Seven-Year Results of Thinning and Burning Restoration Treatments on Old Ponderosa Pines at the Gus Pearson Natural Area

Abstract We examined the 7‐year effects of three restoration treatments on leaf physiology and insect‐resistance characteristics of pre‐settlement age ponderosa pines (Pinus ponderosa Dougl. ex Laws.) at the Gus Pearson Natural Area (GPNA) in northern Arizona. Restoration treatments were: (1) thinned in 1993 to approximate pre‐Euro‐American settlement stand structure, (2) thinned plus prescribed burned in 1994 and 1998, and (3) untreated control. Tree physiological and insect‐resistance characteristics were measured in year 2000, 7 years after thinning, using the same procedures as an earlier study performed in 1996. Consistent with the 1996 results, pre‐dawn water potential in 2000 was consistently lower in the control than both thinned treatments. Both thinned treatments continued to have increased foliar nitrogen concentration in leaves 7 years after treatment. However lower leaf nitrogen concentration in the thinned and burned compared with the thinned treatment suggests lower nitrogen availability to trees in repeatedly burned plots. Analysis of leaf gas exchange characteristics and carbon isotope content (δ¹³C) suggests continued stimulation of photosynthesis by both thinning treatments. Differences among treatments in resin volume, a measure of bark beetle resistance, depended on season of measurement. Trees in both thinning treatments continued to have increased leaf toughness, a measure of resistance to insect folivores. Our results show that many beneficial effects of restoration treatments on carbon, water, and nitrogen relations and insect‐resistance characteristics of pre‐settlement ponderosa pines continue to be expressed 7 years after treatment at the GPNA.     

Historical Stem‐Mapped Permanent Plots Increase Precision of Reconstructed Reference Data in Ponderosa Pine Forests of Northern Arizona

Forest structural reference conditions are widely used to understand how ecosystems have been altered and guide restoration and management objectives. We used six stem‐mapped permanent plots established in the early twentieth century to provide precise structural reference conditions for ponderosa pine forests of northern Arizona prior to Euro‐American settlement. Reference conditions for these plots in 1873–1874 included the following historical attributes: tree densities of 45–127 trees/ha, mean tree diameter at breast height (dbh) of 43.8 cm with a corresponding quadratic mean diameter range of 41.5–51.3 cm, and a stand basal area of 9.2–18.0 m²/ha. The reconstructed diameter distributions (for live ponderosa pine trees with dbh ≥9.14 cm) prior to fire exclusion varied in shape but generally displayed an irregular unimodal distribution. We suggest that management objectives for the structural restoration of ponderosa pine forests of northern Arizona emphasize: (1) conservation and retention of all pre‐settlement (>130 years) trees; (2) reduction of tree densities with a restoration objective ranging between 50 and 150 trees/ha having a large‐tree component between 25 and 50% of the total trees per hectare, respectively; (3) manipulation of the diameter distribution to achieve a unimodal or irregular, uneven‐aged shape (possibly targeting a balanced, uneven‐aged shape on cinder soil types) through the use of harvest and thinning practices that mimic gap disturbances (i.e., individual tree selection system); and (4) retention of 3–11 snags and logs per hectare resulting from natural mortality.     

Forest ecosystems of an Arizona Pinus ponderosa landscape: multifactor classification and implications for ecological restoration

Aim We developed an ecosystem classification within a 110,000‐ha Arizona Pinus ponderosa P. & C. Lawson (ponderosa pine) landscape to support ecological restoration of these forests. Specific objectives included identifying key environmental variables constraining ecosystem distribution and comparing plant species composition, richness and tree growth among ecosystems. Location The Coconino National Forest and the Northern Arizona University Centennial Forest, in northern Arizona, USA. Methods We sampled geomorphology, soils and vegetation on 66 0.05‐ha plots in open stands containing trees of pre‐settlement (c. 1875) origin, and on 26 plots in dense post‐settlement stands. Using cluster analysis and ordination of vegetation and environment matrices, we classified plots into ecosystem types internally similar in environmental and vegetational characteristics. Results We identified 10 ecosystem types, ranging from dry, black cinders/Phacelia ecosystems to moist aspen/Lathyrus ecosystems. Texture, organic carbon and other soil properties reflecting the effects of parent materials structured ecosystem distribution across the landscape, and geomorphology was locally important. Plant species composition was ecosystem‐specific, with C₃Festuca arizonica Vasey (Arizona fescue), for instance, abundant in mesic basalt/Festuca ecosystems. Mean P. ponderosa diameter increments ranged from 2.3–4.3 mm year^?1 across ecosystems in stands of pre‐settlement origin, and the ecosystem classification was robust in dense post‐settlement stands. Main conclusions Several lines of evidence suggest that although species composition may have been altered since settlement, the same basic ecosystems occurred on this landscape in pre‐settlement forests, providing reference information for ecological restoration. Red cinders/Bahia ecosystems were rare historically and > 30% of their area has been burned by crown fires since 1950, indicating that priority could be given to restoring this ecosystem's remaining mapping units. Ecosystem classifications may be useful as data layers in gap analyses to identify restoration and conservation priorities. Ecosystem turnover occurs at broad extents on this landscape, and restoration must accordingly operate across large areas to encompass ecosystem diversity. By incorporating factors driving ecosystem composition, this ecosystem classification represents a framework for estimating spatial variation in ecological properties, such as species diversity, relevant to ecological restoration.     

Effects of Restoration Thinning on Presettlement Pinus ponderosa in Northern Arizona

In the 100 years following the arrival of Euro-American settlers in northern Arizona, Pinus ponderosa (ponderosa pine) forests changed from open, low-density stands to closed, high-density stands. The increase in tree density has been detrimental to the vigor of old-growth trees that established before settlement (presettlement trees). In this study, we examined whether the vigor of presettlement trees could be improved by restoring the original stand structure by thinning the ponderosa pines that established after settlement (postsettlement trees). The restoration treatment caused the following changes in the presettlement trees and their environment in the first year following thinning: an increase in volumetric soil water content between May and August, an increase in predawn xylem water potential in July and August, a decrease in midday xylem water potential in June and August, an increase in net photosynthetic rate in August, an increase in foliar nitrogen concentration in July and August, and an increase in bud and needle size. The results show that the thinning restoration treatment improved the condition of presettlement ponderosa pines by increasing canopy growth and the uptake of water, nitrogen, and carbon.     

Butterfly Response to Microclimatic Conditions Following Ponderosa Pine Restoration

Efforts to restore ponderosa pine ecosystems to open, park‐like conditions that predominated prior to European‐American settlement result in altered stand structure and increased landscape heterogeneity, potentially altering habitat suitability for invertebrates and other forest organisms. We examined the responses of two butterfly species, Colias eurytheme and Neophasia menapia, to microclimatic changes at structural edges created by experimental restoration treatments in northern Arizona. We monitored microclimate, including air temperature, light intensity, and vapor pressure deficit (VPD), on several mornings during butterfly releases. We placed adult butterflies at east‐ and west‐facing edges approximately one half‐hour before dawn to determine their behavioral response to microclimatic differences between east‐ and west‐facing edges. After sunrise, all three microclimatic variables were higher at east‐facing edges, and the difference in microclimate between the two edge orientations increased through early morning. For both species, butterflies placed at east‐facing edges flew earlier than butterflies at west‐facing edges. Colias eurytheme, an open‐habitat species, tended to move toward the treated forest during initial flight, while movements of Neophasia menapia, a forest‐dwelling species, did not differ from random flight. Our results indicate that butterflies respond to microclimatic factors associated with restoration treatments, while responses to structural changes in habitat vary among species, based on habitat and food plant preferences. These changes in forest structure and microclimate may affect the distribution of many mobile invertebrates in forested landscapes undergoing restoration treatments.     

Preliminary trial of woody debris addition on the return of invertebrates to restored bauxite mines in the jarrah forest of Western Australia

Woody debris is an important component of forest ecosystems, but its use in mine site restoration has been limited and it can be slow to build up naturally. A new technique of spreading snipped wood waste onto restored mine pits prior to seeding has been subjected to a preliminary trial at Alcoa's Huntly mine site, in the northern jarrah forest of south‐western Western Australia. We examined whether the application of snipped wood during restoration encourages the return of ground‐ and litter‐dwelling invertebrates without negatively suppressing plant establishment. Invertebrates were sampled across three seasons from experimental plots treated with 0 t/ha (control), 100 t/ha or 300 t/ha snipped wood waste. Invertebrate communities in treatment plots comprised higher numbers and diversity of wood and litter decomposers such as mites, Diplopoda, Dermaptera and Blattodea than control plots. Plant responses were variable, with wood treatment resulting in lower tree and overall plant density but having no effects on plant species richness or plant cover. Wood treatment plots were associated with higher soil nitrogen than controls. We hypothesise that the use of a fine wood treatment at the lowest rate of 100 t/ha (approximately 30% wood cover) is likely to enhance the diversity and abundance of invertebrates in restored areas, with minimal effect on plant establishment. Encouraging a diverse invertebrate fauna to recolonise restoration should help speed up succession and ecosystem functions such as decomposition and nutrient cycling, and more quickly return the land to previous ecosystem values.     

Slash Pile Burning Effects on Soil Biotic and Chemical Properties and Plant Establishment: Recommendations for Amelioration

Ponderosa pine forest restoration consists of thinning trees and reintroducing prescribed fire to reduce unnaturally high tree densities and fuel loads to restore ecosystem structure and function. A current issue in ponderosa pine restoration is what to do with the large quantity of slash that is created from thinning dense forest stands. Slash piling burning is currently the preferred method of slash removal because it allows land managers to burn large quantities of slash in a more controlled environment in comparison with broadcast burning slash. However burning slash piles is known to have adverse effects such as soil sterilization and exotic species establishment. This study investigated the effects of slash pile burning on soil biotic and chemical variables and early herbaceous succession on burned slash pile areas. Slash piles were created following tree thinning in two adjacent approximately 20‐ha ponderosa pine (Pinus ponderosa) restoration treatments in the Coconino National Forest near Flagstaff, Arizona. We selected 30 burned slash pile areas and sampled across a gradient of the burned piles for arbuscular mycorrhizal (AM) propagule densities, the soil seed bank, and soil chemical properties. In addition, we established five 1‐m² plots in each burned pile to quantify the effect of living soil (AM inoculum) and seeding amendments on early herbaceous succession in burned slash pile areas. The five treatments consisted of a control (no treatment), living soil (AM inoculum) amendment, sterilized soil (no AM inoculum) amendment, seed amendment, and a seed/soil (AM inoculum) amendment. Slash pile burning nearly eliminated populations of viable seeds and AM propagules and altered soil chemical properties. Amending scars with native seeds increased the cover of native forbs and grasses. Furthermore adding both seed and living soil more than doubled total native plant cover and decreased ruderal and exotic plant cover. These results indicate that seed/soil amendments that increase native forbs and grasses may enhance the rate of succession in burned slash pile areas by allowing these species to outcompete exotic and ruderal species also establishing at the site through natural regeneration.     

Modeling Ecological Restoration Effects on Ponderosa Pine Forest Structure

FIRESUM, an ecological process model incorporating surface fire disturbance, was modified for use in southwestern ponderosa pine ecosystems. The model was used to determine changes in forest structure over time and then applied to simulate changes in aboveground biomass and nitrogen storage since exclusion of the natural frequent fire regime in an unharvested Arizona forest. Dendroecological reconstruction of forest structure in 1876, prior to Euro‐American settlement, was used to initialize the model; projections were validated with forest measurements in 1992. Biomass allocations shifted from herbaceous plants to trees, and nitrogen was increasingly retained in living and dead tree biomass over the 116‐year period (1876–1992). Forest conditions in 1992 were substantially degraded compared to reference presettlement conditions: old‐growth trees were dying at accelerated rates, herbaceous production was reduced nearly 90%, and the entire stand was highly susceptible to high‐intensity wildfire. Following an experiment initiated in 1993 to test ecological restoration treatments, future changes were modeled for the next century. Future forest structure remained within the natural presettlement range of variability under the full restoration treatment, in which forest biomass structure was thinned to emulate presettlement conditions and repeated low‐intensity fire was reintroduced. Simulation of the control treatment indicated continuation of exceptionally high tree density, probably culminating in stand‐replacing ecosystem change through high‐intensity wildfire or tree mortality from pathogens. Intermediate results were observed in the partial restoration treatment (tree thinning only); the open forest structure and high herbaceous productivity found immediately after treatment were gradually degraded as dense tree cover reestablished in the absence of fire. Modeling results support comprehensive restorative management as a long‐term approach to conservation of key indigenous ecosystem characteristics.     

Current Loads of Coarse Woody Debris on Southeastern Australian Floodplains: Evaluation of Change and Implications for Restoration

We evaluated the status of coarse woody debris (CWD, fallen wood) on floodplains of the southern Murray‐Darling basin of southeastern Australia. The floodplains are dominated floristically by the river red gum Eucalyptus camaldulensis. Aerial survey techniques were used to estimate the amounts of woody debris within 200 m of the channels along 2,442 km of 11 rivers of the system, including the Murray and Darling Rivers and the Darling Anabranch. Aerially based indices were converted into wood volumes by using ground‐truthing at a selection of sites; there was a strong correlation between index values and measured wood volume densities. For thickly forested sites such as Barmah, Gunbower Island, and the Ovens floodplains, the aerial method was not useful, so ground measurements at randomly positioned sites within the forests were used. Volumes were translated into mass by using conversion factors drawn from the literature. We estimated that total tonnage on approximately 221,000 ha of floodplain forests was 4.175 ± 0.579 × 10⁶ tonne. In the larger forested blocks (>7,000 ha), mean wood densities ranged between approximately 12 tonne/ha on the lower Goulburn up to approximately 24 tonne/ha at Barmah State Forest. The area‐weighted mean for the entire area was approximately 19 tonne/ha. A main purpose of the research was to place these figures into an historical perspective to evaluate implications for restoration. A thorough search of historical documentation revealed that there are no extant data upon which to estimate pre‐European settlement levels. We used information from an apparently undisturbed “unmanaged” site in the Millewa forests of southern New South Wales as a basis. Wood density there corresponded to a mean figure of 125 tonne/ha wood‐mass density. By using this figure we estimate that CWD levels on the southern Murray‐Darling basin may be of the order of 15% of pre‐European settlement levels. Full restoration of the 221,000 ha surveyed would require 23.5 ± 0.579 × 10⁶ tonne, which is equivalent to about 600,000 mature (1 m diameter at breast height) river red gum trees or the amount of timber derived from clear felling about 115,000 ha of river red gum forest at current stocking levels. We discuss the implications of this massive deficit and possible short‐ and long‐term solutions.     

Water and Nutrient Outflow Following the Ecological Restoration of a Ponderosa Pine-Bunchgrass Ecosystem

In the late 1800s, fire suppression, livestock grazing, and a wet and warm climate led to an irruption of pine regeneration in Pinus ponderosa Laws. (ponderosa pine) forests of the southwestern United States. Pines invaded bunchgrass openings, causing stand structure changes that increased the number of stand-replacing fires. Ecological restoration, via thinning and prescribed burning, is being used to decrease the risk of stand-replacing fires and ameliorate other effects of pine invasion. The effects of aboveground restoration on belowground processes are poorly understood. We used a hydrologic model and soil water nutrient concentrations, measured monthly below the rooting zone, to estimate restoration effects on nutrient losses by leaching from a mature ponderosa pine forest near Flagstaff, Arizona. Replicated restoration treatments included thinning to pre-1880 stand densities (partial restoration), thinning plus forest floor fuel reduction followed by a prescribed burn (complete restoration), and an untreated control. Water outflow occurred only between January and May and was lowest from the control (47 and 28 mm in 1995 and 1996) and highest from the partial restoration treatment (67 and 59 mm in 1995 and 1996). The concentrations (typically <0.10 mg/ L) and estimated annual losses (<0.02 kg/ha) of NH₄⁺-N, PO₄^{3 ?} -P, and organic P were similar among treatments. Nitrate and organic N concentrations were as high as 0.80 mg N/L; however, these concentrations and estimated annual losses (<0.13 kg N/ha) were similar among treatments. Our results suggest that restoration will not enhance nutrient loss by leaching or alter stream chemistry in ponderosa pine forests.     

Microclimatic Changes Induced by Ecological Restoration of Ponderosa Pine Forests in Northern Arizona

Restoration of ponderosa pine ecosystems results in altered stand structure, potentially affecting microclimatic conditions and habitat quality for forest organisms. This research focuses on microclimatic changes resulting from forest and landscape structural alterations caused by restoration treatments in southwestern ponderosa pine forests. Three microclimate variables—light intensity, air temperature, and vapor pressure deficit (VPD)—were monitored over two field seasons. Differences in microclimate between the treated forest and the surrounding untreated forest were measured, and microclimatic gradients across the structural edge between these two forest types were quantified. Restoration treatments increased sunlight penetration to the forest floor but did not significantly impact ambient air temperature or VPD. Mean values for air temperature and VPD did not differ significantly between treatments, although temperature and vapor pressure deficit did exhibit a trend in the morning; both variables were higher at the structural edge and in the treated forest during morning hours. Significant edge gradients were detected for air temperature and VPD in the morning and evening, increasing from the structural edge into the untreated forest. Our results show that microclimatic effects of these restoration treatments are generally modest, but the changes are more prominent at specific locations and during certain times of day. Because even modest changes in microclimate have the potential to impact a range of key ecological processes, microclimatic effects should be considered when forest restoration treatments at the landscape scale are being planned and implemented.     

Assessing Targets for the Restoration of Herbaceous Vegetation in Ponderosa Pine Forests

A restoration project is considered a success when the initial target is met, but many targets are plausible. We evaluated the success of a restoration project in its 11th year since treatment in a southwestern ponderosa pine–bunchgrass community and the appropriateness of several targets. We measured the responses of (1) total standing crop; (2) standing crop of five functional groups (C₃ and C₄ graminoids, leguminous forbs, and nonleguminous perennial and annual forbs); (3) graminoid community composition; and (4) standing crop of five common graminoid species (Festuca arizonica, Muhlenbergia montana, Elymus elymoides, Carex geophila, and Poa fendleriana). Targets were quantified in remnant grass patches, which provided the standards for these targets, and were assessed in three other forest patch types (pre‐settlement tree patches, post‐settlement tree patches, and patches where all post‐settlement trees were removed). Patches where all post‐settlement trees were removed reached target levels for total standing crop, C₃ and C₄ graminoid standing crop, graminoid community composition, and M. montana, E. elymoides, and C. geophila standing crops. Standing crop of legumes and of F. arizonica did not increase over time in any patch type. Targets were not met in pre‐settlement patches or in patches where some post‐settlement trees were left standing, suggesting that it is unrealistic to expect equal responses across all patch types. If increasing herbaceous standing crop is a major goal, practitioners should create gaps within the pine forest canopy.     

Fire disturbance and forest structure in an old‐growth Pinus ponderosa forest,southern Cascades,USA

Questions: Did fire regimes in old‐growth Pinus ponderosa forest change with Euro‐American settlement compared to the pre‐settlement period? Do tree age structures exhibit a pattern of continuous regeneration or is regeneration episodic and related to fire disturbance or fire‐free periods? Are the forests compositionally stable? Do trees have a clumped spatial pattern and are clumps even‐ or mixed‐age? How might information from this old‐growth forest inform current restoration and management practices? Location: A 235‐ha old‐growth forest in the Ishi Wilderness, southern Cascade Mountains, California. Methods: Age, size, and spatial pattern of trees were quantified in seven stands. Fire history was reconstructed using fire scar dendrochronology. The influence of fire on stand structure was assessed by comparing fire history with age, size, and spatial structure of trees and identifying and measuring trees killed by two recent fires. Results: Species composition in plots was similar but density and basal area of tree populations varied. Age structure for P. ponderosa and Quercus kelloggii showed periods of episodic recruitment that varied among plots. Fire disturbance was frequent before 1905, with a median period between fires of 12 years. Fire frequency declined after 1905 but two recent fires (1990, 1994) killed 36% and 41% of mostly smaller diameter P. ponderosa and Q. kelloggii. Clusters of similar age trees occurred at scales of 28‐1018 m² but patches were not even‐aged. Interactions between tree regeneration and fire promoted development of uneven age groups of trees. Conclusions: Fire disturbance strongly influenced density, basal area, and spatial structure of tree populations. Fire exclusion over the last 100 years has caused compositional and structural changes. Two recent fires, however, thinned stands and created gaps favorable for Q. kelloggii and P. ponderosa regeneration. The effects of infrequent 20th century fire indicate that a low fire frequency can restore and sustain structural characteristics resembling those of the pre‐fire suppression period forest.     

Restoration of Old Forest Features in Coast Redwood Forests Using Early‐stage Variable‐density Thinning

To accelerate development of old forest features in coast redwood, two thinning treatments and an unthinned control were compared in three treatment areas in north coastal California. One thinning treatment was designed to restore old forest densities of 125 trees/ha and the other 250 trees/ha representing a one‐step and partial treatments to the desired stand density. Four years after treatment, numbers of trees had increased in the thinning treatments due to recruitment of new trees, but had decreased in the control due to self‐thinning. Residual trees increased in stem volume following thinning by 128% in low‐density thinning compared to 70% in the controls indicating thinning accelerated stand development. The thinning treatments also moved the species composition of these stands to a greater proportion of redwood. Considerable slash was produced by the thinning treatments but was decomposing rapidly. Black bears damaged approximately 15% of all trees and more than 38% of residual trees in the thinned treatments compared to less than 2% of all trees in the control. This damage included killing some trees and damaging other trees that survived. Decisions over restoration densities in these stands are complicated by prolonged stand development, and balancing risks and costs. In this case, the bears represent a stochastic factor that dramatically increases risk. Thinning appears to be an effective means of enhancing old forest development by accelerating tree growth, modifying species composition, and increasing stand‐level variability. Continued monitoring will be necessary to evaluate long‐term trends in density relative to effects of bear damage.     

Demographic costs and benefits of natural regeneration during tropical forest restoration

For tropical forest restoration to result in long‐term biodiversity gains, native trees must establish self‐sustaining populations in degraded sites. While many have asked how seedling recruitment varies between restoration treatments, the long‐term fate of these recruits remains unknown. We address this research gap by tracking natural recruits of 27 species during the first 7 years of a tropical forest restoration experiment that included both planted and naturally regenerating plots. We used an individual‐based model to estimate the probability that a seedling achieves reproductive maturity after several years of growth and survival. We found an advantage for recruits in naturally regenerating plots, with up to 40% increased probability of reproduction in this treatment, relative to planted plots. The demographic advantage of natural regeneration was highest for mid‐successional species, with relatively minor differences between treatments for early‐successional species. Our research demonstrates the consequences of restoration decision making across the life cycle of tropical tree species.     

Restoration of Young Forests in Eastern Finland: Benefits for Saproxylic Beetles (Coleoptera)

Effective fire suppression in combination with intensive forestry has caused a large number of dead wood‐dependent (saproxylic) species to become threatened in Fennoscandia. In order to return the fire disturbance dynamics and to increase the amount of dead wood, restoration actions are urgently needed. We studied the effects of restoring young (under 30 years old) pine‐dominated (Pinus sylvestris L.) forest stands on saproxylic beetle assemblages in eastern Finland, focusing especially on rare, red‐listed, and pyrophilous (RRLP) species. Our experiment included a restoration treatment including two tree felling levels for fuel load (10 or 20 m³/ha) followed by burning, and an untreated control. We sampled beetles before restoration in 2005, during the year of restoration in 2006, and in two post‐treatment years in 2007 and 2011. Both restoration treatments increased the number of saproxylic and RRLP species. The species richness increased most in the year of restoration in 2006 and this trend continued in the following year 2007, but no differences in species assemblages were detected between the two fuel load levels. By 2011, however, the species richness and abundance had declined back to the pre‐treatment level. We suggest that restoration burning can also be directed to young forests where biodiversity values are initially low. On the basis of the observed decline in the species richness, we suggest that fire could be introduced in neighboring areas in approximately 5‐year intervals to maintain populations of the most demanding pyrophilous species .     

Tropical forest restoration in Costa Rica: the effect of several strategies on litter production, accumulation and decomposition

Tropical forest restoration strategies have the potential to accelerate the recovery of the nutrient cycles in degraded lands. Litter production and its decomposition represent the main transfer of organic material and nutrients into the soil substrate. We evaluated litter production, accumulation on the forest floor, and its decomposition under three restoration strategies: plantation (entire area planted with trees), island (trees planted in patches of three different sizes) and control (natural regeneration) plots. We also compared restoration strategies to young secondary forest (7-9 yr). Restoration treatments were established in 50 x 50m plots in June 2004 at six sites in Southern Costa Rica. Planted tree species included two native timber species (Terminalia amazonia and Vochysia guatemalensis) interplanted with two N fixers (Erythrina poeppigiana and Inga edulis). Litter was collected every 15 days between September 2008 and August 2009 in 12 0.25m2 litter traps distributed within each plot; litter that accumulated on the soil surface was collected at four locations (0.25m2 quadrats) within each plot in February and May 2009. Total litter production in plantation (6.3Mg/ha) and secondary forest (7.3Mg/ha) did not differ, but were greater than in islands (3.5Mg/ha) and control (1.4 Mg/ha). Plantation had greatest accumulation of litter on the soil surface (10.6 Mg/ha) as compared to the other treatments (SF = 7.2; I = 6.7; C = 4.9). Secondary forest was the only treatment with a greater annual production of litter than litter accumulation on the soil surface. Carbon storage in litter was similar between plantation and secondary forest, and significantly greater than the other treatments. No differences were found for carbon concentration and storage in the soil among treatments. There was also high variability in the production and accumulation of litter and carbon among sites. Active restoration treatments accelerated the production of litter and carbon storage in comparison to areas under natural recovery. However, the nutrient cycle has not necessarily been restored under these conditions, as high litter accumulation on the soil surface indicates a low decomposition rate, which slows nutrient return to the soil.