A note on the root development in dry tropical naturally revegetated coalmine spoils

Root biomass distribution on an age series of naturally revegetated coalmine spoils in a dry tropical environment is described. 65 to 85% of the total root biomass was observed in the upper 0–15 cm of spoil depth. Highest (724 g m^-2) and lowest (186 g m^-2) values of root biomass were found on 12-yr old protected flat surfaces and 5-yr old grazed undulating sites, respectively. There is an increase in root biomass with increase in age of coalmine spoil.     

Living and dead belowground biomass and its distribution in some savanna communities in Cuba

Fiala K. etHerrera R. (1988): Living and dead belowground biomass and its distribution in some savanna communities in Cuba.—Folia Geobot. Phytotax., Praha, 23: 225–237.— The paper sums up the first results obtained from the study of belowground biomass estimated in natural and anthropic savanna communities in different regions of Cuba at the end of the 1984 rainy season. The percentage of living roots in total root biomass of natural savannas was lower (34–50%) than that in the anthropic savanna stands (68–74%). The total belowground biomass in three savanna stands ranged from 1,073 to 1,257 g. m^?2. In the natural savanna stands 433 to 517 g. m^?2 of living belowground biomass was found, which was less than in the anthropic savanna stand (745 g. m^?2). In all the savanna stands studied, more than 80% of both the total and living belowground biomass were found in the upper 0–0.2 m soil layer. The share of the living biomass in the belowground plant organs varied from 71 to 79%.     

Spatial heterogeneity in the herbaceous layer of a semi-arid savanna ecosystem

Despite increasing recognition of the role spatial pattern can play in ecosystem function, few studies have quantified spatial heterogeneity in savanna ecosystems. The spatial distribution of herbaceous biomass and species composition was measured across three scales in a semi-arid savanna in central Kenya, and patterns were related to environmental variables at different scales. Herbaceous biomass declined across a rainfall gradient and from upper to lower topographic positions, but variation within a site (across 5–50 m) was similar in magnitude to among-site variation associated with rainfall and topography. Geostatistical analyses showed that patchiness at scales of 5–25 m explained 20% of total variation in herbaceous biomass. This pattern arose from the presence of both 5–10-m diameter patches containing high herbaceous biomass (> 170 g m^–2) and 5–10-m diameter patches characterized by nearly bare soil surfaces (< 40 g m^–2). Patch structure was contingent on topography, with larger bare patches at ridgeline and upper hillslope positions. Grass species distributions showed the greatest degree of patch structure and species turnover across distances of 5–45 m. Additional community variation was associated with topography, with minimal variation in species composition across the rainfall gradient. Pattern diversity significantly exceeded levels reported for four other grassland ecosystems, suggesting fundamental differences in local processes generating spatial pattern. It is hypothesized that heterogeneously distributed grazing pressure, interacting with the distribution of shrub canopies, is an important factor generating such high levels of small-scale patch structure in this savanna.     

Temporal changes in spatial pattern of fine-root mass and nutrient concentrations in Indian bamboo savanna

Abstract. Temporal variations in the spatial distribution of fine-root mass and nutrient concentrations were studied in recently harvested and mature bamboo savanna sites in the dry tropical Vindhyan region in India. The soil block method and root-free-soil cages were used to investigate fine-root dynamics. The mean annual fine-root biomass was 596 and 690 g/m² in harvested and mature sites, respectively. The fine-root net production calculated by different methods ranged from 486 to 749 g m^-2 yr^-1 in the harvested site and 485 to 875 g m^-2 yr¹ in the mature site. All fine-root mass fractions decreased with increase in distance from the base of bamboo clumps, and the herb root mass showed the reverse trend. Bamboo fine roots were better developed in the 10 - 20 cm soil depth and those of herbs in the upper 10 cm. The ingrowth of fine roots in root-free-soil cages showed maximum biomass accumulation during the rainy season (64.2 - 69.9 g m^-2 mo^-1) and minimum in the summer (4.5 - 7.5 g m^-2 mo^-1). The fine-root nutrient concentrations were strongly related to their diameter. The fine-root nutrient concentrations varied considerably in different seasons. The highest nutrient concentrations in all categories were recorded in summer followed by winter and rainy seasons. Nutrient concentrations in live roots were always greater than those found in dead roots in different diameter classes. We suggest the occurrence of nutrient retranslocation from senescent roots to surviving roots in bamboo savanna. Fine roots in the bamboo savanna increased as a function of N-mineralization and nitrification rates. This tendency further increased after the harvest of bamboo, suggesting the crucial role of fine roots in the bamboo savanna after the harvesting of bamboo culms.     

Effect of winter fire on primary productivity and nutrient concentration of a dry tropical savanna

Burning increased the mean annual canopy and belowground biomass of a dry tropical savanna by 40% and 12%, respectively, while littermass was reduced by 85% in comparison to control savanna. Mean annual aboveground and belowground net primary production were 471 and 631 g m^-2 in control, and 584 and 688 g m^-2 in burned savanna, respectively. Fire caused an increase in mean aboveground net production of 24% and in belowground net production of 9%.Concentration of carbon, nitrogen and phosphorus in vegetation of unburned plots ranged between 34.01–38.59%, 0.85–1.53% and 0.04–0.11% and in soil from 0.95–1%, 0.011–0.13% and 0.017–0.02%, respectively. Fire increased the mean concentrations of N and P by 16% and 42% in vegetation and 18.18% and 17.65% in soil, respectively. Thus winter fire can be an important tool for the management of dry tropical savanna with respect to biomass production and nutritive quality.     

Nest and soil populations of Trinervitermes spp. with particular reference to T. geminatus (Wasmann), (Isoptera), in Southern Guinea savanna near Mokwa,Nigeria

Summary Nest and soil populations of Trinervitermes spp. were estimated on grazed secondary savanna woodland near Mokwa cattle ranch and on primary savanna woodland, 6 km from the ranch. Nest populations were estimated by obtaining a relationship between size of nest and the number of termites in the nest and using the relationship to estimate populations in measured nests within the study area.Mound populations of T. geminatus, by far the most abundant species, were 222 m^-2 at a mound density of 232 ha^-1 at the ranch, and 225 m^-2 at a mound density of 175 ha^-1 on primary savanna woodland. The mound population at the ranch represented a fresh weight biomass of 1.089 g m^-2. Changes in abundance of the mound population of T. geminatus were correlated with breeding and foraging cycles. Maximum numbers (388 m^-2, 2.03 g m^-2) in August/September were reduced by the flight of alates and loss of foragers to predators; thereafter, the population continued to decrease (126 m^-2, 0.57 g m^-2) until the cessation of foraging in April/May and numbers of larvae and nymphs began to increase. Soil and mound sampling in primary and secondary savanna showed that although T. geminatus is a mound inhabiting species, two thirds of the mound plus soil population was outside the mounds giving a total population of 737 m^-2 (3.08 g m^-2). Alate production was estimated at 15.5 m^-2 (0.19 g m^-2) and neuter production at 367 m^-2 (1.66 g m^-2); production/biomass ratio was 1.0 T. togoensis (total population of 21 m^-2) and T. occidentalis (200 m^-2) had 90–96% of the total numbers outside the mounds, indicating that these two species were primarily subterranean.     

Wetland vegetation ecology on a reclaimed coal surface mine in southern Illinois,USA

An early successional wetland complex on a reclaimed surface coal mine in southern Illinois was studied 1985–1987. Seasonally, biomass was low, with above-ground values of 10–210g m^–2 and below-ground biomass of 1.5–2435 g m^–2. Biomass peaked in spring and did not vary much throughout the remainder of the growing season. Stem densities were high (179–1467 m^–2) because large numbers of seedlings became established as falling water levels exposed large areas of mudflats. Fluctuating water levels led to a lack of community zonation. Species diversity (H) was low to moderate over all sites with diversity values ranging between 1.86 and 3.27.     

Root distribution,standing crop biomass and belowground productivity in a semidesert in México

In a semidesert community in México (Zapotitlán de las Salinas, Puebla) the vertical distribution of roots and root biomass was estimated at 0–100 cm depth on two sampling dates, November 1995 (wet season) and January 1998 (dry season). Root productivity at 7 to 14.5 cm depth was estimated with the in-growth core technique every two months from March 1996 to February 1998. The relationship between environmental factors and seasonal root productivity was analyzed. Finally, we tested the effect of an irrigation equivalent to 20 mm of rain on root production. Seventy four percent of the total number of roots were found at 0-40 cm depth. Very fine roots (<1 mm diameter) were found throughout the soil profile (0-100 cm). In contrast, fine roots (1-3 mm diameter) were found only from 0–90 cm depth, and coarse roots (>3 mm diameter) from 0–60 cm depth. The root biomass was 971.5 g m^–2 (S.D. = 557.39), the very fine and fine roots representing 62.9% of the total. Total root productivity, as estimated with the ingrowth core technique, was 0.031 Mg ha^–1 over the dry season and 0.315 Mg ha^–1 over the wet season. Only very fine roots were obtained at all sampling dates. Rainfall was significantly correlated with very fine root production. The difference between fine root production in non-watered (0.054 g m^–2) and watered (0.429 g m^–2) treatments was significant. The last value was the same as that predicted for a rain of 20 mm, according to the exponential model describing the relation between the production of very fine roots and rainfall at the site.     

Seasonal variation of phytoplankton community in Thale Sap Songkhla, a lagoonal lake in southern Thailand

The phytoplankton in Thale Sap Songkhla was investigated at 2–3 month intervals from August 1991 to October 1993. The abundance of phytoplankton ranged from 1.4×10⁶ to 1.3×10⁹ cells m^–3. A total of 6 divisions with 103 genera were identified as Bacillariophyta: 49 genera, Chlorophyta: 21 genera, Pyrrhophyta: 15 genera, Cyanophyta: 12 genera, Chrysophyta: 3 genera and Euglenophyta: 3 genera. Although phytoplankton abundance was distinctly greater in the first year of study (August 1991–June 1992) than in the second year (August 1992–October 1993), their patterns are similar: 2 peaks yearly. The peaks of phytoplankton occurred in the heavy rainy season (northeast monsoon) and the light rainy season (southwest monsoon). The main bloom was found during December–January, with a predominance of blue-green algae (e.g. Aphanizomenon andPhormidium) and green algae (e.g. Eudorina). Their species composition also increased, an effect of the large amount of rainfall resulting in low salinity during the northeast monsoon. The minor bloom was produced by diatoms during June–July when water salinity was moderate to seawater. Both phytoplankton numbers and species composition were high. However, unpredictably heavy rainfall during the southwest monsoon period may reduce diatom production due to rapid immediate replacement by blue-green species. Besides salinity concentration, a low total nitrogen: total phosphorus (TN: TP) ratio tended to support the growth of blue-green algae. The diversity of phytoplankton was lowest in the heavy rainy period.     

Pattern of wood litter fall in five forests located along an altitudinal gradient in Central Himalaya

Wood and total litter fall were measured in five forests of the Kumaun Himalaya situated along an altitudinal gradient from 329 m to 1850 m. Total annual wood litter fall, which ranged from 93 to 197 g m², was inversely related with altitude and positively related with forest basal cover. Total litter fall ranged from 405 to 839 g m² yr¹ and was positively related with forest basal cover. Wood litter fall was highest in the rainy season and that of total litter in the summer season. Abscissed litter accounted for a maximum proportion of seasonal wood fall in summer. Monthly non-abscissed wood fall was positively related to the monthly rainfall on each site. The fractions in lower diameter classes (0–5 to 5–10 mm) dominated the annual wood fall in all forests and accounted for 59 to 78% of the total. Fractions in >10 mm size class fall maximally during the rainy months.Nomenclature follows Kanjilal & Gupta (1969) unless otherwise stated.The authors are thankful to Dr Uma Pandey for helpful suggestions and to the Department of Science and Technology, New Delhi for financial support.     

Influence of grazing and soil conditions on secondary savanna vegetation in India

Abstract. Savanna vegetation and pertinent soil features were studied on 43 sites in a dry tropical forest region of India. Grazing intensity ranged from 0.68 to 0.98. Soil moisture was positively related to the proportion of fine soil particles (< 0.1 mm), and the latter decreased while the proportion of coarse particles (2.0-0.5 mm) increased with increasing grazing intensity. Canopy biomass ranged from 28 to 104 g/m² in grazed communities and from 230 to 337 g /m² in ungrazed communities and was positively related to vegetation cover which ranged between 30–72 % in grazed and 68 - 91 % in ungrazed communities. Vegetation cover was negatively related to grazing intensity. Species richness and diversity were highest at low grazing intensity. Using community coefficients and Detrended Correspondence Analysis, the grazed stands were clustered into six and the ungrazed ones into three communities. The grazed communities were recognised as degradation stages and the ungrazed ones as recovery stages. Only five grass species, in various combinations were able to dominate in one of the different stages. Evidently the harsh climatic conditions (high temperatures, high variability in rainfall and a long dry period) in the region permit only a few species already adapted to these conditions to participate in the succession.     

Biomass and nutrient dynamics in a littoral savanna subjected to annual fires in Congo

The dynamics of nutrient accumulation were studied between two annual fires in a herbaceous savanna of the Congolese littoral region. Trees and shrubs were not studied because of their very low density. After fire the aboveground biomass increased for 10 months up to a maximum of 520 g m^–2. The underground biomass amounted to roughly 630 g m^–2 during the dry season and increased after the fire up to a maximum of 870 g m^–2 during the rainy season. In the aerial parts, the accumulation dynamics differed according to the type of nutrient: Ca accumulation was steady until the following dry season in proportion to the total biomass, while the pattern of K accumulation was similar to the living biomass dynamic and reached a maximum four months after the fire. N, P and Mg followed a middle course. For the underground biomass, N accumulation reached a maximum value at the end of the rainy season (10 g m^–2) and decreased at the beginning of the dry season. Most of this element was incorporated into the root system. During the rainy season, accumulation in the root system was of the same order of magnitude as in the aerial parts for P and Ca, whereas it was much lower for K and Mg. Transfers of nutrients to the atmosphere during the annual burning amounted, respectively, to 85, 25, 39, 21 and 28% of the amounts of N, P, K, Ca and Mg accumulated in the aerial biomass and litter components. Losses during fire were small for P, K, Ca and  Mg compared to the available soil reserves, but not for N. The legume Eriosema erici-rosenii ought to play an important role in N input in this ecosystem.     

Influence of floristic composition on the net primary production and dry matter turnover in a tropical grassland

Plant biomass, net primary productivity and dry matter turnover were studied in a grassland situated in a tropical monsoonal climate at Kurukshetra, India (29°58′N, 76°51′E). Based on differences in vegetation in response to microrelief, three stands were distinguished on the study site. The stand I was dominated by Sesbania bispinosa, stand II represented mixed grasses and stand III was dominated by Desmostachya bipinnata. Floristic composition of the three stands revealed the greatest number of species on stand II (75). The study of life form classes indicated a thero-cryptophytic flora. The biomass of live shoots in all the three stands attained a maximum value in September (424–1921 g m^-2) and below ground plant biomass in November (749–1868 g m^-2). The annual above ground net primary production was greatest on stand I (2143 g m^-2) and lowest on stand II (617 g m^-2). The rate of production was highest during the rainy season (15.34 to 3.18 g m^-2 day^-2). Below ground net production ranged from 1592 to 785 g m^-2 y^-2 and the rates were high in winter and summer seasons. Total annual net primary production was estimated to be 3141, 1403, 2493 and 2134 g m^-2 on stands I, II, III and on the grassland as a whole, respectively. The turnover of total plant biomass plus below ground biomass indicated almost a complete replacement of phytomass within the year. The system transfer functions showed greater transfer of material from total net primary production to the shoot compartment during rainy season and to the root compartment during winter and summer seasons.     

Stream periphyton response to grazing and changes in phosphorus concentration

Grazing by the large caddisfly larva, Dicosmoecus gilvipes (Trichoptera; Limnephilidae), drastically reduced periphyton biomass in laboratory channels at a current velocity of 20 cm s^–1. Reduction in biomass as chl a and AFDW ranged from 88 to 93% and 82 to 85%, respectively. On average, grazing rate increased with in-channel SRP (soluble reactive phosphorus) content from 6 to 10 µg 1^–1. Grazing rates averaged 25.9–29.3 µg chl a m^–2 d^–1 and 10.8–12.2 µg chl a mg^–1 d^–1 based on area and grazer biomass, respectively, with most variability among treatments being due to the grazing effect. Grazing tended to shift the algal community increasingly to filamentous blue-green algae regardless of enrichment. After three weeks, Phormidium comprised over 61% of the community in grazed treatments but only 35% in ungrazed treatments. The stalked diatom Gomphonema comprised only 4% of the grazed community, but 11% in the three ungrazed channels with similar values for Scenedesmus. A model that includes grazing was calibrated to the data and produced a reasonable expectation of periphyton biomass over a range in SRP concentrations. While the model with constant grazer abundance predicts a gradually increasing grazed biomass as SRP increases, grazer production in natural streams may actually increase to accommodate the increased food production.     

Progress in understanding the chemical stratigraphy of metals in lake sediments in relation to acidic precipitation

At the beginning of the reproductive season corixids were excluded from several small-sized rock-pools by netting carried out repeatedly at 5 d intervals. In control pools corixid populations were allowed to develop normally. Changes in the bottom fauna, consisting of chironomid larvae were followed by repeated sampling.When corixids were excluded, the density of chironomid larvae remained at a relatively high level, the total biomass (as organic carbon) was 1.5–4.5 g m^–2 (the midsummer median 2.83 g m^–2). When corixids were present, the chironomid biomass fell to 0.1–1.0 g m^–2 (median at the time of peak corixid biomass was 0.52 g m^–2). The largest chironomid larvae tended to disappear first and ultimately only small larvae were present. When corixids were removed, the density of chironomid larvae returned to the level of the experimental pools in about two weeks.     

Primary production and phytoplankton in two small,polyhumic forest lakes in southern Finland

Primary production and phytoplankton in polyhumic lakes showed a very distinct seasonal succession. A vigorous spring maximum produced by Chlamydomonas green algae at the beginning of the growing season and two summer maxima composed mainly of Mallomonas caudata Iwanoff were typical. The annual primary production was ca. 6 g org. C · m^–2 in both lakes. The mean epilimnetic biomass was 1.1 in the first lake and 2.2 g · m^–2 (ww) in the second one. The maximum phytoplankton biomass, 14 g · m^–2, was observed during the vernal peak in May.     

Seasonal variation in the water quality and chlorophyll a of semi-intensive shrimp ponds in a subtropical environment

Twelve water quality variables were measured at different intervals (biweekly and twice a week) in two ponds during two consecutive production cycles in a semi-intensive shrimp farm on the Northwest coast of Mexico. The average harvest during dry season (March–July, a 95 days period) was 1822 kg ha^-1, with an average size of 16.0 g; while the harvest in the rainy season (August–January, a 165-day period) was 2125 kg ha^-1, with an average size of 11.9 g. In the rainy season, dissolved oxygen concentration tended to decrease as the feeding rates and shrimp and phytoplankton biomass were increasing until harvest. During both culture cycles, mean values of temperature, salinity and ammonium in ponds were not significantly different from those in inlet water, whereas pH levels in ponds were higher than in incoming water.Nitrogen nutrients, pH and particulate organic matter levels in ponds and inlet water did not exhibit seasonal differences; the better yield and feed conversion obtained during rainy season reflect the fact that temperature and mean chlorophyll a were higher during the rainy than in the dry season; salinity and total suspended solids were lower during the dry than in the rainy season. The mean levels of the water quality variables in the two ponds were not significantly different in both grow out cycles. Therefore, the differences in productivity between ponds cannot be explained by the water quality.     

Water and element input into native, agri- and silvicultural ecosystems of the Brazilian savanna

The interaction of rain water with the vegetation canopy results in changes of the water quantity and quality. We examined these canopy effects in different ecosystems of the Brazilian savanna, the Cerrado. The ecosystems were 20 yr-old Pinus caribaea Morelet plantations (PI), productive (PP) and degraded Brachiaria decumbens Stapf pastures (DP), continuous corn-soybean rotation (CC), and native typical cerrado (CE). We collected rainfall, throughfall, and, in PI and CE, stemflow from three plots of each ecosystem. Dry deposition and canopy leaching were estimated with a Na-tracer method. Between May 1997 and April 1999, the mean annual rainfall was 1656 mm of which 145 mm fell during the dry season (May–September). The throughfall percentage of the rainfall increased in the order, PI (75–85%) < CC (76–89%) < CE (89–100%) < PP (90–100%) < DP (99–100%); stemflow was < 1% of the rainfall. The volume-weighted mean (VWM) pH in rainfall was higher in the dry (6.5) than in the rainy season (5.4). The VWM pH in throughfall decreased in the order, CC (rainy season: 5.9/dry season: 6.2) > PP (5.5/6.0) > CE (5.2/6.0) > DP (5.2/5.6) > PI (4.8/5.7). The rainfall deposition of the dry season contributed one third of the annual element input with rainfall because of higher element concentrations than in the rainy season. The mean Na deposition ratios, i.e. the ratio of throughfall (+ stemflow) to rainfall deposition as a measure for dry deposition, increased in the order, CE (1.5) = CC (1.5) < PP (1.7) < PI (1.9) < (DP 2.1). Total deposition (rainfall + dry deposition) accounted for 104–164% of the K and Ca fertilizer application in PP and for 6.1–12% of the K, Ca, and Mg fertilizer application in CC. The P concentrations were below the detection limit of 0.2 mg L^–1 in all samples. Net canopy uptake, i.e. a smaller throughfall(+ stemflow) than rainfall + dry deposition, of Ca, K, Mg, S, Cu, and Zn in at least one of CE, PI, DP, and PP indicate that plant growth may be limited in part by these nutrients. During the vegetation period, between 28 and 50% of the applied K and Ca were leached from the canopy in PP and between 8.7 and 17% of the applied K, Ca, Mg, and S in CC. Our results demonstrate that PI causes larger water losses and enhanced acid inputs to the soil compared with all other ecosystems. However, the PI and pasture canopies scavenge more nutrients from the atmosphere than CE and CC.     

Nitrogen mineralization,nitrification and denitrification in upland and wetland ecosystems

Summary Nitrogen mineralization, nitrification, denitrification, and microbial biomass were evaluated in four representative ecosystems in east-central Minnesota. The study ecosystems included: old field, swamp forest, savanna, and upland pin oak forest. Due to a high regional water table and permeable soils, the upland and wetland ecosystems were separated by relatively short distances (2 to 5 m). Two randomly selected sites within each ecosystem were sampled for an entire growing season. Soil samples were collected at 5-week intervals to determine rates of N cycling processes and changes in microbial biomass. Mean daily N mineralization rates during five-week in situ soil incubations were significantly different among sampling dates and ecosystems. The highest annual rates were measured in the upland pin oak ecosystem (8.6 g N m^–2 yr^–1), and the lowest rates in the swamp forest (1.5 g N m^–2 yr^–1); nitrification followed an identical pattern. Denitrification was relatively high in the swamp forest during early spring (8040 g N₂O–N m^–2 d^–1) and late autumn (2525 g N₂O–N m^–2 d^–1); nitrification occurred at rates sufficient to sustain these losses. In the well-drained uplands, rates of denitrification were generally lower and equivalent to rates of atmospheric N inputs. Microbial C and N were consistently higher in the swamp forest than in the other ecosystems; both were positively correlated with average daily rates of N mineralization. In the subtle landscape of east-central Minnesota, rates of N cycling can differ by an order of magnitude across relatively short distances.     

Carbon balance of a tropical savanna of northern Australia

Through estimations of above- and below-ground standing biomass, annual biomass increment, fine root production and turnover, litterfall, canopy respiration and total soil CO₂ efflux, a carbon balance on seasonal and yearly time-scales is developed for a Eucalypt open-forest savanna in northern Australia. This carbon balance is compared to estimates of carbon fluxes derived from eddy covariance measurements conducted at the same site. The total carbon (C) stock of the savanna was 204±53 ton C ha^–1, with approximately 84% below-ground and 16% above-ground. Soil organic carbon content (0–1 m) was 151±33 ton C ha^–1, accounting for about 74% of the total carbon content in the ecosystem. Vegetation biomass was 53±20 ton C ha^–1, 39% of which was found in the root component and 61% in above-ground components (trees, shrubs, grasses). Annual gross primary production was 20.8 ton C ha^–1, of which 27% occurred in above-ground components and 73% below-ground components. Net primary production was 11 ton C ha^–1 year^–1, of which 8.0 ton C ha^–1 (73%) was contributed by below-ground net primary production and 3.0 ton C ha^–1 (27%) by above-ground net primary production. Annual soil carbon efflux was 14.3 ton C ha^–1 year^–1. Approximately three-quarters of the carbon flux (above-ground, below-ground and total ecosystem) occur during the 5–6 months of the wet season. This savanna site is a carbon sink during the wet season, but becomes a weak source during the dry season. Annual net ecosystem production was 3.8 ton C ha^–1 year^–1.