Next Article in Journal
Towards a Carbon Accounting Framework for Assessing the Benefits of Biogenic Wood Carbon to Net Zero Carbon Targets
Previous Article in Journal
Drought Exerted a Stronger Controlling Effect on Soil Carbon Release than Moisturizing in a Global Meta-Analysis
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Forests
Volume 14
Issue 10
10.3390/f14101958
forests-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

How Weed Control Affects Eucalyptus globulus Labill. Productivity: Results from Two Long-Term Trials

by
Sofia Corticeiro
1,2,3,*,
Paula Maia
1,4,
Catarina Gonçalves
3,
Jan Jacob Keizer
2 and
Carlos Valente
3
1
Centre for Environmental and Marine Research (CESAM), University of Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal
2
Department of Environment and Planning, University of Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal
3
RAIZ, Instituto de Investigação da Floresta e Papel, Quinta de S. Francisco, Eixo, 3800-783 Aveiro, Portugal
4
Department of Biology, University of Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal
*
Author to whom correspondence should be addressed.
Forests 2023, 14(10), 1958; https://doi.org/10.3390/f14101958
Submission received: 19 July 2023 / Revised: 21 September 2023 / Accepted: 26 September 2023 / Published: 27 September 2023
(This article belongs to the Section Forest Ecology and Management)
Browse Figures
Review Reports Versions Notes

Abstract

:
Weed control is considered a critical management operation for the establishment and growth of Eucalyptus globulus and is often performed during the first two years following planting. However, no information has been found related to the effects of weed management on the long-term growth of E. globulus. This study aims to better understand how adjusting the timing of weed control, beyond tree establishment, affects the productivity of eucalypts on two commercial plantations in Central Portugal. Two treatments were considered: weed control when vegetation cover occupied at least 50% of the area with a mean height of at least 50 cm, and no weed control. At the northernmost site, weed control operations were performed during the first 3 years following planting. At the southernmost site, weed control interventions occurred between 1.5 and 8.2 years. Weed control demonstrated to be effective at diminishing vegetation cover density to a maximum of 79% and 94% in the northmost and southmost sites, respectively. Weed control significantly increased E. globulus productivity by the end of the study, demonstrating that the timing of weed management must be adjusted in accordance with the understory competing vegetation cover (weeds) and during the entire rotation, not only during tree establishment.

1. Introduction

Eucalyptus globulus Labill. is a highly relevant forest species widely planted around the world, providing raw material for the pulp and paper industry because of its rapid growth and high cellulose production [1,2,3,4,5,6]. E. globulus plantations in Portugal account for 26% of the total forest area, occupying 812,000 ha in 2015, with a mean productivity of 10 m3 ha−1 yr−1 [7].
Considerable efforts have been made to increase the productivity of E. globulus plantations and, thereby, provide guarantees for economic rentability [6]. In general, productivity is a crucial topic in forest management [8]. It is of particular relevance in areas where low rentability easily leads to land abandonment and, in turn, to the encroachment of invasive species and the intensification of rural fire regimes [9,10]. Forest managers and producers have been investing in the implementation of efficient silviculture practices, such as adequate soil tillage, genotype selection, weed control, and fertilization, to increase forest productivity [11].
Among forestry practices, special attention has been given to weed management operations, as weeds may significantly affect the establishment and productivity of forest species [12,13]. Weeds can interfere with tree growth directly through competition for light, water, and nutrients [13,14,15], but also indirectly by increasing fire hazards due to their contribution to the accumulation of biomass as fuel load [16]. The negative effects of weeds on E. globulus establishment and growth also depend on vegetation cover and composition [15]. Herbaceous species have a greater direct effect during planting and establishment, whereas shrub species have a stronger impact after canopy closure [17].
The impact of weeds during the early phases of E. globulus establishment is well documented [13,14,15,18]. The initial timing of weed control operations is essential for its effectiveness during the so-called critical weed-free period [19], as at that stage, vegetation usually grows faster than young trees [15]. Although in the case of E. globulus, the critical weed-free period was reported to be 20 months after planting [14], in Portugal, weed control on E. globulus plantations typically involves mechanical and/or chemical operations during the first year following planting and, if needed, again between the third and fifth years [6].
Few studies have monitored the impact of weed control on the productivity of E. globulus for a whole rotation [15,20], and even in those, vegetation management operations were only performed during the first two years following planting. A gap of knowledge was found in how varying the timing of weed control interventions beyond the critical weed-free period may affect the long-term growth of E. globulus. This knowledge is highly relevant, not only to the productivity of E. globulus but also to the sustainability of plantations. Sustainable forest management must be based on procedures that raise forest productivity without endangering the ecosystems or the ecosystem services provided by the forests [21]. To achieve this balance, forest managers need to have a thorough understanding of how to adjust management practices to local physical and environmental conditions, thus minimizing the environmental impacts of their interventions.
This study was conceived under the hypothesis that we can potentiate the productivity of E. globulus throughout the whole rotation by adjusting the timing of weed management according to site conditions. Thus, the goal of this study was to determine how weed control, scheduled according to the understory vegetation development, can affect the long-term productivity of E. globulus. Under this context, two field trials were installed on plantations of E. globulus located in Central Portugal: one in the province of Beira Litoral, and another further south in the province of Estremadura.

2. Materials and Methods

2.1. Site Selection and Characterization

The present study was conducted in two E. globulus plantations in the Central Region of Portugal managed by The Navigator Company (Figure 1). The southernmost trial was established in 2009 at Torres Vedras, district of Lisbon (39°05′11.56″ N, 9°10′31.78″ O, 242 m above sea level (ASL)). The northernmost trial was established in 2015 at Soure, district of Coimbra (40°05′11.56″ N, 8°35′45.96″ O, 80 m ASL). Both study areas are located in the Mediterranean Csb climate zone, temperate climate with dry and mild summers, according to the Köppen classification [22,23], with an annual mean precipitation (1971 to 2000) of 638 mm in Torres Vedras [22] and of 741 mm in Soure [23]. In Torres Vedras, the annual mean temperature was 15.8 °C, with a mean monthly minimum temperature of 9.2 °C and a mean monthly maximum temperature of 24.1 °C [22]. In Soure, the annual mean temperature recorded was 15 °C, with mean monthly temperatures ranging from 4.4 °C to 28.6 °C [23]. Soils in both study areas are mainly epileptic and endoletic arenic Regosols [24] (Table 1).
These E. globulus plantations were being managed for wood. The establishment and development of both plantations were performed as previously defined in the forest company’s management planning: clear-cut harvesting of the preceding E. globulus planting followed by soil tillage with a single iron plough and planting of young cuttings. E. globulus clones were planted with 4.0 × 2.0 m spacing (1250 trees ha–1) at both sites.

2.2. Experimental Design

At both sites, plots were randomly established following plantation lines. A plot consisted of a group of three or four measurable consecutive trees in the same plantation line, at Torres Vedras and Soure, respectively. The first plot was installed in the second line of the plantation after the road. Successive plots were separated by at least one tree, with a minimum distance of 4 m between the nearest trees of plots of the same plantation line. No plots were installed in consecutive plantation lines. A plantation line with control and herbicide plots was separated from the following by a line of trees with no plots installed, with a minimum distance of 8 m between the nearest trees of plots of different plantation lines (Figure 2). Two to ten plots were established on each plantation line, depending on the local site conditions and the number of viable trees within a line. Overall, at Torres Vedras 55 plots (165 trees) were distributed along 9 plantation lines. At Soure, a total of 32 plots were established, 16 plots by treatment, corresponding to a total of 120 measured trees. One of the following treatments was applied to each plot: no weed control or chemical weed control. The treatment applied to each plot was defined following a random sampling strategy to account for differences in vegetation characteristics and/or soil characteristics. In each plot and along the plantation line, herbicide was applied from 1 m before the first tree to 1 m after the last tree of the plot. Plots with different treatments were separated by a minimum of 3 m between the last point where herbicide was applied until the first tree of the following control plot. Between plantation lines, the herbicide was applied up to 2 m on each side of the trees in the plot.

2.3. Weed Control Treatments

The plots with no weed control treatment, hereby denominated as control plots, did not suffer any intervention related to the understory vegetation. In the plots subjected to weed control, hereby denominated as herbicide plots, vegetation control was carried out in the designated areas with the herbicides Roundup UltraMax, 360 g/L glyphosate (1% to 5%), or Garlon, 480 g/L triclopyr (0.3%). Glyphosate (Roundup UltraMax) was used for the broad-spectrum control of weeds, while triclopyr herbicide (Garlon) was only used for selective control of brambles (Rubus spp.). The weed control interventions and their timing were defined following internal guidelines of best practices and forest certification standards (FSC 2016; PEFC 2018). The treatment was repeated when the understory vegetation cover developed to a percentage of 40% to 60% of the plot area, with a mean height of at least 50 cm. Following these criteria, in Torres Vedras, the vegetation development required weed control operations to be conducted at 1.5 years (T0), 5.2 years (T1), and 8.2 years (T2). The final evaluation of trees (height (H) and diameter at breast height (DBH) (1.30 m from the ground)) was performed at 8.8 years (T3), only 0.6 years after the last herbicide application. In Soure, herbicide applications were performed at 1.6 years (T0), at 1.9 years (T1), and at 3.1 years (T2). The interval in herbicide application between T0 (fall) and T1 (spring) was due to the considerable increase in herbaceous plant cover during early spring. These were mainly graminoids and short-lived species that could, nonetheless, compete with the young, recently planted eucalyptus. The final measurement of trees in the Soure site was considered as 5.2 years after planting (T4), since a windstorm that occurred after that period caused considerable and unexpected damaged to the plantation, due to the high number of trees that fell in the experimental area.

2.4. Vegetation Cover

The floristic inventory (weed cover) was recorded by sampling five control plots and five herbicide plots, randomly selected within each treatment. The main weed species identified in the Torres Vedras trial were Cistus crispus, Rubus ulmifolius, Andryala integrifolia, Brisa maxima, and Holcus lanatus. In the Soure experiment, Juncus effusus and Rubus ulmifolius were the dominant weed species found. Plant cover was visually estimated at the plot level, considering cover estimations of 10% intervals: 0%–10%, 10%–20% … 90%–100% [25,26]. To examine the influence of “weed” functional groups on E. globulus productivity, plant species were classified as herbaceous or woody plants. To evaluate the effectiveness of weed control treatments on weed cover, the cover of the understory vegetation was assessed at the ages of 2.5 (2010), 5.6 (2013), 6.2 (2014), 7.2 (2015), 8.2 (2016), and 9.2 years (2017) in Torres Vedras and at the ages of 2.1 (2016), 3.1 (2017), and 5.0 years (2019) after planting in Soure.

2.5. E. globulus Productivity

Tree height (H) and DBH were initially measured at the age of 1.5 (Torres Vedras) and 1.6 years (Soure), immediately prior to the first application of herbicide to determine the initial biometric variables of E. globulus and, therefore, for initial productivity assessment. H and DBH measurements were repeated once a year for growth monitoring until the ages of 8.8 and 5.2 years in Torres Vedras and Soure, respectively. The timber commercial volume of each individual tree (VOL) was calculated from DBH and H values, following the equation Vol = (0.2105 × (DHB/100)1.8191 × H1.0703) × 1000 [27].
E. globulus tree growth was also calculated according to the time intervals of herbicide application. In the Torres Vedras trial, the increment in H, DBH, and VOL was quantified from 1.5 years (T0) to 5.2 years (T1), from 5.2 years to 8.2 years (T3), and from 8.2 years to 8.8 years (T4). In Soure, the differences in H, DBH, and VOL were measured from 1.6 years (T0) to 1.9 years (T1), from 1.9 years to 3.1 years (T3), and from 3.1 years to 5.2 years (T4). The time frame between applications was not equal in both experiments due to the criteria defined for weed control (vegetation cover of 40% to 60% of the area around E. globulus trees with an average height equal to or over 50 cm). Thus, in order to standardize growth, the mean annual increment in tree volume (MAI) was calculated as the ratio between the increment in VOL and the corresponding period of time to which the increment was calculated (in years).

2.6. Statistical Analysis

The vegetation cover and height data were analyzed using generalized linear models (GLMs) created through a stepwise backward selection procedure, following a Gaussian error distribution and using the identity link function. The E. globulus H, DBH, and VOL data were analyzed using one-way ANOVA. Changes in DBH, H, and VOL between consecutive herbicide applications were analyzed using two-way ANOVA and ensuing least square difference (LSD) post hoc tests. ANOVA assumptions of normality of the errors and homoscedasticity were tested and validated prior to data analysis. All statistical analyses were carried out with the R software package, version 4.1.2, using routine stats (4.1.2), ggpubr (0.4.0), and dplyr (1.0.8), and using an α of 0.05.

3. Results

3.1. The Effect of Weed Control Treatment on Plant Cover

At both the Torres Vedras (Figure 2) and at Soure (Figure 3) study areas, vegetation cover was consistently lower in the herbicide plots, throughout the entire duration of the experiment. Overall, the plant cover registered in the herbicide plots was 21.9%, while the plant cover in the control plots was 44.6%. In relation to the control plots, the maximum vegetation cover reduction in the herbicide plots was 79% and 94% in Soure and in Torres Vedras, respectively.
Woody plant cover was consistently lower than herbaceous plant cover in both study areas (20% vs. 29% in Soure and 34% vs. 43% in Torres Vedras). Notwithstanding, the data suggest that the herbicide treatment produced a higher reduction on woody than on herbaceous cover, since the relative differences between control and herbicide plots were higher for the woody (87.8% in Soure and 52.1% in Torres Vedras) than for the herbaceous cover (69.5% in Soure and 23.8% in Torres Vedras) (Figure 3 and Figure 4).
Time after herbicide application (years after treatment, Table 2) had a direct relation with overall plant cover in both study areas; however, according to the GLM, the differential effect of the herbicide for the two functional types does not seem to be consistent for all the sampling years. The best model estimated for explaining the variability in the plant cover data, to a significant extent, included all of the above-mentioned factors: treatment (control or herbicide), plant functional type (herbaceous or woody), and years after treatment, in addition to the effect of the year of the plant cover assessment and the interaction therein between treatment and cover assessment year. The study area was not a significant factor in explaining the variability in plant cover, so it was removed from the first version of the model estimation (Table 2).

3.2. E. globulus Productivity

At the Torres Vedras study site, the mean height of E. globulus in the control plots ranged from 4.1 m at 1.5 years to 17.5 m at 8.8 years, while the DBH varied from a mean value of 3.5 cm to 14.9 cm (Table 3). In terms of mean VOL per tree, values went from 2.3 dm3 (at 1.5 years) to 148.8 dm3 by the final evaluation of the trial (at 8.8 years). In the Soure study site, the E. globulus H in the control plots was 6.0 m at 1.6 years, reaching 15.2 m at an age of 5.2 years, when the experiment was terminated due to a windstorm (Table 4). The DBH of the same E. globulus trees ranged from the mean value of 9.0 cm (3.2 years) to 12.5 cm. Regarding the VOL of E. globulus trees in the same time frame (from 3.2 to 5.2 years), the variation found was from 38 dm3 to 91.4 dm3.
Growth (increment) in H, DBH, and VOL was calculated for the periods between herbicide applications. The results suggest that the H growth increment (Figure 5) was significantly higher (p < 0.05) in the herbicide-treated plots in the time period between the first (T0) and second (T1) applications and between the third application (T2) and the end of the trial (T3). No significant differences in H increase were observed between the second and the third applications among treatments.
Similar results were obtained relating to the average tree DHB. Differences in the increment of the DBH (Figure 6) were significantly higher in the herbicide-treated plots during the first (T0) and second (T1) applications and between the last application (T2) and the end of the trial (T3). No significant differences were observed between the second and the third applications in both variables.
By analyzing the increment in VOL during the same periods of time (Figure 7), differences between treatments were more evident. Trees from the herbicide-treated plots presented a significantly (p < 0.05) higher increment between all application intervals. Between T0 and T1, the VOL increased by 51.9 dm3 and 42.2 dm3 in the herbicide-treated plots and in the control plots, respectively. The same tendency was observed between T1 and T2, with trees from the herbicide plots increasing by an average of 102.7 dm3 in VOL, while trees from the control plots grew 85.1 dm3. From the last herbicide treatment (T2) to the final evaluation (T3), trees increased by 27.1 dm3 and 19.2 dm3 in the herbicide and control plots, respectively.
The mean annual increment in tree volume (MAI) was significantly higher (p < 0.05) in trees from the herbicide plots than in trees from the control plots for the timing T2-T3 (Table 5), showing a rising trend throughout the three periods of study with MAI values of 16.7 dm3 year−1 (T0-T1), 30.2 dm3 year−1 (T1-T2), and 33.9 dm3 year−1 (T2-T3).
The increment in height was significantly different (p < 0.05) between treatments after the first time period (T0-T1), with trees from the weed control plots presenting a growth of 4.4 m, higher than the increase of 3.9 m determined in the height of trees from the control plots (Figure 8). In the second period (T1-T2), the trees had a mean height increment of 1.7 m, and in the third period (T2-T3) of 3.7 m, irrespective of the treatment.
The DBH increment was only determined in the third time period after the last herbicide application (T2-T3). During this period, the diameter of trees from the herbicide plots differed significantly from the DBH of trees from the control plots, with increments of 3.8 and 3.5, respectively (Table 6).
Similar to the DBH results, the increment in volume in the Soure trial was only determined in the time frame between the final herbicide application and the last biometric evaluation of trees (T2-T3). The results showed that the volume increment in trees from the herbicide plots was significantly higher (p < 0.05) than the mean increment in volume of trees from the control plots: 63.8 dm3 and 53.4 dm3 in the herbicide and control plots, respectively. The MAI calculated between the ages of 3.1 years and 5.2 years (T2-T3) was also significantly higher (p < 0.05) in trees from the herbicide plots compared with those of the control plots, with values of 31.9 dm3 year−1 and 26.7 dm3 year−1, respectively.

4. Discussion

Weed competition is a highly relevant issue in forest management [14,18], particularly due to the limited resources available in young tree establishment and development in spite of understory vegetation growth [15,17]. For this reason, the timing of weed control operations is a major concern to forest managers trying to increase productivity and maintain plantation sustainability at a reasonable cost. In the current work, the prescription for herbicide application was dependent on an objective criteria of understory vegetation development: area occupied by the understory vegetation cover (covering 40% to 60% of the plot) and a mean height of at least 50 cm. This resulted in a differential timing for the application of herbicide: in the Torres Vedras plantation, weed control operations were performed at 1.5, 5.5, and 8.2 years; in Soure, weed control treatments were concentrated in the first 3 years after planting, namely at 1.6, 1.9, and 3.1 years.
Both trials evidenced the effectiveness of the weed control treatments approximately one to two months after herbicide application, with vegetation cover reductions to half (2013) and to one-third (2016) at Torres Vedras and Soure, respectively. The effects of weed control on understory vegetation are usually noticeable within the first weeks after treatment [28,29], but often do not last long depending on site conditions and the composition of the plant community [30]. The duration of the herbicide effect was longer in the Soure study area: two years after the last weed control operation, the area occupied by understory vegetation was about six times lower in treated plots than in control plots. In the present study, the different timings of weed control interventions in both trials were a direct consequence of the criteria established for vegetation control, which did not allow to concentrate the three herbicide applications during the first years after planting at Torres Vedras. A more intensive regime of herbicide applications that allows the concentration of all the interventions in a short period of time may have more effect on controlling the growth and modulating the composition of the understory vegetation, as it could highly diminish competition between forest species and weeds, often during the entire rotation [29]. Herbaceous vegetation typically competes heavily for nutrients and water during the early years of stand formation, while woody vegetation normally develops into a powerful competitor even after crown closure [15].
Both trials presented a positive growth response to the herbicide treatment in terms of H, DHB, and VOL, as reported by other authors [13,14,15,18,29,31]. A possible explanation may be the increment in the leaf area index and the photosynthetic rate of trees with less competition from understory vegetation, as reported by other authors [18,32]. The results from the experiment demonstrated that at both sites, the understory vegetation development and E. globulus growth responded to weed control differently, possibly due to climate conditions, the timing of weed control operations, and understory vegetation composition [13,17,30,31]. Differences between treatments were evident from the start of the trial in Soure, being only perceived in Torres Vedras at a later stage of the experiment. Such a response can be attributed to differences in the competition between weeds and E. globulus among the two areas due to the differential herbicide application regime [29].
Weed competition impacted E. globulus productivity mainly during the early phase of plantation establishment [5,14,15], but our results demonstrated that later weeding operations also contributed to growth increment at the end of the plantation cycle. Inail et al. [31] found significant differences between the DHB and VOL of E. globulus pellita from plots treated with herbicide during the first year and plots treated with herbicide after initial tree establishment, with the last presenting larger mean diameters. Nevertheless, the maintenance of such differences in stand productivity at the end of the rotation was dependent on the site productivity potential [31]. During tree establishment, competition for light, water, and nutrients is limiting for young E. globulus trees [14]. In Soure, during the first five years after plantation establishment, the weed control effect on E. globulus growth profile was more evident in DBH than in H. The same tendency was observed in Torres Vedras, although no significant differences were detected between treatments at that time. No phytotoxic impacts related to herbicide applications were observed in E. globulus trees in either trial. Weed competition substantially reduced the growth of Eucalyptus pellita, particularly in low-productivity sites, but this effect was less evident in high to medium-productivity sites [31]. Our results suggested that, although there were differences in site productivity between the two trials, productivity gains during the first five years of E. globulus growth, before canopy closure, were apparently more dependent on the timing of weed control operations (defined by weed cover) than on the number of operations per se, as differences in the height increment between treatments at Soure were significantly different between 1.6 and 1.9 years.
After tree establishment, competition for light decreases due to higher average tree height in relation to understory vegetation [17]. While Vargas et al. [15] observed a tendency of decreasing in stand productivity loss with the increase in the intensity of weed control treatments, this effect was reported to be more pronounced before canopy closure, and gradually diminishes with age [31]. However, late weed control, such as what occurred in Torres Vedras, may produce significant effects at the end of the E. globulus rotation, even in sites where differences were not evident at younger ages. The continued response to weed control at Torres Vedras may be due to the type of competing weeds, as brambles easily dominated the control and herbicide plots soon after weed control treatments had been applied. The late increment in E. globulus productivity may be explained by a late reduction in the competition for water and nutrients [15,31], as a relevant decrease in the understory vegetation cover of the herbicide plots was only observed after the last treatment.
Weed control led to significant increments in E. globulus growth and stand productivity in both trials, even though the herbicide application timings were different. Under the scope of sustainable forest management, efforts need to be made to reduce the number of chemical interventions in forestry and to mitigate its ecological impacts [33,34]. Studies such as the present one are extremely important in understanding in which situations the gains obtained in terms of forest productivity justify the use of herbicides.

5. Conclusions

Our results evidenced the effectiveness of weed control on diminishing vegetation cover density to a maximum of 79% and 94% in the Soure and Torres Vedras sites, respectively. Moreover, weed control operations led to significant increments in E. globulus growth and, hence, on final stand productivity. Both experiments revealed this effectiveness, even if in different manners. The effects of herbicide application were more noticeable during the first 3 years following planting in the case of the Soure study site, while they only became significant towards the end of the first rotation in the case of the study site of Torres Vedras. This contrast between the two sites could be related to differences in edaphoclimatic conditions and in understory vegetation composition. This study demonstrated that managing the understory vegetation on eucalyptus plantations requires more than one weed control timing. To maximize productivity, however, the timing of the operations must be regulated according to weed cover development throughout the first rotation cycle of E. globulus, rather than being limited to the initial establishment of the trees. Given the present and future challenges to increasing the resilience of forests under imminent climate changes, as well as the need to minimize herbicide application to comply with the norms of forest certification standards, the definition and widespread implementation of simple protocols for monitoring the effectiveness of chemical and other weed control measures would seem highly recommendable.

Author Contributions

S.C.: investigation, formal analysis, visualization, writing—original draft, funding acquisition. P.M.: formal analysis, writing—review and editing. C.G.: conceptualization, methodology, investigation, writing—review and editing. J.J.K.: formal analysis, writing—review and editing. C.V.: conceptualization, methodology, funding acquisition, project management, supervision, writing—review and editing. All authors have read and agreed to the published version of the manuscript.

Funding

Financial support was received from the FirEProd Project (PCIF/MOS/0071/2019) funded by FCT through national funds, and from the ERA Chair BESIDE project financed by the European Union’s Horizon Europe under grant agreement no. 951389. We acknowledge financial support to CESAM by FCT/MCTES (UIDP/50017/2020+UIDB/50017/2020+ LA/P/0094/2020) through national funds.

Data Availability Statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Acknowledgments

The authors wish to thank RAIZ, Instituto da Floresta e Papel and The Navigator Company, for providing the study areas and for their assistance in establishing, measuring, and maintaining the field experiments. The assistance in the early stages of this work by field staff from The Navigator Company and from RAIZ is also gratefully acknowledged.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Food and Agriculture Organization of the United Nations. Global Forest Resources Assessment 2010, Main Report. FAO Forestry Paper No. 163; Food and Agriculture Organization of the United Nations: Rome, Italy, 2009. [Google Scholar]
  2. Barreiro, S.; Tomé, M. Analysis of the Impact of the Use of Eucalyptus Biomass for Energy on Wood Availability for Eucalyptus Forest in Portugal: A Simulation Study. Ecol. Soc. 2012, 17, 14. [Google Scholar] [CrossRef]
  3. McEwan, A.; Marchi, E.; Spinelli, R.; Brink, M. Past, Present and Future of Industrial Plantation Forestry and Implication on Future Timber Harvesting Technology. J. For. Res. 2020, 31, 339–351. [Google Scholar] [CrossRef]
  4. Deng, Y.; Yang, G.; Xie, Z.; Yu, J.; Jiang, D.; Huang, Z. Effects of Different Weeding Methods on the Biomass of Vegetation and Soil Evaporation in Eucalyptus Plantations. Sustainability 2020, 12, 3669. [Google Scholar] [CrossRef]
  5. Wirabuana, P.Y.A.P.; Sadono, R.; Juniarso, S.; Idris, F. Interaction of Fertilization and Weed Control Influences on Growth, Biomass, and Carbon in Eucalyptus Hybrid (E. pellita × E. brassiana). J. Manaj. Hutan Trop. 2020, 26, 144. [Google Scholar] [CrossRef]
  6. Tomé, M.; Almeida, M.H.; Barreiro, S.; Branco, M.R.; Deus, E.; Pinto, G.; Silva, J.S.; Soares, P.; Rodríguez-Soalleiro, R. Opportunities and Challenges of Eucalyptus Plantations in Europe: The Iberian Peninsula Experience. Eur. J. For. Res. 2021, 140, 489–510. [Google Scholar] [CrossRef]
  7. ICNF. 6.o Inventário Florestal Nacional; Instituto da Conservação da Natureza e das Florestas: Lisbon, Portugal, 2019. [Google Scholar]
  8. Bontemps, J.D.; Bouriaud, O. Predictive Approaches to Forest Site Productivity: Recent Trends, Challenges and Future Perspectives. Forestry 2014, 87, 109–128. [Google Scholar] [CrossRef]
  9. Shakesby, R.A. Post-Wildfire Soil Erosion in the Mediterranean: Review and Future Research Directions. Earth-Sci. Rev. 2011, 105, 71–100. [Google Scholar] [CrossRef]
  10. Valente, S.; Coelho, C.; Ribeiro, C.; Marsh, G. Sustainable Forest Management in Portugal: Transition from Global Policies to Local Participatory Strategies. Int. For. Rev. 2015, 17, 368. [Google Scholar] [CrossRef]
  11. De Moraes Goncalves, J.L.; Alvares, C.A.; Higa, A.R.; Silva, L.D.; Alfenas, A.C.; Stahl, J.; Ferraz, S.F.d.B.; Lima, W.d.P.; Brancalion, P.H.S.; Hubner, A.; et al. Integrating Genetic and Silvicultural Strategies to Minimize Abiotic and Biotic Constraints in Brazilian Eucalypt Plantations. For. Ecol. Manag. 2013, 301, 6–27. [Google Scholar] [CrossRef]
  12. Vasić, V.; Konstantinovic, B.; Orlović, S. Weeds in Forestry and Possibilities of Their Control; IntechOpen Limited: London, UK, 2012; ISBN 978-953-51-0159-8. [Google Scholar]
  13. Carrero, O.; Stape, J.L.; Allen, L.; Arrevillaga, M.C.; Ladeira, M. Productivity Gains from Weed Control and Fertilization of Short-Rotation Eucalyptus Plantations in the Venezuelan Western Llanos. For. Ecol. Manag. 2018, 430, 566–575. [Google Scholar] [CrossRef]
  14. Adams, P.; Beadle, C.L.; Mendham, N.J.; Smethurst, P.J. The Impact of Timing and Duration of Grass Control on Growth of a Young Eucalyptus Globulus Labill. Plantation. New For. 2003, 26, 147–165. [Google Scholar] [CrossRef]
  15. Vargas, F.; Rubilar, R.; Gonzalez-Benecke, C.A.; Sanchez-Olate, M.; Aracena, P. Long-Term Response to Area of Competition Control in Eucalyptus Globulus Plantations. New For. 2018, 49, 383–398. [Google Scholar] [CrossRef]
  16. Mirra, I.M.; Oliveira, T.M.; Barros, A.M.G.; Fernandes, P.M. Fuel Dynamics Following Fire Hazard Reduction Treatments in Blue Gum (Eucalyptus globulus) Plantations in Portugal. For. Ecol. Manag. 2017, 398, 185–195. [Google Scholar] [CrossRef]
  17. Little, K.; Rolando, C.; Morris, C. An Integrated Analysis of 33 Eucalyptus Trials Linking the Onset of Competition-Induced Tree Growth Suppression with Management, Physiographic and Climatic Factors. Ann. For. Sci. 2007, 64, 585–591. [Google Scholar] [CrossRef]
  18. Garau, A.M.; Ghersa, C.M.; Lemcoff, J.H.; Barañao, J.J. Weeds in Eucalyptus globulus Subsp. Maidenii (F. Muell) Establishment: Effects of Competition on Sapling Growth and Survivorship. New For. 2009, 37, 251–264. [Google Scholar] [CrossRef]
  19. Zimdahl, R.L. The Concept and Application of the Critical Weed-Free Period; CRC Press: Boca Raton, FL, USA, 1988. [Google Scholar]
  20. Vargas, F.; Gonzalez-Benecke, C.A.; Rubilar, R.; Sanchez-Olate, M. Modelling the Effect of Weed Competition on Long-Term Volume Yield of Eucalyptus Globulus Labill. Plantations across an Environmental Gradient. Forests 2018, 9, 480. [Google Scholar] [CrossRef]
  21. Oettel, J.; Lapin, K. Linking Forest Management and Biodiversity Indicators to Strengthen Sustainable Forest Management in Europe. Ecol. Indic. 2021, 122, 107275. [Google Scholar] [CrossRef]
  22. IPMA. Área Geográfica Centro—Oeste. Available online: http://portaldoclima.pt/pt/ (accessed on 2 May 2023).
  23. IPMA. Área Geográfica Centro—Região de Coimbra. Available online: http://portaldoclima.pt/pt/ (accessed on 2 May 2023).
  24. Schad, P.; Van Huyssteen, C.; Micheli, E.; Anjos, L.; Gaistardo, C.; Deckers, J.; Dondeyne, S.; Eberhardt, E.; Gerasimova, M.; Harms, B.; et al. World Reference Base for Soil Resources 2014 International Soil Classification System for Naming Soils and Creating Legends for Soil Maps, 3rd ed.; Peter Schad Cornie van Huyssteen, E.M., Ed.; FAO: Rome, Italy, 2015; Volume World Soil; ISBN 978-92-5-108369-7. [Google Scholar]
  25. Maia, P.; Keizer, J.; Vasques, A.; Abrantes, N.; Roxo, L.; Fernandes, P.; Ferreira, A.; Moreira, F. Post-Fire Plant Diversity and Abundance in Pine and Eucalypt Stands in Portugal: Effects of Biogeography, Topography, Forest Type and Post-Fire Management. For. Ecol. Manag. 2014, 334, 154–162. [Google Scholar] [CrossRef]
  26. Maia, P.; Vasques, A.; Pausas, J.G.; Viegas, D.X.; Keizer, J.J. Fire Effects on the Seed Bank of Three Mediterranean Shrubs: Implications for Fire Management. Plant Ecol. 2016, 217, 1235–1246. [Google Scholar] [CrossRef]
  27. Tomé, M.; Tomé, J.; Ribeiro, F.; Faias, S. Equações de Volume Total, Volume Percentual e de Perfil Do Tronco Para Eucalyptus Globulus Labill. Em Portugal. Silva Lusit. 2007, 15, 25–40. [Google Scholar]
  28. Guynn, D.C.J.; Guynn, S.; Wigley, T.; Miller, D. Herbicides and Forest Biodiversity—What Do We Know and Where Do We Go from Here? Wildl. Soc. Bull. 2004, 32, 1085–1092. [Google Scholar] [CrossRef]
  29. Zhou, X.; Zhu, H.; Wen, Y.; Goodale, U.M.; Li, X.; You, Y.; Ye, D.; Liang, H. Effects of Understory Management on Trade-Offs and Synergies between Biomass Carbon Stock, Plant Diversity and Timber Production in Eucalyptus Plantations. For. Ecol. Manag. 2018, 410, 164–173. [Google Scholar] [CrossRef]
  30. Little, K.M.; Rolando, C.A. Regional Vegetation Management Standards for Commercial Eucalyptus Plantations in South Africa. South. For. J. For. Sci. 2008, 70, 87–97. [Google Scholar] [CrossRef]
  31. Inail, M.A.; Hardiyanto, E.B.; Mendham, D.S.; Thaher, E. Growth Response to Weed Control and Fertilisation in Mid-Rotation Plantations of Eucalyptus Pellita in South Sumatra, Indonesia. Forests 2021, 12, 1653. [Google Scholar] [CrossRef]
  32. McInnis, L.M.; Oswald, B.P.; Williams, H.M.; Farrish, K.W.; Unger, D.R. Growth Response of Pinus taeda L. to Herbicide, Prescribed Fire, and Fertilizer. For. Ecol. Manag. 2004, 199, 231–242. [Google Scholar] [CrossRef]
  33. FSC-STD-PRT-01-2016 V1-1 Portuguese All Scope EN; The FSC National Forest Stewardship Standard of the Portuguese Republic. FSC (Forest Stewardship Council): Lisboa, Portugal, 2016.
  34. PEFC ST 1003:2018; Sustainable Forest Management—Requirements. PEFC: Lisboa, Portugal, 2018.
Forests 14 01958 g001
Figure 1. Location of the commercial plantations of E. globulus selected for the establishment of the two study areas: Soure, in Coimbra district, and Torres Vedras, in Lisbon district.
Figure 1. Location of the commercial plantations of E. globulus selected for the establishment of the two study areas: Soure, in Coimbra district, and Torres Vedras, in Lisbon district.
Forests 14 01958 g001
Forests 14 01958 g002
Figure 2. Representative scheme of the experimental design at the study sites. Individual trees are represented by green circles, herbicide-treated plots are delimited by black rectangles, and control plots (no weed control) are delimited by red rectangles. The numbers indicate the respective plantation line.
Figure 2. Representative scheme of the experimental design at the study sites. Individual trees are represented by green circles, herbicide-treated plots are delimited by black rectangles, and control plots (no weed control) are delimited by red rectangles. The numbers indicate the respective plantation line.
Forests 14 01958 g002
Forests 14 01958 g003
Figure 3. Mean weed cover in Torres Vedras study area, evaluated as percentage of woody species cover (black bars) and herbaceous species cover (grey bars), observed at control (C) and herbicide (H) plots in the sampling years of 2010 (1.0 year after the last herbicide application), 2013 (0.1 years after the last herbicide application), 2014 (1.0 year after the last herbicide application), 2015 (2.0 years after the last herbicide application), 2016 (3.1 years after the last herbicide application), and 2017 (1.0 years after the last herbicide application).
Figure 3. Mean weed cover in Torres Vedras study area, evaluated as percentage of woody species cover (black bars) and herbaceous species cover (grey bars), observed at control (C) and herbicide (H) plots in the sampling years of 2010 (1.0 year after the last herbicide application), 2013 (0.1 years after the last herbicide application), 2014 (1.0 year after the last herbicide application), 2015 (2.0 years after the last herbicide application), 2016 (3.1 years after the last herbicide application), and 2017 (1.0 years after the last herbicide application).
Forests 14 01958 g003
Forests 14 01958 g004
Figure 4. Mean weed cover in Soure study area, evaluated as percentage of woody species cover (black bars) and herbaceous species cover (grey bars), observed at control (C) and herbicide-treated (H) plots in the sampling years of 2016 (0.2 years after the last herbicide application), 2017 (1.1 years after the last herbicide application), and 2019 (1.9 years after the last herbicide application).
Figure 4. Mean weed cover in Soure study area, evaluated as percentage of woody species cover (black bars) and herbaceous species cover (grey bars), observed at control (C) and herbicide-treated (H) plots in the sampling years of 2016 (0.2 years after the last herbicide application), 2017 (1.1 years after the last herbicide application), and 2019 (1.9 years after the last herbicide application).
Forests 14 01958 g004
Forests 14 01958 g005
Figure 5. Mean increment in H (m) of E. globulus trees at Torres Vedras study area in the control treatment (dark grey) and in the weed control treatment (light grey) between herbicide applications: from 1.5 years to 5.2 years (T0_T1), from 5.5 years to 8.2 years (T1_T2), and from 8.2 years to 8.8 years (T2_T3). Different letters indicate significant differences (p < 0.05) in height increment as detected by the two-way ANOVA test.
Figure 5. Mean increment in H (m) of E. globulus trees at Torres Vedras study area in the control treatment (dark grey) and in the weed control treatment (light grey) between herbicide applications: from 1.5 years to 5.2 years (T0_T1), from 5.5 years to 8.2 years (T1_T2), and from 8.2 years to 8.8 years (T2_T3). Different letters indicate significant differences (p < 0.05) in height increment as detected by the two-way ANOVA test.
Forests 14 01958 g005
Forests 14 01958 g006
Figure 6. Mean increment in DBH (cm) of E. globulus trees at Torres Vedras study area in the control treatment (dark grey) and in the weed control treatment (light grey) between herbicide applications: from 1.5 years to 5.5 years (T0_T1), from 5.5 years to 8.2 years (T1_T2), and from 8.2 years to 8.8 years (T2_T3). Different letters indicate significant differences (p < 0.05) in DBH increment as detected by the two-way ANOVA test.
Figure 6. Mean increment in DBH (cm) of E. globulus trees at Torres Vedras study area in the control treatment (dark grey) and in the weed control treatment (light grey) between herbicide applications: from 1.5 years to 5.5 years (T0_T1), from 5.5 years to 8.2 years (T1_T2), and from 8.2 years to 8.8 years (T2_T3). Different letters indicate significant differences (p < 0.05) in DBH increment as detected by the two-way ANOVA test.
Forests 14 01958 g006
Forests 14 01958 g007
Figure 7. Mean increment in individual volume (dm3) of E. globulus trees at Torres Vedras study area in the control treatment (dark grey) and in the weed control treatments (light grey) between herbicide applications: from 1.5 years to 5.2 years (T0-T1), from 5.5 years to 8.2 years (T1-T2), and from 8.2 years to 8.8 years (T2-T3). Different letters indicate significant differences (p < 0.05) in volume increment as detected by the two-way ANOVA test.
Figure 7. Mean increment in individual volume (dm3) of E. globulus trees at Torres Vedras study area in the control treatment (dark grey) and in the weed control treatments (light grey) between herbicide applications: from 1.5 years to 5.2 years (T0-T1), from 5.5 years to 8.2 years (T1-T2), and from 8.2 years to 8.8 years (T2-T3). Different letters indicate significant differences (p < 0.05) in volume increment as detected by the two-way ANOVA test.
Forests 14 01958 g007
Forests 14 01958 g008
Figure 8. Mean increment in height (m) of E. globulus trees at Soure study area in the control (dark grey) and in the herbicide plots (light grey) between herbicide applications: from 1.6 years to 1.9 years (T0-T1), from 1.9 years to 3.1 years (T1-T2), and from 3.1 years to 5.2 years (T2-T3). Different letters indicate significant differences (p < 0.05) in height increment as detected by the two-way ANOVA test.
Figure 8. Mean increment in height (m) of E. globulus trees at Soure study area in the control (dark grey) and in the herbicide plots (light grey) between herbicide applications: from 1.6 years to 1.9 years (T0-T1), from 1.9 years to 3.1 years (T1-T2), and from 3.1 years to 5.2 years (T2-T3). Different letters indicate significant differences (p < 0.05) in height increment as detected by the two-way ANOVA test.
Forests 14 01958 g008
Table 1. General characteristics of the two study areas, including information on the weed control treatment application: Torres Vedras (Lisbon) and Soure (Coimbra).
Table 1. General characteristics of the two study areas, including information on the weed control treatment application: Torres Vedras (Lisbon) and Soure (Coimbra).
SitePlantationMonitoring Period Weed Control Operations (n)LithologySoil Type (FAO)
[24]
Torres VedrasMarch 20082009–2017
(tree age 1.5–9 years)		3
(1.5, 5.2, and 8.2 years)		Conglomerates, sandstones, and limestonesEpileptic and endoleptic arenic Regosols
SoureMay 20142015–2019
(tree age 1.6–5.2 years)		3
(1.6, 1.9, and 3.1 years)		Sandstones and pebbleEpileptic and endoleptic arenic Regosols
Table 2. Factors affecting vegetation cover at Torres Vedras and Soure study areas, obtained by applying a generalized linear model (GLM) to the observed data.
Table 2. Factors affecting vegetation cover at Torres Vedras and Soure study areas, obtained by applying a generalized linear model (GLM) to the observed data.
EstimateStd. Errort Valuep (>|t|)
Intercept5550.911861.44882.9820.00337
A—Treatment (herbicide)6742.602235.46713.0160.00303
B—Functional type (herbaceous)8.502.94902.8810.00457
C—Years after treatment6.751.72553.9120.00014
D—Cover assessment year−1.670.8075−2.0660.04059
Interaction between A and D−3.361.1092−3.0260.00293
Table 3. E. globulus height (H), diameter at breast height (DBH), and individual tree volume (VOL) of the trees in control plots by age from the start of the experiment in 2009 (1.5 years) to its conclusion in 2017 (8.8 years) in the Torres Vedras study area.
Table 3. E. globulus height (H), diameter at breast height (DBH), and individual tree volume (VOL) of the trees in control plots by age from the start of the experiment in 2009 (1.5 years) to its conclusion in 2017 (8.8 years) in the Torres Vedras study area.
Plantation Age (Years)H (m)DBH (cm)VOL (dm3)
1.54.1 ± 0.73.5 ± 0.82.3 ± 1.3
2.35.4 ± 1.05.7 ± 1.17.6 ± 3.8
3.68.9 ± 1.48.5 ± 1.826.5 ± 13.3
4.610.6 ± 1.310.3 ± 1.944.5 ± 19.4
5.612.4 ± 1.511.5 ± 2.164.9 ± 28.4
6.113.1 ± 1.512.1 ± 2.275.6 ± 31.9
7.216.0 ± 1.713.7 ± 2.4116.7 ± 46.2
8.016.4 ± 1.714.3 ± 2.6129.6 ± 51.8
8.817.5 ± 1.914.9 ± 2.7148.8 ± 58.6
Table 4. E. globulus height (H), diameter at breast height (DBH), and individual tree volume (VOL) of the trees in control plots by year since the establishment of the experiment in 2015 (1.6 years) until its end in 2019 (5.2 years) in Soure study area (na means no available data).
Table 4. E. globulus height (H), diameter at breast height (DBH), and individual tree volume (VOL) of the trees in control plots by year since the establishment of the experiment in 2015 (1.6 years) until its end in 2019 (5.2 years) in Soure study area (na means no available data).
Plantation Age (Years)H (m)DBH (cm)VOL (dm3)
1.65.9 ± 1.1nana
2.59.8 ± 1.3nana
3.211.5 ± 1.39.0 ± 1.438.0 ± 14.0
3.813.0 ± 1.39.8 ± 1.550.4 ± 17.2
5.215.2 ± 1.412.5 ± 1.991.4 ± 30.8
Table 5. Mean annual increment in individual volume (MAI) of E. globulus trees at Torres Vedras in the control plots (dark grey) and in weeding plots treated with herbicide (light grey), determined between herbicide applications: from 1.5 years to 5.2 years (T0-T1), from 5.5 years to 8.2 years (T1-T2), and from 8.2 years to 8.8 years (T2-T3). Different letters indicate significant differences (p < 0.05) in volume increment as detected by the two-way ANOVA test.
Table 5. Mean annual increment in individual volume (MAI) of E. globulus trees at Torres Vedras in the control plots (dark grey) and in weeding plots treated with herbicide (light grey), determined between herbicide applications: from 1.5 years to 5.2 years (T0-T1), from 5.5 years to 8.2 years (T1-T2), and from 8.2 years to 8.8 years (T2-T3). Different letters indicate significant differences (p < 0.05) in volume increment as detected by the two-way ANOVA test.
Time PeriodMAI (dm3 Year−1)
ControlHerbicide
T0-T1 (1.5 to 5.2 years)13.6 ± 6.0 a16.7 ± 6.7 a
T1-T2 (5.5 to 8.2 years)25.0 ± 10.1 cd30.2 ± 10.2 bc
T2-T3 (8.2 to 8.8 years)24.0 ± 12.7 d33.9 ± 10.2 b
Table 6. Mean increment in diameter at breast height (DBH), increment in individual volume (VOL), and mean annual increment in individual volume (MAI) of E. globulus trees in the control and in the herbicide plots between herbicide applications from 3.1 years to 5.2 years (T2-T3). Different letters indicate significant differences (p < 0.05) in the measured variable increments as detected by the two-way ANOVA test.
Table 6. Mean increment in diameter at breast height (DBH), increment in individual volume (VOL), and mean annual increment in individual volume (MAI) of E. globulus trees in the control and in the herbicide plots between herbicide applications from 3.1 years to 5.2 years (T2-T3). Different letters indicate significant differences (p < 0.05) in the measured variable increments as detected by the two-way ANOVA test.
Time PeriodDBH Increment
(cm)		VOL Increment
(dm3)		MAI
(dm3 Year−1)
ControlHerbicideControlHerbicideControlHerbicide
T2-T3 (3.1 to 5.2 years)3.5 ± 0.8 a3.8 ± 0.7 b53.4 ± 18.7 a63.8 ± 14.4 b26.7 ± 9.3 a31.9 ± 7.2 b
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

Corticeiro, S.; Maia, P.; Gonçalves, C.; Keizer, J.J.; Valente, C. How Weed Control Affects Eucalyptus globulus Labill. Productivity: Results from Two Long-Term Trials. Forests 2023, 14, 1958. https://doi.org/10.3390/f14101958

AMA Style

Corticeiro S, Maia P, Gonçalves C, Keizer JJ, Valente C. How Weed Control Affects Eucalyptus globulus Labill. Productivity: Results from Two Long-Term Trials. Forests. 2023; 14(10):1958. https://doi.org/10.3390/f14101958

Chicago/Turabian Style

Corticeiro, Sofia, Paula Maia, Catarina Gonçalves, Jan Jacob Keizer, and Carlos Valente. 2023. "How Weed Control Affects Eucalyptus globulus Labill. Productivity: Results from Two Long-Term Trials" Forests 14, no. 10: 1958. https://doi.org/10.3390/f14101958

APA Style

Corticeiro, S., Maia, P., Gonçalves, C., Keizer, J. J., & Valente, C. (2023). How Weed Control Affects Eucalyptus globulus Labill. Productivity: Results from Two Long-Term Trials. Forests, 14(10), 1958. https://doi.org/10.3390/f14101958

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop