Methods of Evapotranspiration Assessment and Outcomes from Forest Stands and a Small Watershed

Evapotranspiration is a very complex phenomenon, comprising different aspects and processes (hydrological, meteorological, physiological, soil, plant and others). Farmers, agriculture advisers, extension services, hydrologists, agrometeorologists, water management specialists and many others are facing the problem of evapotranspiration. This book is dedicated to further understanding of the evapotranspiration problems, presenting a broad body of experience, by reporting different views of the authors and the results of their studies. It covers aspects from understandings and concepts of evapotranspiration, through methodology of calculating and measuring, to applications in different fields, in which evapotranspiration is an important factor. The book will be of benefit to scientists, engineers and managers involved in problems related to meteorology, climatology, hydrology, geography, agronomy and agricultural water management. We hope they will find useful material in this collection of papers.


Introduction
The authors have used ET terms related to evapotranspiration terms which outcome from common equation of water budget. The components of water budget equation have been employed by the author prof. Kantor since 1970s. The terms are in compliance with terms of Encyclopedia of Forest Sciences (Burley et al., 2004) and with Committee of Hydrologic Impacts of Forest Management (National Academy of Sciences of the United States of America 2008). All study and understanding begins with good definitions (Hewlett, 1982); therefore at first we present fundamental terms of forest hydrologic cycle taken into consideration for evapotranspiration of forest stands and small forested catchments (Bruijnzeel, 2004): Precipitation (P) -rain, snow, and fog (occult precipitation) Throughfall (Tf) -sum of direct throughfall reaching the forest floor without touching the canopy plus crown drip, once the storage capacity of the canopy has been filled Stemflow (Sf) -part of precipitation travelling along the branches and trunks into the forest floor Net precipitation (Pn) -sum of throuhgfall and stemflow [Pn = Tf + Sf] Interception losses (Ei) -portion of the precipitation intercepted by the canopy and evaporated back to atmosphere -evaporation from a wet canopy [Ei = P -(Tf + Sf)] Transpiration -water taken up by the roots and returned to the atmosphere via the process of transpiration (Et) -evaporation from dry canopy Evaporation from the litter and soil surface (Es) Evapotranspiration (ET) -total evapotranspiration equals to sum of Ei + Et + Es (in millimetres of water per time unit) The interlocked character of the chief components of the hydrological cycle is summarized by the site or catchment water budget equation: 1 Atmosphere 2 Precipitation 3 Throughfall (direct throughfall and drip) 4 Stemflow 5 Evaporation 6a Interception loss 6b Transpiration of crown, canopy 6c Evaporation from slope surface, forest floor and ground vegetation 7 Water uptake 8 Overland flow 9 Subsurface flow 10a Groundwater flow 10b Vertical percolation into the underlying bedrock and recharging deep groundwater 11 Stream flow 12 Slope 13 Stream The system of research areas in piedmont and mountain region of the Orlické hory Mts., where the water regime of montane forest ecosystems and silvicultural methods are continuously investigated, is composed of three long-term experimental silviculturehydrologic stations: Deštenská stráň -experimental balance areas on moderate WSW slope (alt. 890 m) to complex study water budget of Norway spruce and European beech stands during progress of forest reproduction, Česká Čermná -experimental runoff areas on steep south hillside of Dubovice (alt. 500 m) to study effect of regeneration felling (clear and shelterwood) on hillslope runoff and soil erosion, U Dvou louček -experimental watershed on SW slope with moderate gradient (alt. 880-950 m) to study effect of hydroameliorative treatment and conversion of forest ecosystem after salvage felling due to air pollution on water regime of the catchment.
For the main aims of various projects, which have involved optimization of tree species composition to ensure hydrologic services of forestry, technical and biological treatments to Methods of Evapotranspiration Assessment and Outcomes from Forest Stands and a Small Watershed 77 resulted in the marked reduction of their ecological and static stability. Subsequent enforced felling measures reduced their wood-producing as well as non-wood-producing functions/effectiveness. From the hydrological point of view it is important that extensive air pollution clear-cut areas have been created associated with the marked extension of the forest road network. The density of forest roads reaches about 51 m/ha, 1/3 of the figure being represented by logging roads with frequently unsatisfactory slope conditions and drainage. Regeneration of forest stands after the disasters is aggravated by difficult air pollution and ecotope conditions. It is particularly difficult to ensure forest ecosystem biodiversity. It is expected that richer tree species composition will be applied particularly through the increase in the percentage of broadleaved species (European beech, sycamore maple, Scotch elm, and soil-improving species) as well as silver fir.

Characteristics of experimental areas and watershed 1.2.1 The Deštenská stráň Hillside experimental area
The Deštenská stráň Hillside experimental area in the Orlické hory Mts serves for studying the water balance of spruce and beech ecosystems as representatives of two most important tree species of middle-altitude mountain locations in the Czech Republic. A couple of balance plots forms the research area. Both balance plots (40x30 m each) are 50 m apart being situated on a slope with WSW aspect and mean gradient 16° at an altitude of 890 m; their latitude and longitude determines 50°19´20´´ N and 16°21'45' E respectively. Mean annual temperature is 4.9°C and mean annual precipitation 1200 mm. From the forest site classification point of view, the spruce and beech stands belong to the most widespread forest type of the spruce/beech FVZ, to the forest type of an acid spruce/beech stand with Deschampsia (6K1). From the pedological point of view, both stands can be ranked among typical acid Cambisols of higher altitudes, sandy loam to loamy sand with the 50% mean admixture of skeleton, the proportion of which reaches 90 -98% at a depth of 0.7-1.0 m (weathered parent rock being mica schist).
The study was started in autumn 1976. During the first five hydrological years (1 November 1976 to 31 October 1981), hydrological effectiveness of mature stands was studied. In winter 1981/1982, both stands were felled at a time and immediately (in spring 1982) the research plots were reforested again with Norway spruce and beech so that since 1 November 1982, the study of the water regime of both species could be continued under changed unfavourable air pollution/ecotope conditions. In the course of the experiment, the following components of the water regime have been continuously studied by the same procedures and using the same measuring devices on both plots, both in mature stands and in newly established plantations/present young growths (Kantor, 1992(Kantor, , 1995: interception and transpiration of trees (in mature stands being determined numerically as the only unknown quantity in the equation of water balance), evaporation from the soil surface, interception and transpiration of ground vegetation, changes in soil moisture, overland flow, lateral water flow through the soil, water seepage to the subsoil (with the following subsurface flow), snowpack depth, density and water equivalent value, air temperature and humidity.

The U Dvou louček experimental watershed
The U dvou louček experimental watershed was established for the purpose of studying the problems of draining a waterlogged forested watershed situated on a mountain slope. The In terms of hydrogeology the watershed belongs to the crystalline complex of the Eagle Mts. with a low even medium crevice permeability of the rock and unstable penetration of the quaternary mantle. The rock consists namely of honeycombed muscovite gneiss of the Proterozoic Era and mica schist of the strontian series. Results of hydrodynamic tests in boreholes carried out in the watershed showed that gneiss in the hydrogeology structure of the watershed plays the role of a collector and the mica schist acts as a hydrogeology insulator. In the upper part of the watershed where gneiss appears the supply of ground water is dependent exclusively on atmospheric precipitation. Due to the good permeability of gneiss and gradient of the terrain the water continually flows off via the underground. In contrast, the area of the mica schist in the lower and middle parts of the watershed prevents the water to flow via the underground and it swells to the surface. Since the supply of water from the gneiss area is virtually non-stop, the area is continually waterlogged. On the gneiss cambisol (brown forest soil) developed and partly also humus podzols; on the mica schist humus podzols, gleyey peat podzol and peat soils. The soil is loamy-sand, in some horizons clayey, extremely stony (20 to 50%) and non-homogeneous. From the point of view of the dynamics of soil water the soils are permeable, with a high infiltration capacity. The soil areas with gleyey podzolic soil and peat are oversupplied with water and have a high water level caused by a high inflow of water from the higher areas.
In the U Dvou louček watershed, hydrological and silvicultural research programmes are conducted. The hydrological programme includes watershed calibration in hydrological years 1991/92-1995/96 including recording hydropedological characteristics, manual implementation of the ecological draining measure in summer 1996 and the study of the effects of draining on further hydrology and hydropedology of the watershed. Draining measures aimed at the restoration of functions of the existing drainage system and the www.intechopen.com Methods of Evapotranspiration Assessment and Outcomes from Forest Stands and a Small Watershed 79 interception of runoff from spring areas and areas with insufficient drainage were carried out on the area greater than 2 ha. The length of drainage ditches reached about 500 m. The silvicultural research programme follows up the implemented drainage. It deals with the improvement of the survival and growth of established spruce young plantations and increasing their biodiversity and ecological stability. In part of the drained area protected by fencing of an area of about 1 ha with three degrees of the soil profile moisture (moist, moister, moistest), European beech, sycamore maple and silver fir are interplanted into the spruce plantation and on unforested places since 1997. Various types of mound and ridge planting and amelioration using natural materials (basic rock meals) are applied.
To obtain input data, precipitation in the watershed is measured by 8 station rain gauges and 2 ombrographs connected with the automatic meteorological station. Other characteristics of the soil and air sphere are also determined as well as the dynamics of air pollution flow by a summation method. Vertical flow onto the bedrock using ten buried open lysimeters (deep infiltrometers) placed at a depth of 0.75 m with a total area of 1 m 2 and lateral subsurface flow from the interface of organic and mineral horizons and from the interface of more loose and more compacted mineral layers was also measured. Groundwater table is monitored by 52 water table measuring perforated pipes placed in two transects (perpendicularly to the contour and across ditches) and on four microplots differing in vegetation cover (peat moss, reed, hairgrass and mixed growth). Suction pressure of soil is measured by 288 tensiometers at a depth of 0.15-0.60 m in two elementary runoff-balance plots (10 x 10 m each) differing in moisture. On the elementary runoffbalance plots (ERBP) groundwater table is measured by 42 soil water level measuring perforated pipes built-in depth of 0.7 m. The flow (discharge) in the closing profile of watercourse was monitored by mechanical float water level recorder. Since 1996, the flow is recorded by manometric water level recorder with automatic data collection.

Method of ET calculation from water balance equation
2.1 ET calculated from complex of measured water balance components in a forest stand with only single unknown Et 2.1.1 Outline of method procedure The method of ET assessment is performed on the Deštenská stráň Hillside experimental area. Investigations, from which the ET components for mature Norway spruce and European beech stands were obtained, we have done in a continuous sequence of five water years (from 1 Nov. 1976to 31 Oct. 1981. Two components of vaporization process, i.e. interception losses (Ei) and evaporation from the litter and soil surface (Es) were ascertained directly; the tree canopy transpiration (Et) was estimated by calculation from water budget equation as the only unknown: where P is precipitation, Tf throughfall, Sf stemflow, Es evaporation from forest floor and ground vegetation, OlF overland flow, SslF subsurface lateral flow, VF vertical flow (infiltration) to bedrock, and ΔS change in soil water storage. From these components of water budget, the total evapotranspiration (ET) was possible to calculate as the sum of Ei + Et + Es (in millimetres of water per time unit).
Interception losses were calculated as the difference of open area precipitation P and net precipitation Pn = Tf + Sf. In both stands, throughfall (Tf) were measured by a series of 10 station ombrometers each of them with an orifice of 500 cm 2 , located in spacing 4 m along a contour line. Stemflow (Sf) was collected from spiral collar attached to a tree trunk and conveyed by hose into gauge barrels; Sf was recorded from 3 trunks in the spruce stand and 11 trunks in the beech stand of all diameter classes. Evaporation from forest floor and ground vegetation (Es) were found out using modified Popov's evaporimeter. It is a double cylinder container; the inner cylinder, its bottom is covered by wire mesh, is filled by an undisturbed soil monolith. Its circular cross section is 160 square centimetres and depth 0.2 m. The evaporimeter is inserted in the soil so that its upper edge may be levelled with surrounding soil surface. Evaporimeters were weighted in regular time intervals usually once a day. After precipitation, water infiltrated through evaporimeter was measured. From weight difference of the soil monolith the evaporation was calculated. In each of both balance plots 20 evaporimeters were installed. In the spruce stand, 6 evaporimeters were planted by bilberry (Vaccinium myrtillus) and six ones by hairgrass (Deschampsia flexuosa), two main dominant plants of ground vegetation. Evaporation from soil surface was estimated from weighing eight evaporimeters. In the beech stand, all twenty evaporimeters were used to determination of evaporation from soil surface. The further necessary components of water balance were obtained by following procedures.
Precipitation ( Soil moisture was surveyed using a gravimetric method. Disturbed soil samples were taken from 4 soil horizons (characterizing a soil profile) on three places both in a spruce and in a beech stand. Sampling was done usually once a week during growing season in rainless days.
Measurements of water budget components during growing season were done in all days with measurable precipitation, i.e. usually 2 to 3 times in a week. Interception losses of both stands were further decreased by occurrence of occult precipitation which increased near equally value of net precipitation in observed growing seasons by 40 to 70 mm, i.e. by 5 to 10% of summer precipitation. Practically the same ability of both stands to obtain occult precipitation is explainable on one hand by greater intercepting area of needle biomass, on the other hand by better conditions for precipitation travelling along the branches and trunks and lower storage capacity of the broadleaved tree. Evaporation from forest floor and ground vegetation (Es) did not present any important difference between both stands during investigated growing seasons. They equal to 75.6 mm, i.e. 10.4% and 72.0 mm, i.e. 9.9% in the spruce and beech stand, respectively. The plantations influenced the water budget in years after reforestation only insubstantially. Both interception (Ei) and transpiration (Et) was negligible due to small needle and foliage biomass. Water regime in both established plantations was being successively influenced by evaporation from ground vegetation. In 5 years after logging weeds had been infesting near the whole surface of both balance plots. In vegetation cover the following species dominated: Rubus idaeus L., Carex sp., Avenella flexuosa (L.) Drejer, Calamagrostis arundinacea (L.) Roth, Deschampsia caespitosa (L.) Beauv., and Calamagrostis epigeios (L.) Roth. In summer mean dry matter of weed above ground biomass exceeded 3 metric tons per hectare. Evaporation of ground vegetation (Es), i.e. its interception and transpiration ranged from 268 to 332 mm (i.e. from 40.8% to 67.7% of summer precipitation) during 5 growing seasons after clearcutting.

ET obtained from complex of partially measured water balance components and partially derived from tensiometric measurements in a small watershed with only single unknown E(t, s) 2.2.1 Outline of method procedure
The method of ET assessment is performed on the U Dvou louček experimental watershed (UDL). Two constructed elementary runoff-balance plots -ERBP (each of size 10 x 10 m and slope 6°) were used for ET determination by above mentioned method (viz heading).

Description of ET model design
The scheme of the measurement assessment for the ET estimation is shown on the Figure 3.
The ET is calculated as the single unknown from the following water budget equation: Et + Es = E(t, s) transpiration of forest stand and evaporation from forest floor and ground vegetation Q(c, sc) vertical flow (outflow or inflow) of capillary and semicapillary soil water Δ SW(a) change in content of capillary and semicapillary soil water in aeration (unsaturated) zone Δ SW(g) change in content of gravitation water in soil layer with depth of 0.6 m Q(g) discharge of gravitation water E(t, s) = Et + Es transpiration of spruce pole-stage stand and evaporation from forest floor and ground vegetation The model (design, pattern) given by Figure 3 and Equation (3) takes in consideration two interlinking forest soil zones, i.e. zone gravitational and zone capillary including semicapillary. Soil moistures are derived from suction-pressure measurements using retention curves created by a laboratory. Vertical capillary and semicapillary outflow (or inflow) is determined on the basis of Darcy equation for an aeration (unsaturated) zone using measurement of suction pressures and responding coefficients of unsaturated hydraulic conductivity. These coefficients have been found out by treatment of volume soil samples using the vaporization method by Schindler with the instrument Ku-pFUGT Muenchebeck applying approximation of measured data by Van Genuchten equations in the programme RETC, version G. Suction pressures for calculation of capillary, semicapillary soil water including its flow in partially saturated soil and hydraulic depths of gravitational water in auger holes are measured by tensiometers and perforated water-level pipes, respectively. These data are collected in 6 -10 day intervals.

Results of model
Results of the model are presented for the growing period of moderately dry year 2008 (precipitation equalled to 77% of long-term mean) in Tables 4 and 5.
Overland flow (OlF ) attains only negligible values. Therefore, these values are not taken in consideration at computing the model outcomes by the Equation (1). Interception losses (Ei) of forest stand in common with E(t, s) results in estimating total evapotranspiration (ET) equal to 303 -309 mm per evaluated time period of 148 days in the growing season. Interception equal to 22.9% of precipitation (P) is consistent with data from other research areas mentioned in the Orlické hory Mts. Q(c, sc) represents ascending or descending vertical flow of capillary and semicapillary soil water in the aeration zone. The soil of both ERBP has not been during the whole growing period 2008 saturated. The soil moisture has not descended under lentocapillary point at pF 2.8 (point of decreased availability). Gravitational water of auger holes and buried open lysimeters (subsoil infiltrometers) is located in irregular space network of gravitational pores. These devices serve for observing level of gravitational water and bring information on its dynamics. During dry period the instruments present vertical fluctuation of gravitational water; during stormflow they give level of gravitational water in gravitational pores in combination with its lateral and vertical movement. Volume portion of gravitational pores ranges between 3.3 -11.7% in ERBP "Nad cestou" and 2.2 -6.4% in ERBP "Pod cestou" in dependence on soil depth (cf . table 3). Gravitational pores create in forest soil irregular network spatially restricted. Flow in gravitational pores is evident and its proportion ranges between 28 -30% of total discharge from the ERBP. The total runoff of gravitational water during observed growing period amounted to depth of 30.7 mm (cf. tables 4 and 5). The value was estimated using similarity with hydrological balance model of the Divoká Orlice River basin (Horský 1970), where the experimental watershed is located. The time behaviour of gravitational water discharge was then derived in relation to time behaviour of runoff obtained from the closing profile of the "U Dvou louček" watershed. The procedure is substantiated by knowledge of hydrogeological survey that the portion of gravitational water vertically percolates into the underlying bedrock and recharges deep groundwater. At computing procedure the measurements from open subsoil infitrometers and auger holes were taken into account. Transpiration of tree canopy and evaporation from forest floor and ground vegetation E(t, s) of depth 212 -218 mm per evaluated time period of 148 days in the growing season, i.e. in average 1.4 -1.5 mm per day, is consistent with measurements on other examined research areas in the Orlické hory Mts.  -7.0 -9.0 -11.2 -24.9 -15.7 -12.6 -11.6 -9.6 -18.8 -10.9 -23.9 -9.5 -16.     Table 5. The hydrologic balance of the ERBP no. 2 "Pod cestou" in the "U Dvou louček" The presented model of hydrological balance brings values of ET and Q consistent with experimental results from comparable conditions and with runoff data issued for streams in upper part of the Divoká Orlice River basin by Czech Hydrometeorological Institute (Horský 1970). The results of investigation represent conditions of mountain land covered predominantly by coniferous forest in the Czech Republic.

Hydropedological and hydrological methods of ET assessment
3.1 ET obtained from continuous measuring the volumetric moisture of soil profile in a forest stand and calculation of E(t, s) from soil moisture differences 3.1.1 Outline of method procedure A long-term observation of the water balance elements during the forest regeneration of a Norway spruce stand and that of European beech on the Deštenská hillside in the Orlické hory Mts serves also for comparing evapotranspiration (ET) of both stands. ET represents interception losses from the canopy (Ei), transpiration of forest stand (Et), and evaporation from forest floor and ground vegetation (Es). The first observations were made in mature spruce-and mature beach stands (1976)(1977)(1978)(1979)(1980)(1981); a clearcut harvest with hole planting of spruce and beech transplants followed in 1982. Then, the research continued during the growth, development, and thinning of both stands from the stage of young plantation up to the small pole stage . The transpiration of the tree species was calculated as the only unknown of the water budget equation (cf. Kantor 1985, and the first method in this ET chapter). In 1998, we started a continuous measuring of soil moisture by volume in layers of the soil segments with the aim to come up with a new procedure of direct determination of ET. In 2005, a method of E(t, s) determination in the spruce and beech stands was devised on the basis of volumetric moisture changes in the soil profile -E(t, s) by Soil Water Content Variation, in abbreviation E(t, s)-SWCV and the results obtained in the growing season of 2005 were published (Šach et al. 2006). In the winter of 2005/06, the young spruce stand (25 years old) was completely damaged by crown and stem snowbreaks. The extreme disturbance of the forest environment in the young experimental spruce stand after the snow breakage disaster in the winter of 2005/2006 became an impulse to carry out the next study. The investigation was based on two methodical procedures: -assessing the E(t, s) of the forest stands based on continuous measuring of the water content in the root zone of the soil profile, -intermittent measuring of the evaporation from the soil surface including the ground vegetation separately by dominant species (Es). A comparative investigation was simultaneously done in the young experimental beech stand. The aim of this part of the ET chapter is to present the method E(t, s)-SWCV, examples of its use, its validation, and some results obtained during recent growing seasons.

Description of forest stand development and present feature of the spruce and beech stands on experimental balance plots
1976-1981: observation of the water balance components run in the mature spruce and beech large-diameter stands (interception and transpiration of trees, soil surface evaporation, soil moisture changes, surface runoff, seepage of water, snow cover parameters, air temperature and humidity). 1982: forest regeneration by the clear felling method and hole planting of spruce and beech.
www.intechopen.com 1983-2005: following the observation of the water balance components during the growth and progress, and tending both stands from plantation to small pole stage and pole stage stands (Kantor 1992(Kantor , 1995 including foliage biomass. 2005-2006: 25-year-old spruce stand was severely damaged in winter by crown and stem snow breaks, the young beech stand was afflicted with snow breaks only minimally; 98% spruce trees were affected by snow breakage, the stand density decreased from 1550 to 950 trees per ha, the needle foliage of the stand was reduced to about 40%, and the stand canopy was markedly disturbed. 2006: following the observation of the water balance components in the remedying spruce and beech stands after the snow disaster. 2007: stand gaps began to get infested by forest weeds whose cover reached up to 80% in the summer and autumn.

E(t, s) of the young spruce and beech stands
During the growing season from May 1 to October 31 in 2005October 31 in , 2006October 31 in and 2007 was determined by the calculation obtained through the continuous measurement of the volumetric moisture changes in the soil profile (Šach et al. 2006). An analogous procedure, e.g. Tesaį et al. (1992) and Vilhar et al. (2005), was also used but with discrete data from discontinuous observations. By rooting through depth, we induced the thickness of the root zone equal to 500 mm for calculating E(t, s). The volumetric soil moisture was measured with the VIRRIB transducers belonging to sufficiently precise ones in the estimation of the changes in the volumetric soil moisture. The transducers were placed into the deductive root zone in depths of 50 mm, 200 mm, and 500 mm with 3 repetitions. The repeating followed the forest stand variability. In 3 depths with 3 repetitions the total of 9 transducers were placed into each forest stand.

Procedure of calculating E(t, s) of a forest stand
The calculation of E(t, s) in mm for a particular soil layer per month was done by using the formula: (4) where E(t, s) LM E(t,s) for a soil layer (mm/month) ∑W V sum of volumetric soil moisture decrements as decimal number D SL soil layer thickness (mm) S VP skeleton volumetric proportion as decimal number (an especially important entry) The sum of E(t, s) LM for three observed soil layers of the root zone represents the E(t, s) of the forest stand in the respective month. Using the newly devised method, we can also calculate the daily values of E(t, s) (Šach et al. 2006).

-
We included the changes of the volumetric soil moisture into the calculation (the mean of 3 repetitions in the same depth) if the volumetric soil moisture in the subsequent record was lower than that in the preceding one. - We did not usually include into the calculation small decreases in the volumetric soil moisture at night considering six-hour intervals (0, 6, 12, 18, 0, 6... hours of Central European Time -CET = UTC + 1).

www.intechopen.com
Methods of Evapotranspiration Assessment and Outcomes from Forest Stands and a Small Watershed 91 -Decreases in the volumetric soil moisture during 12 hours after rain were considered to be a vertical flow and, especially at a low air temperature and a high air humidity (usually 100%), we did not take them into calculation (similarly Cheng 1987 under comparable conditions). -During rain, when the volumetric soil moisture usually increases, we did not take E(t, s) into consideration.

Evaporation from soil surface including ground vegetation in the young spruce and beech stands -Es
Es was determined by intermittent accurate weighing sets of Popov's evaporimeters in the summer hydrologic half year of 2005, 2006, and 2007. The evaporimeter with evaporative circle cross-section equal to 160 square centimetres indicated the evaporation from the soil layer 0-20 cm. The evaporimeter set (more than 10, usually 16) represented the proportional soil cover in the spruce and beech stands, and newly for comparison also on the clearcut.
The observations were realised during characteristic rainless periods on the beginning, in the middle, and at the end of the growing season. They provided the results about the morning, afternoon, and night evaporation. The determination of evaporation is based on an accurate weighing of evaporative vessels at regular time intervals 3 to 4 times per day by digital scales with accuracy of ± 0.1 g. Three comparable 5-day cycles were evaluated from the 14 th to the 19 th June in the respective years.

Validation and examples of the method use
The results of devised method are consistent with those represented by Kantor (cf. with the first method in this ET chapter). Average E(t, s) during growing season 1977-1982 equalled to 258 mm for the mature spruce stand and 234 mm for the undisturbed pole stage spruce stand in growing season 2005 found by the method of E(t, s)-SWCV (Šach et al. 2006). Similarly, average E(t, s) during growing season 1977-1982 equalled to 248 mm for the mature beech stand and 221 mm for the undisturbed small pole stage beech stand in growing season 2005 found by the method of E(t, s)-SWCV (Šach et al. 2006). Methodical procedure of ET finding based on computing changes of recorded volumetric moisture in a soil profile and obtained results are consistent with procedures and data of further authors doing research in comparable mountain conditions; these results were discussed in papers by Šach et al. (2006) and Černohous, Šach (2008). However, accuracy of the devised method E(t, s)-SWCV is increased by continuous recording volumetric moisture and computing ET from its differences, and also taking volumetric stoniness into account in particular layers of a soil profile. The method of E(t, s) computation in the young spruce-and the young beech stands based on the volumetric moisture changes in the soil profile (soil water content variation -SWCV) as applied on the Deštenská hillside in the Orlické hory Mts, comes from similar principals as the method based on tensiometric measuring of the suction pressure in the soil profile in mature spruce and beech stands on a NE slope in the experimental object Zdíkov-Liz in the Šumava Mts. (Mráz et al. 1990). Also the observed soil profile depths (100, 200 and 500 mm) corresponded practically to those on the Deštenská hillside including the features and course of drawing water for E(t, s). In the observed growing seasons of 1986-1989, the calculated ET in the mature spruce stand were equal to 274 mm on average. ET in the mature beech stand was calculated only in the growing season of 1989 and its value exceeded 300 mm.

92
The procedure of E(t, s) determination from differences of continuously measured soil moisture by volume presents sufficient accuracy, if E(t, s) values added into water budget equation completed hitherto the single unknown, cf. Table 6 and 7 by Kantor et al. (2008).   The E(t, s)-SWCV method may be made more precise if the results obtained from Es determination are used, especially at open canopy and weed infestation, e.g. after snow breakage of a forest stand (Fig. 4). The method of E(t, s) estimation by SWCV may be employed also in computation of daily E(t, s) values (Tab. 8).
3.2 ET calculation from daily variation of baseflow between day and night from the small forest watershed 3.2.1 Outline of method procedure Within the soil water regime observation of the partially waterlogged mountain catchment U Dvou louček (near the village Įíčky in the Orlické hory Mts, Czech Republic), we recorded diurnal streamflow variation during the precipitation free period. The fluctuation emerged after doing the drainage in support of the forest stand regeneration and forest plantation growing out. The decline of the streamflow in the daytime against that in the night time, without influence of precipitation, is caused by fluctuation of baseflow from the catchment. We suppose the cause of the streamflow fluctuation in declining total evaporation in the night time. The principal component of total evaporation in rainless period is E(t, s), i.e. transpiration of forest stand and evaporation from forest floor and ground vegetation.
The changes of baseflow during day and night time in conditions of the catchment U Dvou louček were most likely caused by loss of soil water from saturated horizons of waterlogged and drained part of the catchment and from saturated horizons of natural watercourse surroundings connected directly hydraulically with streambed. The water loss was, in all likelihood, induced by total evaporation comprising transpiration of tree species and ground vegetation, evaporation from soil, interception evaporation from vegetation cover and evaporation from water surface of streams. The interception practically did not occur in precipitation free period and evaporation from water surface of streams was negligible in relation to total evaporation from the catchment area. To express water loss from the catchment, just only transpiration of trees and ground vegetation and evaporation from soil, i.e. evapotranspiration, were substantial. The deduction resulted from known increased soil water uptake by fine roots of trees on daytime transpiration, when concurrently under the drought period the evaporation from soil and transpiration of ground vegetation raised. We submitted the hypothesis on the basis of recording of outflow decrease at daytime against night time, when at night processes of total evaporation were not running so intensively, and on the basis of stream discharge analyses performed on the catchment U Dvou louček in the Orlické hory Mts. Constantz et al. (1994Constantz et al. ( , 1998 were interested in similarly causes of diurnal variation of streamflow. He substantiated the fluctuation by increasing of water temperature in a watercourse during daytime against night time and consequently greater infiltration into a streambed. Ronan et al. (1998) contemplated the relation between streamflow decrease in daytime and associated evapotranspiration of riparian vegetation, nevertheless, as the main factor of that streamflow decrease during daytime they considered temperature changes of water infiltrating into a streambed. We believe the theory can be accepted only in arid regions with conditions similar to those in the Middle West of the United States, where  Constantz et al. (1994Constantz et al. ( , 1998 and Ronan et al. (1998) carried out their investigations. Under those conditions the evaporation from water surface and infiltration of water into streambeds represent the main loss factors. In humid regions, where streams predominantly drain landscape, the theory by Kobayashi et al. (1990Kobayashi et al. ( , 1995 and Bren (1997) better corresponds with local natural conditions. They consider the evapotranspiration of riparian vegetation to be the main cause of diurnal streamflow fluctuation.
The new method of calculating evapotranspiration from decrease of day discharge in comparison with night one was developed. The procedure was demonstrated on computed example by Fig. 5 and Tab. 9 and it is a part of the next passage.  We examined the decrease of day discharge against night one from streamflows received by a manometric water-level recorder in the closing profile of the catchment. We analysed drought periods with streamflow fluctuation in revealing growing seasons 1997 -1998, and the more recent year 2008. In our special conditions the water loss per daytime ranged from 2,000 to 87,000 litres. By the theory of variable source areas (Hewlett & Hibbert, 1967), we determine 3 variable source areas (VSA). We derived VSA size from length of open ditches and natural streams (forming drainage network) and their possible lateral drainage reach. Hydraulic drainage reach was calculated using the Czech Government Standard No. ČSN 75 4200: "Treatment of water regime of agriculture soils by drainage". We carried out the enumeration by average weight portion of soil particles sized up to 0.01 mm and saturated hydraulic conductivity K (m per day) of the soils in our catchment. It is possible to suppose that use of agricultural procedures was not fully precise on computing values for non-homogeneous forest soils; nevertheless, its precision was sufficient for given purpose. From total length of natural streams and drainage ditches with hydraulic reach 8 m on each of both sides resulted in 3 VSA.  Table 9. Example of computing daily E(t, s) of length equal to 500 m. The second VSA sized 18,240 square metres includes length of drainage ditches and main watercourse equal to 1,140 m in total. The third VSA sized 27,344 square metres involves strips passing along all natural and man-made watercourses on the watershed of the whole length equal to 1,709 m.

Results
If we recalculated the mean daytime loss of discharge equalled to 45,000 l on the probable area 27,344 square meters or 18,240 square meters influenced by drainage network, we obtained expression of runoff-depth loss in millimetres, i.e. 1.6 or 2.5 mm per day. The daily values of runoff-depth loss, computed for discharge loss interval 2,000 to 87,000 l from the VSA of 8,000 square meters, ranged between 0.25 to 2.5 mm per day. Decreases of 20,000 to 87,000 l related to the VSA of 18,240 or 27,344 square meters then response to 1.1 -4.8 mm or 0.7 -3.2 mm per day. The range of values corresponded with transpiration and evapotranspiration data reported by Ladefoged (1963), Stįelcová et al. (2004), Kantor (Krečmer et al., 2003), and most recently Šach et al. (2006). Ladefoged (1963)  optimal climatic conditions. On the basis of his experiments in spruce stands he further specified interval of transpiration: 1.3-3.2 mm per day for the period of May to August (September). Stįelcová et al. (2004) stated daily smoothed values of transpiration for a mature beech group in the Polana Mt. (the Slovak Republic) at an elevation of 850 m, calculated on the one hand from the measured sap flow and on the other hand using the SVAT model, for the period June−July 1996 in the range from 1.1 to 6.0 mm with mean value 1.6 mm per day.
Kantor (Krečmer et al., 2003) used his many years' investigation of water balance in Norway spruce stand and European beech one in the Orlické hory Mts for estimation of daily value of evapotranspiration. Young spruce or beech stand (from thicket to pole stand) gives daily evapotraspiration for cloudy spring, summer, and autumn day 1.1, 2.0, 1.0 mm respectively, for sunny spring, summer, and autumn day 2.6, 4.4, 2.5 mm respectively. Clear-cut with heavy weed infestation shows practically the same data: 1.1, 2.2, 1.0 mm respectively 2.6, 3.7, 2.4 mm. The interval of evapotraspiration 1.0-4.4 mm per day by Kantor is very close to interval 1.1-4.1 that we established for similar natural conditions on our experimental catchment U Dvou louček. Šach et al. (2006) derived daily evapotranspiration from decrements continuously recorded volume soil moisture in young closed stands in the Orlické hory Mts. Daily evapotranspiration of spruce pole-stage stand and beech pole-stage one ranged in rainless days of June 2005 from 1.3 to 3.3 mm respectively from 1.2 to 4.5 mm. The evapotranspiration range from 1.2 to 4.5 mm again supported the range 0.7 -4.8 mm from the examined catchment. If we return the computing, i.e. from the known runoff loss and from E(t, s) per day, estimated by different procedure, it is possible to determine the concrete size of VSA for given streamflow per day.

Conclusion
The study resumes our hitherto knowledge on influence of evapotranspiration on discharge variations in conditions of the partially waterlogged catchment in growing seasons 1997, 1995, i. e. years of phenomenon revelation, and recent growing season in 2008. The daily discharge fluctuations of streamflow, which drained waterlogged pedon, were in accordance with the theory of Kobayashi et al. (1995) and Bren (1997) who also consider the evapotranspiration of riparian vegetation to be the main cause of that phenomenon in similar natural conditions. Our observations and computations in small forested watershed in the Orlické hory Mts. (Czech Republic, EU) correspond to this theory. Values of E(t, s) 0.25 to 4.8 mm per day found by us are consistent with those of other researchers. From total length of natural streams and drainage ditches with hydraulic reach 8 m on each of both sides resulted in 3 variable source areas (VSA) different in size and area of drainage. If we return the computing procedure, i.e. from the known runoff loss and the E(t, s) per day, estimated by different procedure, it is possible to determine the concrete size of VSA for given streamflow per day.

General conclusion of the chapter
The evapotranspiration ET (Ei +Et + Es) plays an important role in ensuring of hydrologic and soil conservation services of forest and forestry.