Hans Riekerk

A comparison of the watershed hydrology of coastal forested wetlands and the mountainous uplands in the Southern US

Abstract Hydrology plays a critical role in wetland development and ecosystem structure and functions. Hydrologic responses to forest management and climate change are diverse in the Southern United States due to topographic and climatic differences. This paper presents a comparison study on long-term hydrologic characteristics (long-term seasonal runoff patterns, water balances, storm flow patterns) of three watersheds in the southern US. These three watersheds represent three types of forest ecosystems commonly found in the lower Atlantic coastal plain and the Appalachian upland mountains. Compared to the warm, flat, and shallow groundwater dominated pine flatwoods on the coast, the inland upland watershed was found to have significantly higher water yield, Precipitation/Hamons potential evapotranspiration ratio (1.9 for upland vs 1.4 and 0.9 for wetlands), and runoff/precipitation ratio (0.53±0.092 for upland vs 0.30±0.079 and 0.13±0.094 for wetlands). Streamflow from flatwoods watersheds generally are discontinuous most of the years while the upland watershed showed continuous flows in most years. Stormflow peaks in a cypress–pine flatwoods system were smaller than that in the upland watershed for most cases, but exceptions occurred under extreme wet conditions. Our study concludes that climate is the most important factor in determining the watershed water balances in the southern US. Topography effects streamflow patterns and stormflow peaks and volume, and is the key to wetland development in the southern US.

Effects of timber management on the hydrology of wetland forests in the Southern United States

Abstract The objectives of this paper are to review the hydrologic impacts of various common forest management practices that include harvesting, site preparation, and drainage. Field hydrological data collected during the past 5–10 years from ten forested wetland sites across the southern US are synthesized using various methods including hydrologic simulation models and Geographic Information Systems. Wetland systems evaluated include red river bottoms, black river bottoms, pocosins, wet mineral flats, cypress domes, and pine flatwoods. Hydrologic variables used in this assessment include water table level, drainage, and storm flow on different spatial and temporal scales. Wetland ecosystems have higher water storage capacity and higher evapotranspiration than uplands. Hydrologic impacts of forest management are variable, but generally minor, especially when forest best management practices are adopted. A conceptually generalized model is developed to illustrate the relative magnitude of hydrologic effects of forest management on different types of wetlands in the southern US. This model suggests that in addition to soils, wetland types, and management practice options, climate is an important factor in controlling wetland hydrology and the magnitude of disturbance impacts. Bottomland wetlands, partial harvesting, and warm climate usually offer conditions that result in low hydrologic impact.

GROUND-WATER-TABLE RISE AFTER FOREST HARVESTING ON CYPRESS-PINE FLATWOODS IN FLORIDA

Forest removal represents one of the large-scale ecosystem disturbances that concern water quality degradation, species composition change, and wildlife habitat alteration along the Florida coast. We conducted a five-year study with the objective to address effects of two forest management scenarios on the water regimes of cypress-pine flatwoods ecosystems in the lower coastal plain. Three experimental blocks (16–21 ha) were used in this study, with one representing control (C), one wetlands-harvest-only (W), and one wetlands+uplands harvest (ALL). Within the center of each block, a representative cypress wetland and its surrounding pine upland were extensively instrumented to quantify the changes of each hydrologic variable induced by tree removal. Water levels in cypress wetlands in both treatment areas were significantly elevated about 32–41 cm on average, and outflow doubled in the five-month dry period immediately following the tree harvesting. The ground-water table in the upland was also raised by about 29 cm on average due to ALL, but it was not affected significantly during the entire post-treatment period by W. During wet periods, the treatment effects for both wetlands and uplands were not significant. Causes for spatial and temporal variability of hydrologic responses to forest harvesting are speculated to be 1) total evapotranspiration does not change significantly in flatwoods after tree removal during wet seasons; 2) specific yield of the flatwoods soils is variable in time and space; and 3) lateral water movement from uplands to wetlands. From this study, we conclude that harvesting both uplands and wetlands causes greater response than harvesting wetlands only. The impacts lasted for more than two years but were most pronounced only in the dry periods. Temporal and spatial variations of each hydrologic component should be considered in evaluating the hydrologic impact of forest management on the flatwoods landscape.

Effects of clearcutting and site preparation on water yields from slash pine forests

Abstract Three operational-sized watersheds in poorly-drained pine flatwoods forests on sandy soils of the Lower Coastal Plain in north Florida were isolated and continuously monitored for more than 3 years. Recording flumes assessed quantity of runoff water. After 1 year of calibration monitoring, two of the watersheds were harvested, site-prepared, and planted under distinct harvest and regeneration practices. Minimum practices, imposed on one watershed, consisted of manual shortwood harvest of pinelands, roller drum chopping of harvest residues and residual understory, bedding, and planting. Maximum practices, imposed on another, consisted of tree length logging of pinelands with heavy equipment, extraction of lightwood stumps, burning and windrowing of logging residue and residual understory, harrowing, bedding, and planting. Both systems increased water yields. Following minimum practices, yield increased intermittently depending on weather. Following maximum practices the increase in water yield was quicker, larger, and more persistent over various seasons and weather conditions. In both cases water yields returned to preharvest levels or less within the year following planting.

Simulation of evapotranspiration from Florida pine flatwoods

An evapotranspiration model (ETM), including three submodels of transpiration, rainfall interception and substrate (soil or water surface) evaporation, was developed for the simulation of evapotranspiration (ET) from cypress (Taxodium ascendens) wetlands and slash pine (Pinus elliottii) uplands in Florida flatwoods. Transpiration was scaled up from the leaf level to the ecosystem level by incorporating information on ecosystem structure (e.g. species composition, tree density, leaf area index), turbulence transport in and from the canopy, stomatal conductance and meteorological information. The rainfall interception submodel was physically derived from the interception processes. The substrate evaporation submodels for the open surfaces in the wetlands and forest floors in the pine uplands were developed based on field observations. The model was tested with an independent data set acquired by eddy correlation. Daily and seasonal patterns of ET, bulk aerodynamic conductance, bulk stomatal conductance and canopy dryness index of these two ecosystems were simulated and the sensitivity of ET to environmental conditions was evaluated.

Leaf litterfall, leaf area index, and radiation transmittance in cypress wetlands and slash pine plantations in north-central Florida

The seasonal dynamics of leaf litterfall and leaf area index (LAI, all-sided basis), light penetration and the vertical distribution of surface area index, and the feasibility of estimating LAI from radiation transmittance were studied from April 1993 to March 1994 in the canopies of three cypress (Taxodium ascendens) wetlands and their surrounding slash pine (Pinus elliottii) uplands in Florida flatwoods. Annual leaf litterfall ranged from 324 to 359 g m−2 in the wetlands, which was very close to the average for 11 sites throughout Florida of 340±26 g m−2. The seasonal pattern of the normalized LAI obtained for the dominant tree species in the ecosystems could be used to construct the seasonal dynamics of LAI at the ecosystem scale. The vertical distribution of surface area index in the wetlands was significantly different from that in the surrounding pine uplands. The maximum LAI of cypress wetlands in this area was about 8 m2 m−2, which was higher than the maximum of slash pine plantations of 6 m2 m−2. Cypress leaves were strongly erectophile in space. Results showed that the LAI-2000 canopy analyzer could generally be used to estimate forest LAI, whether the forest canopy was closed or not, if an overall clumping index of 2.00 was applied. However, as LAI decreased, the relative error contained in the radiation-based LAI estimates increased. This indicated that foliage clumping at the stand scale was more important than that at the tree or branch scale.

The hydrology of cypress wetlands in Florida pine flatwoods

The poorly drained pine flatwoods of the Lower Coastal Plain of the Southern United States, including Florida, contain many pond cypress (Taxodium ascendens) wetlands, which cover about one-third of the area. Management of the resource includes pine silviculture and cypress harvesting for lumber, plywood, paper, and mulch. Concern about the ecological impacts and hydrologic effects prompted a cooperative study of cypress wetlands integrating several disciplines. This paper reports results of clear-cut harvesting on the wetland hydrology. Three wetlands of about 0.5 ha were selected and instrumented to measure the climatic and hydrologic variables before and after treatments from January 1993 to January 1997. Silvicultural treatments were wetland-only clear-cut harvesting, wetland plus surrounding upland clear-cut harvesting, and an undisturbed control. The absence of observable soil surface runoff (high infiltration rate) and slow ground-water movement in the upland pine flatwoods suggested that normally the precipitation and evapotranspiration balanced each other and that the wetlands generated most of the runoff from the landscape mosaic as a whole. However, the results showed that open-water evaporation after wetland harvesting exceeded evapotranspiration of the control, explaining in part a decrease in outflow after wetland-only harvesting. Increased runoff from the pine upland, generated by reduced evapotranspiration and expanded saturated areas after elear-cut harvesting, apparently was buffered to some extent by increased evaporation from the embedded clear-cut cypress wetland. The average open-water area was about fifty percent larger than the wetland area as defined by the vegetation. However, excess wetland water balance data suggested the presence of a rain-catchment area that was 2–3 times larger than the vegetative wetland area because of semi-saturated soil in the low slopes. Therefore, the actual catchment area of a cypress wetland in the pine flatwoods may be variable in time, space, and silviculture depending on the topography, the extent of open water, and saturated soil. The application of this information in water management is for better control of first-year runoff from the pine-cypress landscape as a whole. Furthermore, silvicultural Best Management Practices for cypress wetland water management need to consider a variable source area for surface-water pollution that is larger than the wetland area as defined by the vegetation.

Water balance of Eucalyptus globulus and Quercus suber forest stands in south Portugal

Abstract The pulp industry is one of the major export industries of Portugal at present. Eucalyptus globulus is widely planted for this purpose. Nevertheless, the use of Eucalyptus is confronted with strong objections because of possible environmental effects. The effect on soil moisture appears to be particularly relevant in the southern and interior semi-arid zones of the country. The study utilized two watersheds near the town of Odemira in Alentejo, a southern province of Portugal. The region has a relatively humid climate because of its proximity to the sea. One 6-ha watershed contained scattered native cork oak ( Quercus suber ) with a newly developing understory of shrubs. An adjacent 19-ha watershed contained a mature Eucalyptus globulus stand, which was harvested for the study. The study included measurements of precipitation, runoff and soil water storage. Measurements of the tree stands and canopies were included for the characterization of throughfall and steamflow. The results demonstrated that the summer-dry climate depleted soil moisture to such a degree that wintertime recharge of the cork-oak watershed generated only one percent of annual rainfall into runoff. Annual rainfall increased during the study period, confounding wetter soil conditions with the effect of Eucalyptus harvesting. Canopy interception by the mature Eucalyptus stand was 16.8%, and stemflow was 2.7% of annual rainfall. Stemflow from the scattered oak trees was 1.3% of annual rainfall, but canopy interception could not be evaluated and was estimated at 4.2%. Interception by the shrub vegetation in the cork oak watershed was not measured. Analyses of water balances including periodic soil moisture measurements showed no difference in monthly evapotranspiration ( E t ) between the two vegetation types during the pre-treatment period. However, the 11 mm month −1 reduction of E t lasted only for three months. After this period the regenerating Eucalyptus coppice matched water use of the cork-oak stand and appeared to exceed this by 4 mm month −1 at the end of the drought cycle. Harvesting increased first-year runoff only by 4 mm and second-year runoff by 36 mm after soil water recharge by heavier rainfall.

Effects of clearcutting and site preparation on stormflow volumes of streams in Pinus elliottii flatwoods forests

Abstract Three operational-sized watersheds in pine flatwoods of the Lower Coastal Plain in north Florida were isolated and continuously monitored for 3 1 2 years. Recording flumes assessed quantity of runoff water. After 1 year of calibration monitoring, two of the watersheds were, harvested, site-prepared, and planted under distinct harvest and regeneration practices. Minimum practices, imposed on one watershed, consisted of manual shortwood harvest of pinelands, roller drum chopping of harvest residues and residual understory, bedding, and planting. Maximum practices, imposed on another, consisted of tree-length logging of pinelands with heavy equipment, extraction of lightwood stumps, burning and windrowing of logging residue and residual understory, harrowing, bedding, and planting. Both silvicultural systems increased stormflow volumes; maximum practices moreso. Under both systems stormflow volumes apparently increased immediately following tree harvests. The increase persisted through site preparation, planting, and subsequently for 1 1 2 years without further detectable change. Results of this and two earlier reports on the effects of silviculture on hydrologic regimes of flatwoods forests are reviewed and summary conclusions drawn.

Effects of different harvesting and site preparation operations on the peak flows of streams in Pinus Elliottii flatwoods forests

Abstract Stream flows on three poorly drained and contiguous pine flatwoods catchments were monitored for 3.5 years. One was left untouched. Pine timber from another was manually harvested at the end of the 1st year, residues were chopped, terrain was bedded, and pine seedlings planted — a minimum series of forest operations. The third was subjected to a maximum series of forest operations — harvesting of tree-length logs with heavy equipment, lightwood stump removal, burning, windrowing, discing, bedding, and planting. Unlike the minimum series, the maximum series of forest operations significantly increased the peak outflows, particularly following windrowing. The increase was estimated to exceed sixfold immediately following windrowing and to decrease slowly and linearly with time thereafter (equation 3).