Competing Vegetation Effects on Soil Carbon and

0 downloads 0 Views 675KB Size Report
Jun 20, 2013 - treatment received 5 yr of intensive vegetation control (her- bicide) from ... glycine, isopropylamine salt) broadcast with backpack sprayers before planting ... interval, all samples in that core were discarded and a new core.
Published August 8, 2014

North American Forest Soils Conference Proceedings

Competing Vegetation Effects on Soil Carbon and Nitrogen in a Douglas-fir Plantation Erika Knight* Paul Footen Robert Harrison

School of Environmental and Forest Sciences College of the Environment Univ. of Washington Box 352100 Seattle, WA 98195-2100

Thomas Terry

Sustainable Solutions 5935 Swayne Rd. NE Olympia, WA 98516

Scott Holub

Weyerhaeuser NR Company PO Box 275 Springfield, OR 97477

Application of herbicides to control competing vegetation and improve crop tree growth is a common silvicultural practice. Vegetation control has the potential to change pools of soil C and N and thus affect soil quality and C sequestration. In this study, the effects of vegetation control (primarily for herbaceous vegetation) on soil C and N were compared for a bole-only harvest with and without 5 yr of annual herbicide application (+VC and −VC, respectively). Soil C and N were measured in six depth increments (forest floor and 0–15, 15–30, 30–45, 45–60, and 60–100 cm) in a 12-yr-old Douglas-fir [Pseudotsuga menziesii (Mirb.) Franco] plantation at the Fall River Long-Term Soil Productivity site in western Washington. Deep-soil (60– 100-cm) C concentration was significantly higher (a = 0.10) with vegetation control (14.7 g kg−1 for +VC, 10.4 g kg−1 for −VC). Nitrogen concentration was significantly higher in the forest floor treatment without vegetation control (11.2 g kg−1 N for +VC, 12.7 g kg−1 N for −VC); however, the N content of the −VC 0- to 15-cm mineral soil was significantly lower than the +VC (2920 kg N ha−1 for +VC, 2720 kg N ha−1 for −VC). The root concentration (kg roots kg soil−1) was higher in the +VC treatment at both the 30- to 45and 45- to 60-cm depth intervals. Despite these differences, there were no significant differences in total C or N content to 100 cm with and without vegetation control. The longer term impact of the greater root concentration at 30 to 60 cm on soil C and N pools needs to be assessed. Abbreviations: LTSP, Long-Term Soil Productivity.

P

acific Northwest forest ecosystems have the potential to sustain some of the largest annual productivities and total pools of forest biomass in the world (Smithwick et al., 2002). These forests contain among the highest combined aboveground tree biomass and belowground organic matter of any forests in the contiguous United States (Kern, 1994; Smithwick et al., 2002). Soils can potentially store C for long periods of time, and management for increased soil C content has been proposed as a method to help mitigate global climate change (Lorenz et al., 2011; Peng et al., 2008). Soil C accounts for much of total ecosystem C in forests in the region. At the Fall River Long-Term Soil Productivity site in southwestern Washington, for instance, the mineral soil, which was sampled to a depth of 80 cm, accounted for 43% of the total C pool on the site including soil, coarse woody debris, the forest stand, and aboveground understory vegetation (Ares et al., 2007). The potential for large amounts of C sequestration to occur in Pacific Northwest forests, as well as the potential for loss, highlights the need to understand how C stocks are altered by forest management. This is especially important This work was presented at the 12th North America Forest Soils Conference, Whitefish, MT, 16–20 June 2013, in the Production Systems for Biomass and Bioenergy session. Soil Sci. Soc. Am. J. doi:10.2136/sssaj2013.07.0320nafsc Received 31 July 2013. *Corresponding author ([email protected]). © Soil Science Society of America, 5585 Guilford Rd., Madison WI 53711 USA All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Soil Science Society of America Journal

where timber harvest is part of sustainable management—management to maintain and enhance soil organic C is requisite for sustaining soil productivity. Management of forest residuals after harvest as well as management of the regenerating forest stand can affect post-harvest soil C stores. Within the first few years post-harvest, there is often a spike in soil C content as roots and harvest slash decompose into the forest floor and mix with the soil, but soil C levels often then begin to decline until new vegetation becomes established and begins contributing substantial new root growth, litter, and dissolved and particulate organic matter into the soil ( Johnson and Curtis, 2001; Sanchez et al., 2006). The litter production and root turnover of regenerating vegetation is a significant contributor to the re-accumulation of soil C, and intensive management that increases tree growth can, in turn, increase soil C through the production of coarse roots (Fox, 2000). The use of herbicides to control vegetative growth that competes with crop trees for moisture, nutrients, and sunlight is a common forest harvest practice used to increase crop tree survival and growth. However, the effects of vegetation control on soil C are unclear and may vary with soil type and climate. The results of a study in the U.S. Southeast indicated that soil C to 100 cm decreased with the use of vegetation control treatments for 17 yr (Shan et al., 2001). Slesak et al. (2011b) suggested that vegetation control may reduce the amount of recent organic matter inputs to the soil and that soil microbes may consume more preexisting soil organic matter as a result. However, this overall C loss might not be the case if increased tree growth results in production of less easily decomposed organic matter that balances losses from the decomposition of organic matter in the soil. In addition to affecting soil C, competing vegetation control can also change N cycling, with the potential to reduce the soil N content (Shan et al., 2001). The uptake of N by vegetative growth can retain N near the soil surface and in plant biomass (Marks and Bormann, 1972; Vitousek et al., 1979). Control of competing vegetation has the potential to reduce N uptake on a site, and studies by Smethurst and Nambiar (1995) and Vitousek and Matson (1985) have observed increases in NO3 leaching when competing vegetation is controlled. However, a study by Strahm and Harrison (2006) indicated that there is a substantial ability in some soils, particularly soils of volcanic origin or volcanic ash such as occur in the Pacific Northwest, to adsorb NO3. The accumulation of C and N in soil is a result of the balance between inputs of organic matter from litterfall and root turnover and losses from decomposition and leaching. Rates of decomposition, leaching, and adsorption vary with both soil type and environmental factors, and there is high variability in the response of soil C and N to forest harvest removals and regeneration practices. This study investigated the effect of using herbicides to control competing vegetation on soil C and N, as well as forest productivity, in a Douglas-fir plantation in the Pacific Northwest. ∆

MATERIALS AND METHODS Site Description

Research was conducted at the Fall River Long-Term Soil Productivity study, which is an affiliate study of the Long-Term Soil Productivity (LTSP) network (Powers, 2006). The site is located in western Washington (46°44¢ N, 123°24¢ W) at an elevation between 300 and 375 m and has a maritime climate. Slopes are gentle with a 10 to 15% grade (Ares et al., 2007). The soil at Fall River is of the Boistfort series (medial over clayey, ferrihydritic over parasesquic, mesic Typic Fulvudands), which is a very deep and well-drained soil with few rocks that developed from weathered basalt. Soil texture ranges from silt loam in the A horizon to silty clay below 53 cm (Soil Survey Staff, 1999). Study installation began in 1999 with harvesting of 12 treatment plots replicated in four complete blocks. Treatment plots were 30 by 85 m and had an internal measurement plot of 15 by 70 m (0.105 ha). Preharvest forest composition and slope position determined blocking. Soil sampling before plot installation (1998) revealed that mineral soil C and N contents to 80 cm were 249 and 13.14 Mg ha−1, respectively (Ares et al., 2007), which is higher than average for soils derived from igneous parent materials in the Pacific Northwest (Littke et al., 2011). The 12 treatments installed on this LTSP site included ground-based harvesting (soil compaction), vegetation control, and organic matter (biomass) removal intensity treatments. In this study, two of these treatments were selected for study of the effects of vegetation control on soil C and N: the bole-only harvest with competing vegetation control (+VC) treatment and bole-only harvest without competing vegetation control (−VC). In the bole-only harvest, tree boles to an 8- to 13-cm top were cable-yarded, and all remnant coarse woody debris was distributed evenly across the plot area. The plots were planted with Douglas-fir seedlings in 2000 at a spacing of 2.5 by 2.5 m (1600 trees ha−1). All plots specified for the vegetation control treatment received 5 yr of intensive vegetation control (herbicide) from 2000 to 2004, applied through a combination of backpack sprayers and spot application. This vegetation control treatment was not intended to simulate standard operational vegetation control but rather to completely eliminate competing vegetation, and specific herbicides varied by year to achieve the desired level of vegetation control (Ares et al., 2007). Herbicide types, concentrations, and application methods were reported in detail by Ares et al. (2007) and included: sulfometuron methyl (2-[[[[(4,6-dimethyl-2-pyrimidinyl)amino]carbonyl]amino] sulfonyl]benzoic acid) and glyphosate (N-(phosphonomethyl) glycine, isopropylamine salt) broadcast with backpack sprayers before planting seedlings in 2000; atrazine [6-chloro-N-ethylN¢-(1-methylethyl)-1,3,5-triazine-2,4-diamine] and glyphosate applied in spring 2001; atrazine, sulfometuron methyl, clopyralid (3,6-dichloro-2-pyridinecarboxylic acid, monoethanolamine salt), and glyphosate applied in spring 2002; hexazinone [3-cyclohexyl-6-(dimethylamino)-1-methyl-1,3,5-triazine2,4(1H,3H)-dione], clopyralid, and glyphosate applied in spring 2003; and hexazinone applied in spring 2004. All clopyralid Soil Science Society of America Journal

and glyphosate applications after planting were directed spot applications. Five years after planting seedlings, the total cover of competing vegetation (forbs, grasses, vines, shrubs, and nonplanted trees) was 74 and 4% in the −VC and +VC treatments, respectively (Devine et al., 2011). At the time of soil sampling for this study, the trees on the site were 12 yr old and approaching canopy closure. A more detailed description of the Fall River study installation was provided by Ares et al. (2007).

Plot Sampling Forest floor and mineral soil samples were collected from February to May 2012 from +VC and −VC treatment plots in each replicate block (a total of eight plots). Within each plot, six 1.0-m2 subplots were randomly selected for sampling. When surface obstacles such as planted trees, stumps, or decomposing wood obstructed sampling at one of the randomly selected subplots, a new random subplot location was chosen. Forest floor and mineral soil samples were collected at each subplot.

Forest Floor Sample Collection Within each subplot, forest floor samples were collected at three locations from within a 0.05-m2 area and composited into one sample. Forest floor material included all identifiable plant detritus (rotten wood, needles, twigs, etc.) 50% decomposing wood by volume in any depth interval, all samples in that core were discarded and a new core was collected.

Sample Processing and Analysis All samples (tissue and mineral soil) were oven dried at 60°C. The bulk density was calculated based on dry weight and sample volume. No differences in moisture content were observed when the drying temperature was increased to 100°C, so www.soils.org/publications/sssaj

the oven-dry weight at 60°C was used for the bulk density value. Dry mineral soil samples were sieved with a 4.75-mm sieve. Although the fraction of material 4.75 mm were removed and weighed. The root concentration was calculated as the weight of these >4.75-mm roots divided by the weight of the