Shrub encroachment in arid and semiarid rangelands, a worldwide phenomenon, results in a heterogeneous landscape characterized by a mosaic of nutrient-depleted barren soil bordered by nutrient-enriched shrubby areas known as “fertile islands.” Even though shrub encroachment is considered as a major contributor to rangeland degradation, little is known about mechanisms favoring the reversibility of the early stages of this process. Here we synthesize the interactions between fires and soil erosion processes, and the implications of these interactions for management of rangelands. The burning of shrub vegetation develops relatively high levels of soil hydrophobicity. This fire-induced water repellency was shown to enhance the soil erodibility in and around burned shrub patches. The fire-induced enhancement of local-scale soil erosion results from changes in the interparticle bonding forces between the soil grains, thus altering the way moisture is retained in the soil. It has been shown—with a number of wind-tunnel studies, field-scale manipulative experiments, microtopographic measurements, and isotopic tracer studies—how the fire-erosion interactions affect the dynamics of fertility islands. Further we propose a new conceptual model of resource “island” dynamics that explains some of the findings previously reported in the literature on the interactions between aeolian processes and arid-land vegetation. In particular, we highlight the ability of fires to enhance the erodibility of nutrient-rich soils accumulated under the shrubs favoring the redistribution of soil resources, thereby contributing to the reversibility of the early stages of shrub encroachment.

This study evaluated the efficacy of prescribed fire applied within landscape-scale rotational grazing treatments to reduce mesquite (Prosopis glandulosa Torr.) encroachment and restore herbaceous productivity and cover. One-herd, multiple-paddock rotational grazing was used to accumulate herbaceous fine fuel for fires via prefire deferment and to provide periodic postfire deferment for grass recovery. Treatments were an unburned continuous-grazed control, a four-paddock-1 herd system with fire (4:1F), and an eight-paddock-1 herd system with fire (8:1F), with two replicates per treatment (1 294–2 130 ha per replicate). The management plan was to burn 25% of each system (one paddock in the 4:1F; two paddocks in the 8:1F treatments) and defer grazing during all or portions of the 9 mo (May to January) prior to burning. Deferral was “internalized” by grazing on the remaining 75% of each treatment without reducing stocking rate determined for the entire system. Mesquite cover increased on clay-loam soils from 22% to 40% in unburned paddocks over 7 yr (1995–2001). This increase, coupled with extended drought, reduced fine fuel amounts for fire and limited the number and intensity of fires that were applied. It was possible to burn one paddock in the 8:1F treatment (12.5% of total area), but not in the 4:1F treatment (25% of total area) during drought. Fires reduced mesquite and cactus (Opuntia spp.) cover by 25–79% and 24–56%, respectively, but cover of these species increased to prefire levels within 6 yr. All fires reduced (P ≤ 0.05) total herbaceous biomass for 1 yr postfire. The 8:1F treatment increased (P ≤ 0.05) grass biomass on loamy-bottom soils and reduced (P ≤ 0.05) bare ground on clay-loam and loamy-bottom soils in unburned paddocks compared to the unburned continuously grazed control. The 8:1F treatment, through internalized grazing deferment, facilitated the application of fire to reduce woody cover during extended drought without degrading the herbaceous understory.

Prescribed fire is used to reduce the rate of woody plant encroachment in grassland ecosystems. However, fire is challenging to apply in continuously grazed pastures because of the difficulty in accumulating sufficient herbaceous fine fuel for fire. We evaluated the potential of rotationally grazing cattle in fenced paddocks as a means to defer grazing in selected paddocks to provide fine fuel for burning. Canopy cover changes from 1995 to 2000 of the dominant woody plant, honey mesquite (Prosopis glandulosa Torr.), were compared in three landscape-scale grazing and mesquite treatment restoration strategies: 4-paddock, 1-herd with fire (4:1F), 8-paddock, 1-herd with fire (8:1F), and 4:1 with fire or aerial application of 0.28 kg · ha⁻¹ clopyralid   0.28 kg · ha⁻¹ triclopyr herbicide (4:1F/H), and a continuously grazed control with mesquite untreated (CU). Prescribed burning took place in late winter (February–March). Droughts limited burning during the 5-yr period to half the paddocks in the 4:1F and 8:1F strategies, and one paddock in each 4:1F/H strategy. Mesquite cover was measured using digitized aerial images in 1995 (pretreatment) and 2000. Mesquite cover was reduced in all paddocks that received prescribed fire, independent of grazing strategy. Net change in mesquite cover in each strategy, scaled to account for soil types and paddock sizes, was 34%, 15%, 5%, and −41% in the CU, 4:1F, 8:1F, and 4:1F/H strategies, respectively. Thus, rotational grazing and fire strategies slowed the rate of mesquite cover increase but did not reduce it. Fire was more effective in the 8:1F than the 4:1F strategy during drought because a smaller portion of the total management area (12.5% vs. 25%) could be isolated to accumulate fine fuel for fire. Herbaceous fine fuel and relative humidity were the most important factors in determining mesquite top-kill by fire.

Beef cattle production from rangelands in the Southern Great Plains has decreased in concert with herbaceous forage production declines in response to woody plant encroachment by honey mesquite (Prosopis glandulosa Torr.) over the past 120 yr. Combinations of livestock overstocking and fire suppression are considered to be primary drivers of these changes. This experiment evaluated cow–calf production responses over a 7-yr (1995–2001) period to ranch-scale (1 294–2 130 ha) integrated restoration strategies involving prescribed fire and grazing management. Restoration strategies tested in this year-round grazing ecosystem were 4-pasture, 1-herd rotation with fire (25% of pasture acreage burned each year; 4:1F); an 8-pasture, 1-herd rotation, with fire (8:1F); and a 4-pasture, 1-herd, with fire and aerial application of 0.28 kg · ha⁻¹ clopyralid   0.28 kg · ha⁻¹ triclopyr herbicide (4:1F / H). Restoration strategies were compared to a continuous grazing strategy with no mesquite treatment. All cattle stocking rates were moderate (7.5–15 ha · animal unit⁻¹ · year⁻¹) and all fires were applied during late winter. Beef cattle (cow–calf) production variables measured included conception rate, weaned calf percentage, weaning weight, weight of calf per exposed cow, weight of calf per hectare, and supplement fed per cow. We observed significant differences in beef production among strategies primarily during the first 2 yr where the continuous grazing strategy exhibited better overall livestock production than the integrated restoration strategies. Differences in livestock production among strategies were minimal over the last 5 yr of the study. These livestock production results suggest livestock and management adapted to restoration strategies after the first 2 yr. Results point to the need to cautiously transition into integrated grazing and fire restoration strategies when cattle and management are changed and intensified from prior historical protocols.

Distinguishing between anthropogenic and natural sources of sediment and nutrients is important for water resource management in irrigated basins. Water quality of flood irrigation was monitored at the field scale in the upper Klamath Basin, Oregon, on two unfertilized cattle pastures that were 2 ha (Site 1) and 70 ha (Site 2) in area. Water samples were analyzed for concentrations of sediment, total dissolved nitrogen (TDN), total dissolved phosphorus (TDP), orthophosphate, ammonium-N (), and nitrate-N (). At both sites the TDN concentration was significantly greater in surface runoff than in applied irrigation water (P < 0.05). Site 1 sediment and TDP concentrations were significantly greater in irrigation surface runoff than in applied irrigation water (P < 0.05). A first flush during irrigation was observed at Site 1 where nutrient concentration was at maximum value during the first 3 h of surface runoff. At Site 2 the surface runoff sediment and TDP concentrations were not significantly (P > 0.05) higher than the applied irrigation, except when cattle were present. When export was measured, the mean yield of sediment and TDN per irrigation was 23.9 kg N · ha⁻¹ and 0.26 kg N · ha⁻¹, respectively, and there was a net retention of TDP of 0.04 kg P · ha⁻¹. export occurred during one irrigation event yielding 0.15 kg N · ha⁻¹. export was minimal or undetected. A late summer storm event resulted in pasture surface runoff concentrations of TDN and TDP that were 33 and 3 times higher, respectively, than irrigation source water concentrations. The TDN was significantly higher in subsurface runoff than it was in applied irrigation water (P < 0.05). Improved irrigation efficiency might prevent many of the nutrient and sediment transport mechanisms observed during this study.

We examined the independent and combined effects of prescribed fire and livestock grazing on herbaceous plant production in shortgrass steppe of northeastern Colorado in the North American Great Plains. Burning was implemented in March, before the onset of the growing season. During the first postburn growing season, burning had no influence on soil moisture, nor did it affect soil nitrogen (N) availability in spring (April–May), but it significantly enhanced soil N availability in summer (June–July). Burning had no influence on herbaceous plant production in the first postburn growing season but enhanced in vitro dry matter digestibility of blue grama (Bouteloua gracilis [Willd. ex Kunth] Lag. ex Griffiths) forage sampled in late May. For the second postburn growing season, we found no difference in herbaceous plant production between sites that were burned and grazed in the previous year versus sites that were burned and protected from grazing in the previous year. Our results provide further evidence that prescribed burns conducted in late winter in dormant vegetation can have neutral or positive consequences for livestock production because of a neutral effect on forage quantity and a short-term enhancement of forage quality. In addition, our results indicate that with conservative stocking rates, deferment of grazing during the first postburn growing season may not be necessary to sustain plant productivity.

A longitudinal study of California hardwood rangelands shows significant change in landowner characteristics and goals. Results of three studies spanning 1985 to 2004 were used to develop and evaluate a multiagency research and extension program known as the Integrated Hardwood Range Management Program. Program-sponsored education and research aimed at encouraging landowners to change woodland management has been reflected in a significant reduction in oak cutting and an increase in oak planting. Recent changes have come with the times: landowners were as likely to have consulted land trusts about oaks as Cooperative Extension, and the number engaged in production of crops or livestock continued to decline. On the other hand, the proportion of landowners, including ranchers, reporting that they live in the oak woodland to benefit from ecosystem services such as natural beauty, recreation, and lifestyle benefits significantly increased. Though owners of large properties and ranchers were more strongly against regulation and “government interference” than other respondents, this did not appear to affect oak values and management. Property size remained significantly related to landowner goals, values, and practices, with those producing livestock owning most of the larger properties. There has been a decline in the number of properties being studied due to conversion of some from oak woodland to other uses, though the remaining respondents still own at least 10% of the woodlands. Landowners with conservation easements or those who are willing to consider them, who believe oak recruitment is inadequate, or who use advisory services were significantly less likely to cut oaks and more likely to plant them. Policy, management, and outreach that support synergies between production and conservation activities, and that combine ecosystem service-based income streams that encourage keeping land intact and increased land-use stability, are needed to support conservation of private rangelands.

Sound rangeland management requires accurate information on rangeland condition over large landscapes. A commonly applied approach to making spatial predictions of attributes related to rangeland condition (e.g., shrub or bare ground cover) from remote sensing is via regression between field and remotely sensed data. This has worked well in some situations but has limited utility when correlations between field and image data are low and it does not take advantage of all information contained in the field data. I compared spatial predictions from generalized least-squares (GLS) regression to a geostatistical interpolator, regression kriging (RK), for three rangeland attributes (percent cover of shrubs, bare ground, and cheatgrass [Bromus tectorum L.]) in a southern Idaho study area. The RK technique combines GLS regression with spatial interpolation of the residuals to improve predictions of rangeland condition attributes over large landscapes. I employed a remote-sensing technique, object-based image analysis (OBIA), to segment Landsat 5 Thematic Mapper images into polygons (i.e., objects) because previous research has shown that OBIA yields higher image-to-field data correlations and can be used to select appropriate scales for analysis. Spatial dependence, the decrease in autocorrelation with increasing distance, was strongest for percent shrub cover (samples autocorrelated up to a distance [i.e., range] of 19 098 m) but present in all three variables (range of 12 646 m and 768 m for bare ground and cheatgrass cover, respectively). As a result, RK produced more accurate results than GLS regression alone for all three attributes when predicted versus observed values of each attribute were measured by leave-one-out cross validation. The results of RK could be used in assessments of rangeland conditions over large landscapes. The ability to create maps quantifying how prediction confidence changes with distance from field samples is a significant benefit of regression kriging and makes this approach suitable for landscape-level management planning.

Plant functional types (PFT) have been used to describe the response of native vegetation to environmental factors (i.e., fertility) and to livestock disturbance, but rarely under conditions of continuous grazing. In this work we investigate whether the long-term response of grassland communities submitted to a gradient of continuous grazing pressure can be described with such an approach. After 15 yr of differentiation of the grazing pressure applied to native grasslands we measured leaf dry-matter content (LDMC) and specific leaf area (SLA) of Poaceae populations of the communities. A grazing pressure gradient was created by levels of daily forage allowance: 4, 8, 12, and 16 kg of dry matter per day per 100 kg of animal live weight, monitored monthly. PFTs were defined by numerical analysis, where an algorithm finds the optimal trait subset based on the agreement between matrices of species × traits, paddocks × grass biomass, and environmental variables (levels of forage allowance and soil characteristics). The results show that it is possible to describe a gradient of grazing pressure by means of LDMC and/or SLA measured only on the Poacea contributing at least 80% of the total Poaceae biomass. Four PFTs were differentiated by these leaf traits. PFTs having low LDMC and high SLA are characteristic of high intensity of use and are made up largely of stoloniferous C₄ species typical of rapid resource capture strategies. Conversely, PFTs characterized by high LDMC and low SLA include species that are representative of low grazing pressure. Variations in the aggregate value of traits are due to changes in the species proportions and not to leaf-size adaptation as hypothesized. We conclude than in the absence of a gradient of fertility, plants with strategies of resource capture tend to be more represented under high grazing pressures. This situation results in a loss of functional diversity, but in particular a reduction in forage availability, which is incompatible with high animal production.

Grasslands of the Edwards Plateau of central Texas have been extensively altered through woody species encroachment, particularly as a result of increasing abundance of the invasive native shrub, Ashe juniper (Juniperus ashei). Over the last several decades there has been widespread mechanical removal of the species. The wood is often left in place to decompose, either mulched or not. Where the wood is left to decompose might have some bearing on its rate of decomposition. This study was conducted to determine the rates of Ashe juniper wood decomposition as a function of open vs. shaded habitat and the potential effect of wood decomposition on nutrient inputs into this system. Wood decomposition in this arid ecosystem might be expected to occur more rapidly in shaded habitat where the moisture and temperature regimes would be more favorable for wood-decomposing fungi. On the other hand, during times of low rainfall we might expect wood to decompose more rapidly when exposed to high levels of ultraviolet radiation. In our experiment, we found no difference between open and shaded treatments. Wood biomass loss occurred rapidly over the first 3–4 mo of the study and slowed for the remaining 2 yr. Wood carbon (C) increased only slightly (7.3%), but nitrogen (N) increased significantly (176%). As a consequence of changes in wood nitrogen, C∶N decreased through time. Results of this study suggest that the wood decomposition process in open and shaded habitats in this arid ecosystem during a time of low rainfall do not differ. Our findings also suggest that land managers aiming to establish native species following felling of Ashe juniper should do so in the first year when nutrient release from decomposing wood is the highest.

Redberry (Juniperus pinchotii Sudw.) and ashe (Juniperus ashei Buchh.) juniper dominate rangelands throughout central Texas. Our objective was to attempt to improve the efficacy of goats as a biological control mechanism for juniper through behavioral training. Conditioning sheep and goats to increase the palatability of chemically defended plants can be a useful tool in brush control. Previous research illustrated that goats can be conditioned to consume more juniper while in individual pens when foraging choices are limited. To test whether this creates a longer-lasting increase in juniper preference, we determined if goats would continue to consume juniper on pasture for one year after being fed juniper in individual pens for 14 d. Female Boer-cross goats (n  =  40) were randomly divided into two treatments: conditioned and naive to juniper. At approximately 12 mo of age, conditioned goats were placed in individual pens and fed redberry juniper 1 h daily for 14 d, while naive goats received only alfalfa pellets to meet maintenance requirements. After the pen-feeding phase of the study, goats were placed in one of four pastures (10 goats · pasture⁻¹) for 12 mo. Two pastures housed conditioned goats, and two pastures housed naive goats at a moderate stocking rate (1 animal unit · yr⁻¹ · 8 ha⁻¹). Bite count surveys were conducted twice per month, while herbaceous standing crop and monoterpene levels were measured once per month. Juniper preference varied monthly; however, conditioned goats consistently ate more (P < 0.05) juniper than naive goats except for April, when the study began, and March, when the study ended. When selection of herbaceous forages decreased, conditioned goats increased selection of juniper, while naive goats increased selection of other palatable shrubs. Seasonal changes of monoterpene levels in juniper had no apparent effect on juniper preference. We contend that feeding juniper at weaning will increase use of the plant in grazing situations.

Suppression of one-seed juniper (Juniper monosperma [Englem.] Sarg.) reinvasion with goats requires achieving levels of defoliation of newly established saplings that eventually kill or suppress plant growth. We tested the effects of stocking density and mixed grazing with sheep on the level of use of one-seed juniper saplings by goats. In summer and spring, groups of 10 does (goats alone, GA) or 5 does and 4 ewes (mixed grazing, MG), grazed 20 × 30 m cells infested with saplings (500–533 · ha⁻¹; mean: 0.8 m tall), either continuously for 6 d (low stocking density, LD) or with daily rotation through 10 × 10 m cells during the 6-d period (high stocking density, HD) in a block design. Feeding activity; juniper in feces; utilization of herbaceous vegetation; frequency of saplings with light, moderate, and heavy foliage and bark use; and branch utilization were determined. Goats in HD spent more time feeding on saplings, less time feeding on herbaceous forages, and tended to consume more juniper than goats in LD. Utilization of herbaceous vegetation ranged from 52% to 73% and was higher for MG than GA and for LD than HD. The MG–HD treatment resulted in the highest frequency of short saplings (< 0.5 m) with heavy defoliation in summer and spring, and lowest frequency of saplings with light debarking in spring. Heavy defoliation was more frequent in short saplings, whereas heavy debarking was more frequent in tall (> 1 m) saplings. Sapling mortality was not affected by treatments (P > 0.05) and averaged 5% across treatments. Branch debarking was greater in spring (P  =  0.02) and explained approximately 80% of branch mortality and 62% and 52% of the reduction in sapling live crown height and volume. Branch utilization (percent length) was not affected by grazing treatments (range: 45–48%), but was influenced by the length and diameter of branches. This study suggests that high stocking density and mixed grazing stimulate feeding behaviors that increase utilization of juniper saplings by goats. Susceptibility of saplings to defoliation and debarking varies with sapling size, branch structure, and season. Targeted grazing in spring appears to have a greater impact on sapling suppression and branch mortality due to higher debarking frequency.

Microhistological analysis of feces is the most applied noninvasive method for assessing diets of wild ungulates. However, the method is complicated by differential digestibility of forage species. To evaluate the efficacy of this method in quantifying browse components in summer diets of moose (Alces alces L.) on Norwegian rangelands, we compared it to parallel field surveys of browsed vegetation on the same range. Although the same principal diet components were identified in the feces and in the field, there were consistent discrepancies between the two methods in estimated proportional diet contents. Birch (Betula spp.) showed the highest field:fecal ratio: 3.3 ± 0.50 compared to 0.9 ± 0.16 for Salix spp., 0.8 ± 0.16 for aspen (Populus tremula L.), and 0.6 ± 0.12 for rowan (Sorbus aucuparia L.). Until in vivo fecal correction factors for differential forage digestibility are available, we caution against broad application of fecal analyses for estimating proportions of browse in moose diet. Although we could not determine the exact amount of discrepancy implicit in each method, previous studies of moose summer diet in the area clearly indicate that fecal analyses gave a less accurate representation of actual moose browse diet than did the field survey. Fecal analyses are nevertheless needed to identify moose diet components other than browse, which are not easily obtained from field surveys.