moisture conditions, ‘ also in the following 3 years of drouth. taf observations showed that the effects of vegeta on soil temperature were reduced asthe y became dry. The lag of soil temperature u air temperature also was less pronounced t; drouth conditions. = EFFECTS OF RAINFALL ON VEGETATIONA; Relation of Rainfall to Available Forage The general relationships between the de _ of forage on the Texas Range Station and amount of rainfall for the 16-year experim period are shown in Figure 11. Total rai“ for the year was plotted against the numbev hits as measured by the point-contact metho The point-contact measurements were ly made about the middle of April. In most », the vegetation had “greened up” by this f and plant identifications were not too diffi- 1 This period also was considered the most able time to record the abundance of bitter- '1 although growth of this plant often occurs a: late fall. E he number of hits recorded by the inclined- Kdered as a measure of the amount of avail- forage on the experimental pastures. Since 5 chartings were made in pastures being fed by livestock, they reflect plant growth, ity of ground cover and amount of forage éized by the grazing animals. "Several general observations are evident from ' relationships shown in Figure 11. Available Vge of all plant species studied is closely related athe amount of rainfall. Most species show a j in available forage of from 1 to 2 years behind 2 rs of high or low rainfall. This effect was in previous studies by Nelson (1934), 'dd;)ck and Forsling (1938) and Clawson 50 . ‘The graph giving total density of all plant ies, both annual and perennial, shows a signi- nt correlation between available forage on i» clay and the rainfall of the preceding year. 's relationship is indicated by the large corre- ion coefficient, r'-=.637. This means that the lximum response of vegetation to rainfall under _stant stocking rates is not immediate but re- "res a delay of about 1 to 2 years. The more .1 iable vegetation on Tobosa clay shows a aggg but significant correlation coefficient, A close relationship was obtained also between rainfall of the preceding year and the density F buffalograss. For this species, the value of ‘ f .635 on Ozona clay was very close to signifi- Vince on the 1 percent level. This correlation fficient indicates that about 36 percent of the r-to-year variability in available forage is ociated with rainfall during the previous year. i The nature of the soil has influenced the . f» ount of variability in the density of vegetation. or example, the lakebed areas of the Randall _ 1' Irion soils series had greater fluctuations in ; amount of vegetation than the other soils. * ost species of plants showed a more stable nsity on the ridgetop soils of the Ozona series. Tobosagrass showed the least amount of year- year fluctuation in density with variation in ainfall. This grass was the most depandable of 1 species for livestock forage during years of ' s rainfall. However, when rainfall conditions ere favorable it did not increase in abundance ; rapidly as buffalograss or curly mesquite grass. i highly significant correlation coefficient, , .651, was calculated for the density of tobosa- , ss on Tobosa clay soil and rainfall of the contact method, as used in this study, is" previous year. A value of r=.610 also was cal- culated for the relationship of rainfall and tobosa density 2 years later. A greater lag in growth following rainfall was noted for the three-awn grasses than for the other species. This lag, which amounted to 2 years in many instances, may have been due partially to the low palatability of these grasses resulting in a carryover of forage from year to year. Variation in growth due to climatic factors was greatest in the annual weeds. This fluctua- tion is demonstrated by the 16-year record of bitterweed density presented in Figure 11. Bitter- weed growth is more closely related to seasonal distribution of moisture than to the total annual rainfall. This poisonous range plant grew pro- fusely in bare areas when there was sufficient late fall and early spring moisture. In contrast, annual broomweed germinated late in the spring and was not abundant during years with scant late-spring rains. Figure 11 indicates that vegetation response following years of low rainfall is slower than ~82 78" JULY 4 and 5', 1951 mo NW6 _ SOIL TEMPERATURE AT é" l 16 ,00_74__ 7H0: ea-rz» """°° 96-70- ""96 9% 409s 92. 94 9o 91 85 g 9o 8e F‘ a 55 a4 v B6 52 M a 62 80- JUNE 29 and 30, 195g Va 6o 7 7: son. TEMPERATURE AT é" 7e 76 |0L—74 |00—7Z 96"70 ~90 94 - 9L -90 ~§ -eu “é ~62 - JUNE 29 and 3o, 1952 4o fi 78 — m" son. TEMPERATURE AT 1" ~76 7+- "6 _ i l W 74 ~ u v , , , : l ; : l : , : ‘ I I : I . I 0o Z 4 6 B 1o ménunogr 2 4 6 6 m v Mo“ _._-- AIR TEMPERATURE -------- TOBOSA S01) -<>—<>-<>—<>—- BARE GRO ' .... .. . "ND ~- BUFFALO AND CURLY MEsqUm; Figure 10. Characteristic effects of vegetation type on soil temperatures at 3 and 6 inches depth. l3 the response to years of high rainfall. This points up the danger of restocking immediately after periods of drouth. Effects of Rainfall on Floristic Composition Fluctuations in the percentage composition of tobosagrass and the turf grasses, buffalo and curly mesquite during 1938-53 are shown in Fig- ure 12. Percentage composition was calculated by dividing the average number of hits of each species by the total number of hits for all plant species. "‘ - ALL PLANT SPECIES- m . ozomn cuw l’ n4 ‘s: ‘i. \ 78-8-43: ‘g f: ' as I no _- ,a\ “.1. ‘f. A\ 5'; w ' : 9 E :/ w C g 90 n a °° " a 10 ll g w 1a 5 . a >0 2s j ‘Iv ll 30 s 1D s v I I 40 4| 4t 4! 44 4! 4w <1 40 n so s: s: u 0 -— EQR —.- '- "SIDE*'OATS G RRMA '- n ' I 1'3 i 9 g 6 a 7 fi ozown cuw Si.‘ " /'\ / \ g 5 / \\ [I \ VALERA cuw I 4 \ I r 3 I \ I / '. \ / \ z I \\ I . \v/."‘\._.\\ ’ _\ I /J‘-~€’)~ I \°\ \\ ¢—’-?§~. o 9'0 n 40 4| 41 41 44 4s 4e 41 w: "19 m1 a: ‘x5: s! '— YEAR — -—THREE ' AWN GRRSSES - w a ,, RANDALL “no I \i \ / \ i" ./'\. I \\ p f \ z \. / \'~’ . \ ‘s’ \ . \ l . 3'“ \/ _ \ 5 ozovm cLav \y/~ \\~/,q\ ,1 \ \- - f'-§@” \\ i"’—‘\._ ' n 9 T; 4n u. ls 94 4s 1s <1 1e 4o n n s1 s; '- YERR "- Figure 11. Relation of vegetation density to rainfall during 1938-53. Vegetation density was measured by poinv f tact chartings made in March or April of each year. 14 — Pnecmrrawwou (sucnss) -— These data show that tobosagrass made v, bulk of the forage during years of low rai 1 Buffalograss and curly mesquitegrass, how show a higher relative coverage than tobosa ‘l in most years. a The graphs showing percentage compositio‘ the above grasses for the 11-year period, 193 1948, might readily lead to an erroneous con sion. The apparent trend toward a predomina of tobosagrass is largely a reflection of rai i Tobosagrass in 1948 actually had the lo density observed. The fact that the density; . \ __ I '\ - BUFFALO amass — w ./\. . ' / I \ RANDALL ctnv so \ l - . f‘- 7 I RAINFALL . - Q E40 U n‘ ozoNA cuw = I! \ / \ k . . g ° 3° z i“ g \ /,/. \\ l / ‘\ K ~ ' \ . é Z0 A . / \ "‘ \ 1 x‘, / / \ a l \\ / . / - \ // l 7 0 )5 s9 4o 4: a: 4a 44 Yés-AR“ 4r 4s 49 so 91 s: - TOBOSA enass — i 4o . a‘. K E'~..*'... "1', I‘: “an, 1|? TOBOSA cuw x '. - a” "R ozonm cLAv ‘g q /_\ /‘-\ ‘ g 1’ "‘ ' \€/ f‘ // \\ ] 2 w \ x, I \s; l \— \s l, a 1e 99 4o <1 n 4s u Y» 4o 42 4a 442 so s‘: r1 '? — EAR - — BITTERWEED — N, . 8 "F?" l. mou era‘ l ‘i, l I\ L‘ Z ‘a 90 5 :- 0: , . :' , / i} _ , _. § TOBOSH cuw __- _.- -_ ; '__- a 2o ' .' '-_ E I ,' z ‘I. '- : ,' n ..V. | ' 5;‘ a ozowr S xi 3 ' \\‘l'; x’ \\_ _. -" 4’ \-a/ o‘ u _ n 4| n a! 44 n n <1 u n 1e n ll -- vsm -— short grasses was reduced even more than of tobosa accounts for the apparent shift 1n inance. omposition trends shown in Figure 12 also i»: out some of the hazards involved in the qrpretation of grazing effects from the vege- , nal survey of a single year. Studies of plant position during dry years may suggest differ- grazing effects than similar studies during ' - of adequate rainfall. Illa’. ge Production ‘iThe forage produced in the temporary exclo- §>= was determined from clippings made in h": st and December of 1951, 1952 and 1953. se yield data are shown graphically in Figure ' Forage production was extremely low from :1 to 1953 because of the shortage of rainfall. »- highest production during this period was ined during the spring and summer-growing a sons of 1952. It amounted to only 738 pounds air-dry forage per acre of tobosagrass and 449 nds of mixed buffalograss and curly mesquite- éss. Buffalograss produced 369 pounds of ge per acre in the lakebeds during the same 'od. All clippings were made in dense sod which should produce much more forage der normal rainfall conditions. 1v" uvvuncb] -—- § On the Ozona clay soils, tobosagrass consist- tly produced more forage per unit of sod area I: the turf grasses, buffalo and curly mesquite. Yffalograss, on the lakebed soils of the Randall f 'es, produced more forage in 1953 than the w er grasses. However, only a limited area was upied by this grass at this time. a Forage production was limited during the _ l-growing seasons for all species of grass. om August to December 1952, the production f: insufficient to measure. Normally, forage wth in the fall cannot be depended on to rnish feed for the winter. It is, therefore, ential for the livestock operator to save some rplus forage from spring and summer growth that the livestock may survive the winter with lminimum of supplemental feed. This practice l‘ Will help to maintain the vigor of the grasses. ; These data on forage production reflect the tensity of the drouth during the period of this vestigation. They also indicate the importance , adjusting stocking rates to conform to low tinfall conditions. * ects of the Drouth on Vegetation As shown by the rainfall records, Figure 2, a ‘Vere drouth has prevailed at the Texas Range tation since September 1950. The rainfall for 51-53 was considerably below the average for i: area. A large proportion of the recorded ginfall occurred in light showers that were inef- j tive for plant growth. This shortage of mois- 1 re affected the forage production and caused a rked reduction in plant density. It has been difficult to separate the effects of the prolonged drouth from the effects of heavy grazing. Many of the changes in the vegetation that have taken place can be attributed to a combination of both drouth and heavy grazing. However, the changes in the exclosures, as shown in Figure 14, were caused by the drouth. Tobosagrass was the most drouth resistant of the forage grasses. Only slight changes in total sod area were recorded for this species. Some reduction in foliage density occurred due to death loss within the sod areas, but this reduc- tion was smaller than that of the other grass species. High death losses occurred in both buffalo and curly mesquitegrass. Figure 14 shows the reduc- tions in sod area and in foliage density that occurred in one of the exclosures on the station m - PERCENTAGE COMPOSITION - (POINT CONTACT memos) TOBOSA GRRSS 8 8 8 _- Psflcsu-r COMPOSITION -—~ B BUFFALO AND cum-v mesoun-s S I6 l9 40 4| 42 43 14 45 46 4! OJ 19 5'0 5| - YEAR —~ ’° - oeusmr- (worm conncr Mzrnoo) . . .1 2 . . BUFFALO AND CURL‘! ME5QLHT£ ’° : Z . ,-._ ' '- : . .' . '. PUTS '8 1S \ TOBOSH GRfl 5S -——- NUMIIR OF S 8 0 Jl 3n 40 4| 41 A; - 44 15 46 47 45 49 S0 3'1 S2 fi) —- Venn- Figure 12. Fluctuations in floristic composition and aver- age density of tobosa and the turf grasses, buffalo and curly mesquite. Percentage com- position can be misleading without taking density into account. Composition changes caused by the rainfall pattern should be re- cognized and separated from changes due to grazing. 15 from 1950 to 1953. Curly mesquite appeared to have been affected by the drouth conditions more than buffalograss. Density measurements 0f side-oats grama, obtained in the shortgrass exclosure in pasture D, showed that some death loss occurred during the same period. Marked reductions in vine mesquite grass were observed in the lakebed exclosure in pasture D. Severe losses were noted for purple three-awn in the exclosures and in the grazed pastures. The drouth apparently facilitated an increase in the density of brush species. Pricklypear seemed to die back temporarily but recovered rapidly following rains. New plants of this species often originated from the breaking up of the old plants by drouth. Plants of cholla cactus became severely desiccated during dry periods but sur- vived the drouth very well. Foliage density of mesquite trees was reduced in 1952 and 1953. Some branches died but very few trees were completely killed by the drouth. The data from the exclosures show a significant increase in the number of plants of both pricklypear and mes- quite during 1950-52. Effects of the drouth on the vegetation varied somewhat in the soil types examined. The fine . I textured soils in the low areas dried out i became very loose. Sod grasses on these died back more than those on the ridgetop Because of the reduction in the vegetation c0 " wind erosion was evident on all soils. - ' RESPONSE or VEGETATION TO GRAZING; History of Grazing Use Detailed records have been maintained si 1938 of the kinds and numbers of livestock g ed on the various pastures of the Texas Ra, Station. These records were used to analyze _ effects of grazing on the vegetation. The area now occupied by the Range Stat’: was privately operated before 1938. Little l formation is available on stocking rates and p " tices used during that period. Available evide q indicates, however, that the area was hea grazed by both sheep and cattle. Since 1938, the pattern of stocking has v ied. All pastures have been subjected to s0' grazing by both sheep and cattle. Certain p, tures have been used in both rotation grazing i continuous grazing management. Stocking ra for all classes of livestock have been changed s --FORAGE PRODUCTION —--_ 70o - TOBOSA GRASS (DZONAQAY) BUFFALO AND CURLY MESQUITE (OZONA CLAY) 6w _ i _ BUFFALO GRASS (RANDALL CLAY) 3 s s 3 is .'¢',‘~"§ =a= 3+ i‘; s F; g 400 l- \Q p‘ m II: H '6.o\ r=. $3 as o ' m - ,0,‘ a N ,4.“ ‘It's .. K W.‘ W" ;» "J "oi 1:: - E m ' 92*‘ t?» I'- O p4.“ ' . I v.3... W» $ ca». a M‘ w. '9'.‘ w“ '0'! w," W“ ’§“‘ V9.6‘ '9‘. l00~ N) Nd "'9, - ‘f; at‘ 5:“ 9V0‘ m M‘ m m m W? 92¢‘ $0; 0 o o via é é é :':~:< {'93 a» a a a -'o‘¢‘< m N‘ " Mg ._-'__: 33g ‘H 0 i" -".- i ‘AUG TO ‘DEC. DEC TO AUG. AUG TO DEC. DEC TO AUG AUG TO DEC. f 1951 1952 1952 1953 1953 ' Figure 13. Forage production during the drouth of 1951-53. The bulk of the forage comes from spring and early sud‘ mer growth. Forage produced in the fall usually is not adequate to carry livestock through the winter. 16 ><“>§">fl>< X>>§gé /“1T:§roao. AML $rh<$5¢ dmnaé (+4) Skulirfl Quukglfi '1" _ -@wuz guru/M; 30m? dnmzllié (an 5mm RAQN a." /KE>\u1~%&/Q\1li<82\\" '3" “@ //l5wm ggmwvwl "35-- /Bo!m qwwék ‘/ Tehran; Ami |9___ “(BQMQ n0t9\_ /'»},E[¢%,q (P) __ /’-?&§u31 411M115 (.1) 5mm kufigx *5" Q 9mm MW s" I952 U953 v’ Figure 14. Vegetational change on a typical plot in the livestock exclosure (pasture N) showing the reduction in 17 Figure 15. Livestock exclosure in pasture D protected from grazing since 1940. Note the abun- dance of sideoats grama coming into this ex- closure. eral times on each pasture during the period. The average stocking rate for the 16-year period var- ied from 14 to 40 acres per animal unit per year. This variability in stocking rates, livestock classes and seasons of grazing makes the problem of accounting for pasture differences very diffi- cult. All of these factors have contributed to the fluctuations in vegetation. However, forage spe- cies differ in their responses to them. It has been assumed in this study that vege- tational differences between pastures were large- ly a result of differential grazing use. This as- sumption is based on the fact that the pastures were fenced in the present pattern in 1938. It is further assumed that, except for soil charac- teristics, the environmental complex was similar throughout the area. Rainfall and temperature measurements taken at the ranch headquarters are believed representative of the entire area. The response of the vegetation to grazing prac- tices was determined in the following manner: ( 1) Differences in the vegetation caused by soil factors were recognized and removed statis- tically by analyses of variance. Thomas (1950) showed that there were significant differences between individual pastures for most of the ma- jor plant species on the station after variability due to soils was removed. These differences were confirmed by a more complete vegetational sur- vey in 1953. (2) The amount of vegetation on each soil in each pasture was then compared with the past grazing history of the pasture. This was done by computing correlation coefficients relating graz- ing factors with the density of each of the plants or combinations of plants studied. (3) These correlation coefficients were used as a basis for formulating hypotheses concerning the effects of grazing. The hypotheses were then checked by examining the past records and fol- 18 lowing trends in vegetation under the vario grazing patterns. However, some of the grazi '_g practices have not been continued for a sufficie length of time to establish final conclusions. i (4) Observations of vegetational compositio i in the exclosures also were useful in interpretin the response of the vegetation to grazing pra tices. * ‘ Effects of Stocking Rate and Season of Use The general relationships between yearlong ‘s; seasonal stocking rates and the amount of -Q tation on the experimental pastures are shown b1 the correlation coefficients in Table 2. These in-f dicate that most grass species were more abund; ant on the lightly stocked pastures than on th heavily stocked pastures. Also heavy spring gra Y ing appears to be more detrimental to most gras- ses than heavy fall grazing. Sideoats grama was rarely encountered in thi pastures stocked heavily with combinations of sheep and cattle. This grass was abundant in exclosures protected from grazing, Figure 15. I also was commonly found along the roadways ad joining the Range Station. The negative corr lation coefficient, r: —~.464, for the relation o density to yearlong grazing furnishes addition, evidence that this species decreased under hea ‘ grazing pressure. This evidence is supported b Dyksterhuis (1949), who states that sideoa grama is a climax “decreaser” in the vicinity o San Angelo, Texas. “Decreaser” species plants whose density is reduced by continuou heavy grazing over a period of several years. Both buffalo and curly mesquitegrass we reduced in coverage by heavy grazing, but the rp lationships between these species and stockin rates are not very pronounced, Table 2. Buffal‘ grass appears to be more resistant to heavy gra ing than curly mesquitegrass. This may be duo to the fact that the crowns of curly mesquitegra, Table 2. Correlation coefficients showing the relationshi‘ between rate of stocking by livestock and tll amount of vegetation on the Texas Range Statio Plant species s*°*=l_1(1__a_1 '2 v_ soed alueaa o an cur y mesqui e _'188 ‘M8 Adjusted density —-303 349 Number of hits __ -153_ -045_ 1§l$i§'lh$§§'$n(l§§§§ll¥v Iiillf "figs lnnual broornweed (numberV .281 —.5281 ‘ uite trees under 4 feet (number) .112 .139 sfiypgqy (nugnrlie!) Indicates significant relationship on the 5 percent level. rom 1950 survey. _ _ Zlndicates significant relationship on the 1 percent level. >0 + Response or m: Tans: nwu Gnnssss TO fiaazme ,,. g —— PASTURE P - sweep outv / 2”" 000000 PRSTURE o - CATTLE omur E E"- i if“... A /\ -- a‘ ""‘-.~ fa’ fWa’ I’ 193! X939 who 1 1 ‘_T,T_' l 3 ,. Rssvouse or m: Tunes AWN Gaassss TO — GRAZING BY SHEEP ONLY i GRAZING BY CATTLE ONLY 0000000 /cmmse T0 earn; (Panama P) >- ul | cunns: ~ro sneer (PISTURE o) AVHMIIUIIIIOIIITSOITIIIAIIIIPIQ m S | r 19149 1950 " 1951 1952 535.5115; 01G r0011 Till OI VIGBATIOI SUBVIY Figure 16. Response of the three-awn grasses to grazing by sheep and cattle. Note the rapid increase in these grasses when the pastures are graz- ed with sheep. even though rainfall conditions were favorable. Pasture A, which was grazed each year during the spring-growing period, showed a serious re- duction in most of the better grasses. Similar ro- tation grazing systems were tested on pastures I, K, L and M. In each case, the pastures that were heavily stocked during the spring-growing season deteriorated faster than those stocked dur- ing the other seasons. This evidence emphasizes the importance of a systematic deferment in the rotation design so that one pasture will not be grazed during the critical spring period every year. Effects of Class of Livestock on Vegetation Some differences in vegetation associated with grazing by sheep and cattle are indicated by the correlation coefficients in Table 3. The relative abundance of three-awn grasses is one of the best indicators of the class of live- stock grazing on the pastures. Pastures stocked at the highest rates with sheep had the largest amounts of these grasses. The three-awn gras- 19 Figure 17. Lakebed area in pasture O heavily stocked with cattle only. Very little three-awn grass three-awn grasses than the cattle pastures but fewer weeds. This picture, taken in 1952, shows close utilization 0f buffalograss around the clumps of purple three-awn. ses, however, tend to decrease in abundance when the pastures are grazed by cattle. A significant correlation coefficient, r2676, was obtained for sheep stocking and three-awn grass density. This relationship was discussed in detail by Thomas and Young (1953). Figure 16 shows the trends in density of three-awn grasses under sheep and cattle grazing. Figures 17 and 18, taken in the lakebed areas of pastures O and P, also show this differential effect of livestock class. The amount of annual weeds, particularly of annual broomweed, also was a good indicator of the kind of livestock grazing on the pastures. In years of favorable rainfall, pastures stocked with cattle supported large amounts of annual broom- weed. This species, nevertheless, was largely eliminated by heavy spring grazing with sheep. The negative correlation coefficient, r: —.528, shows a significant relationship between stocking rate by sheep and the number of annual broom- Figure 18. Lakebed area in pasture P heavily stocked with cattle only. Very little three-awn grass is present. This picture was taken in 1952 at the same time of the year as Figure 17. 20 Figure‘ 19. Fence line separating pasture B on theg from pasture C on the right. Differenceé due largely to the presence of annual b weed in pasture C, grazed by cattle, and ple three--awn in pasture B, grazed by sheep. L?’ weed plants per pasture. A typical fencl contrast between pastures grazed by sheep pastures grazed by cattle is shown in Figur_ Heavy stocking by cattle evidently ca, slight reductions in the amount of tobosagrj the experimental pastures. This effect is cated by the negative correlation coefficie Table 3 and by the point-contact chartings j in Figure 20. Sheep grazing caused a sligi, crease in tobosagrass. Since the point-co, chartings are influenced by the amount of p’ zation, these differences do not necessaril flect an actual change in ground cover. , Under sheep grazing in pasture O. tobospp decreased in density on the Ozona clay soil increased on the Tobosa clay soils. o! The amount of buffalo and curly i? grass apparently is not closely associated; the class of livestock on the pastures. l; drouth, sod areas of both of these stoloni grasses were seriously reduced on all pat As a result. the effects of grazing by shee cattle have been difficult to evaluate. The relationships between the amount ofi and the rates of sheep and cattle stockii shown in Table 3. The numbers of mesquitf apparently were not influenced by the livestock. However, the correlation abundance of pricklypear and the rate o; stocking was highly significant. r=.7.86. g grazing on pricklypear break off and sca v3, tions of these cacti. These sections, or “if take root under favorable moisture co and establish new pricklypear plants. it Utilization Studies Studies of forage Litilization by cattle were made in pastures O and August 1951 to August 1953. Forage a .' w . ERRATA Bulletin 786; "Relation of Soils, Rainfall ant Grazing Management to Vegetation -- Western L Edwards Plateau of Texas" Q The caption to Figure l7, page 20, should re=m "Lakebed area in Pasture 0 heavily stocked wig sheep. This pasture has more of the three-a}? grasses than the cattle pastures but fewer '¥i This picture, taken in 1952, shows close y utilization of buffalograss around the clumpifi purple three-awn1' 3 w Forage available for livestock use in pasture O, heavily stocked by sheep, and pasture P, heavily stocked by cattlel , F195 Foragi available in pounds per acre Aug. 1951 Dec. 1951 Aug. 1952 Dec. 1952 Aug. 1953 1958 2098 1906 781 783 e P 1237 790 917 192 348 q fhuftalo and curly mesciuitegrass ‘e O 1320 1 04 187 200 re P 252 Trace 182 37 168 '9 -+rass - e O 1238 215 29 Trace 50 re P 201 Trace 71 59 196 represent averages of the forage clipped in the grazed areas. llivestock was determined by comparisons of tation clippings made in the grazed areas with we made in temporary exclosures. a he data on forage available for livestock use astures O and P are presented in Table 4. e data show characteristic differences in the punt of forage available under sheep and cattle ling. Both pastures were stocked at the rate acres per animal unit per year during the Tod of study. The pasture grazed by sheep tistently had more forage available than the fle pasture. This difference was maintained the drouth increased in severity. As a result, lemental feeding of cattle became necessary BUFFALO a cuavr-mzsquar: iyChanqt +0 ‘Sheep (l7osI-ure O) Cltdvtqq, +0 Cf-‘IHI; (Posture P) o‘. ‘.90 9 e Q Q Q Q o Q Q e e - z O Q Q0 z .. .0 I‘ ..Q9‘ O I Q O ‘ Q 9 , e e 0 o 9 9 o. .0 °. O. .9 e. .0 9 ‘I C ‘Q Q. Q : '-. = '- e e .. .. Q O e9 , . '. 3 g e I . 9. 9 O I I . .0 o. z. ‘e e.‘ I. .. Q. e 9 e Q... O r + l : : : '. i 1. e e e 4h I; 4's 4'1 4s 4e +7 ‘t8 49 5° 5| $1 55 Pflslurc p ‘.0 0,. pasture O '0 e99 "e go ‘flu. 0°.’ ‘.90 inn‘ ' o‘. 9 l. .. e g. r 1 l_ . 1 l 4 l I l i I L 1,131,, 59 41> 4': 4; J4 4'4 43 4B 4'1 4's 49 so f! 51 $5 - TIM‘ OF VUGCTATIGN CURVQY " - re 20. Response of buffalo, curly mesquite and to- bosagrass to sheep and cattle grazing. Pas- tu.re O was grazed by cattle the first 6 years and has been grazed continuously by sheep since 1944. Pasture P was grazed by sheep for 6 years and by cattle since 1944. Both pastures have been stocked at an average rate of one annual unit per 18 acres yearlong. Figure 22. UTILIZATION 0F T080311 J U LY 1951 Figure 21. Utilization map showing stubble-height meas- urements of tobosagrass after grazing by sheep and cattle. The shaded area has been grazed to a stubble height below 6 inches. at an earlier date than supplemental feeding of sheep. This information indicates that a ratio of 6 sheep to 1 cow may be too low for the Texas Range Station. However, careful examinations of both the sheep and cattle pastures showed that sheep caused equally as much or more damage to the vegetation and soil as did the cattle. This was due to the habit of sheep to “spot graze.” A utilization survey made in July 1951, shows some differences in the amount of utilization of tobosagrass by sheep and cattle at similar rates of stocking, Figures 21 and 22. The pattern of utilization of tobosa by sheep was strongly influ- enced by wind direction and soil condition. In grazing, sheep first concentrated on the ridgetop Picture taken at the location marked in Fig- ure 21. Pasture O on the left has been graz- ed by sheep and pasture P on the right by cattle. 21 soils of the Ozona series and then drifted south- ward into the wind. An experiment was designed in 1951 to check the effect of “previous clipping” on the amount of utilization of tobosagrass.‘ Scattered areas of tobosagrass were moved in August 1951 and the amount of utilization on these areas was checked the following December and in August of 1952, one year after mowing. The data from the December 1951 clipping show a significant increase in the amount of utilization of tobosagrass by cattle on the mowed areas. However, this effect of “previous clip- ping” in the cattle pasture had disappeared by August 1952. In the sheep pasture, there was better utilization on the clipped areas, even 1 year after mowing. These data show that previous utilization of the grass species is one of the major reasons for “spot grazing.” Both sheep and cattle prefer the new growth of vegetation follow- ing close clipping or grazing. ACKNOWLEDGMENTS The authors acknowledge the following men who assisted with certain phases of this project: W. H. Dameron (deceased) and W. T. Hardy, superintendents, Ranch Experiment Station, Sonora; L. B. Merrill, range specialist, Sorvra and E. J. Compton, University of Texas. A. I ‘.- ciation also is extended to V. L. Cory, for range botanist, and to other members o.- Texas Agricultural Experiment Station who have assisted in the collection of field-data and in the organization of this grazing experiment. The contribution of the University of Texas in furnishing the land and other facilities of the Texas Range Station is gratefully acknowledged. LITERATURE CITED Allred, B. W. 1950. Practical grassland management. Publ. by Sheep and Goat Raiser Magazine, San Angelo, Texas. Canfield, R. H. 1939. Effects of intensity and frequency of clipping on density and yield of black grama and tobosa grass. U.S. Dept. Agr. Tech. Bull. 681. Carter, W. T., E. H. Templin and I. C. Mowery. 1938. Soil survey report of the Texas Range Station. Unpublished report. Clawson, Marion. 1950. The western range livestock industry. McGraw-Hill Book Co. New York. Clements, F. E. 1920. Plant indicators: the relation of plant communities to process and practice. Carnegie Inst. Wash. Publ. 290. 22 Cory, V. L. 1927. Activities of livestock on the I Texas Agr. Expt. Sta. Bull. 367. _. Craddock, o. W. and c. L. Forsling. 193s. Thei of climate and grazing on spring-fall sheep r ; Southern Idaho. U.S. Dept. Agr. Tech. Bull. 600s Dameron, Jacob T. 1950. A comparison of the i I, point quadrat, line intercept and square foot f; for evaluating composition of pasture vegetati West Texas. Unpubl. Master Thesis, Agr. and College of Texas. - Dickson, R. E., C. E. Fisher and P. T. Marion. é_ Summer grazing experiments on native grassla, Spur, 1942-47. Texas Agr. Expt. Sta. Progress 1123. Duley, F. L. and C. R. Domingo. 1949. Effect of on intake of water. N ebr. Agr. Exp. Sta. Res. Bull. -_ Dyksterhuis, E. J. 1949. Condition and managemeiv rangeland based on quantitative ecology. Jour. ' V, Mgt. 2: 104-115. ‘ Fraps, G. S. and V. L. Cory. 1940. Composition utilization of range vegetation of Sutton and Edw" Counties. Texas Agr. Expt. Sta. Bull. 586. C Hitchcock, A. S. 1951. Manual of the grasses of? United States. U.S. Dept. of Agr. Misc. Publ. 200. Kroth, E. M. and J. B. Page. 1946. Aggregate formi g in soils with special reference to cementing subs i‘, Proc. Soil Sci. Soc. Amer. 11: 27-34. . Leithead, Horace L. 1950. Field methods used to de strate range conservation. Jour. Range Mgt. 3: i: Nelson, Enoch W. 1934. The influence of precipi Y and grazing upon black grama grass range. U.S. l; Agr. Tech. Bull. 405. A Osborn, Ben. 1950. Range cover tames the rai u U.S. Dept. Agr. Soil Cons. Service. Fort Worth, T p Mimeo. . Potts, R. C. 1946. The relation between certain p and major soil types on the Texas Range Sta rIrJnpubl. Master Thesis. Agr. and Mech. College; exas. J Robinson, n. o. and J. B. Page. 1950. Soil .95 stability. Proc. Soil Sci. Soc. Amer. 15: 25-29. ' Shantz, H. L. 1938. Plants as soil indicators. In and Men. U.S. Dept. of Agr. Yearbook, pp. 835-860.,‘ Thomas, G. W. 1951. Some effects of soil differences " past grazing use on the amount and composition-a vegetation on the Texas Range Station. Unpubl. M1.‘ Thesis, Agr. and Mech. College of Texas. _§ Thomas, G. W. and V. A. Young. 1953. Occurrence of three-awn grasses as affected by grazing managem Tex. Agr. Expt. Sta. Prog. Rept. 1577. I Thomas, G. W. 1954. The relation of the vegetation; the Texas Range Station to soils, precipitation ' girazing. Unpubl. Dissertation, Agr. and Mech. Col f o Texas. i;