ESTIMATING OUTFLOW AND SEDIMENT UPTAKE CHOSEN POLISH SOIL KINDS IN SIMULATED CONDITIONS

Eugeniusz Nowocień, Bogusław Podolski, Rafał Wawer

ABSTRACT

The work presents the research results on soil susceptibility to surface wash out, executed on model object (10 experimental plots of different soil kinds in black fallow at terrain slope 10%) in simulated rain (sprinkling) conditions.

Key words: sediment uptake, outflow, sprinkler, soil susceptibility to wash out, polish soil kinds..

INTRODUCTION AND RESEARCH GOALS

Polish and international data [1, 2, 3, 6] point at erosion as a main soil degradation factor. Although the processes of erosion are considerably well recognized, their quantitative valuation which remains strongly variable between local conditions, still needs continuing and widening of research in all scales, both in cameral and field approaches.

The are two main ways of field research regarding soil erosion: the first, conducted in a passive way – in natural conditions, without intervention in course of erosion processes [4, 6, 10]. The main advantage of such approach is the reflection of real state whereas the main disadvantage remains the long time period required for collecting sufficient amount of data for estimations of suitable quantitative indicators. The second method: a simulated research can be done in shorter time period, which accelerates the estimation of interdependencies between factors and effects of erosion processes [4, 6, 8, 9, 10].

The goal of the research, presented in this article, was to recognize qualitative and quantitative soil loss mechanisms in result of dispersed wash-out in more detailed approach for main polish soil kinds on uniform model micro-plots of 10% slope and uncovered surface treated with simulated precipitation.

RANGE AND METHOD OF RESEARCH

The stationary model research were localized at the experimental area at the Institute of Soil Science and Plant Cultivation in Pulawy. The location of experiments in nearness of the Institute gave the possibility to use good quality meteorogical data from adjacent station, which belongs to the Institute.

The basic element of the experiment were 10 profiles of various soil kinds not degradated by water erosion, which have been chosen in cameral studies. Each of five main groups of soils’ susceptibility to water erosion was represented by two soil kinds [4, 6]: loose sand (pl), weak loamy sand (ps), light loamy sand (pgl), strong loamy sand (pgm), light loam (gl), moderate loam (gs), regular silt (plz), loess (plg), heavy loam (gc) and strong silty loamy sand (pgmp) (tab. 1).

Then probe taking locations were chosen form 1:5000 scale soil maps, fallowed by laboratory soil texture analyses for quality checking. The taken probes were then analysed for texture, structure, hydraulic and chemical properties. Special care was given to upper, humus layer, most endangered by erosion.

Experimental plots for model research (the base of model experiments)

The experimental area for model research on water and wind erosion was made in framework of research project, financed by Polish State Committee for Scientific Research (project number 5P06H/00210), localized on the experimental station of Institute of Soil Science and Plant Cultivation in Pulawy. The experimental area was organized as fallows: 20 bottom-less chests (0.3/1.0/2.0 meters) (Phot. 1) were placed in trenches organized in 2 rows, separated by 3m wide technological path (Phot. 2).

Individual plots in each row were placed at regular intervals of 1m form each other. All plots were shaped to hold the surface steepness of 10% slope. The chests were waterproofed and secured by sheet metal plates.

Each experimental plot have gutter of 5% slope on its lower edge, which refers the overland flow water and sediments to collectors, lying beneath. The chests have been filled up with soil material taken form upper horizons of 10 chosen soil kinds, which were then hold in black fallow (with no plant cover) during hole experiment period. On each of experimental plots 2 probes for humidity measurements (digital humidity meter TDR) were installed. Large part of experimental area was assigned as storage area for soil material and equipment (sprinkler, water container, collectors for runoff water and sediments).

Soil susceptibility to overland flow in conditions of simulated precipitation

Soil matter, detached form experimental plot by simulated erosional precipitation, was piped away by 5% steep gutter through release pipe to measurement collector.

Simulated precipitation was realized with sprinkler, designed in Institute of Soil Science and Plant Cultivation [5, 7], performed in Institute of Agrophysics of Polish Academy of Sciences in Lublin (Fig. 1, Phot. 3).

Simulated raining was carried out in period from early October to late March in favorable weather conditions (positive temperature with absence of natural precipitation). Each simulation was fallowed by measurements of quantity of runoff and sediment. Simulations were ran in 30, 40, 45, 50 and 60 minutes time intervals, fallowed by measurements of simulation parameters: time of the beginning of overland flow, rainfall intensity, soil humidity before and 24 hours after the sprinkling.

The soil loss in experimental plots caused by simulations was up-to-date filled up with fresh soil material of the same origin, stored at experimental area’s storage place. Each time the surface of the micro-plot was prepared to achieve fallow – like soils’ surface structure.

RESEARCH RESULTS

In the result of model experiments we obtained 26 data series of severe variables: outflow [dm³], sediment [g], initial soil humidity [%], amount of precipitation [mm] and sprinkling time [min] for each soil kind. The average values for hole experiment period are shown in on figure 2.

On the base of severe variables derivative variables were distinguished:

• rain intensity I [mm/h], I=P/t, where P – amount of precipitation [mm], t- time of precipitation [h] (13);

• rainfall kinetic energy (13) E₀ [J/m²], E₀=(206+87*logI)*P;

• outflow coefficient w_out, w_out=O/P, where O – outflow [mm];

• unit sediment uptake off U_s [g/m²dm³], U_s =S/O, where S – sediment wash uptake [g/m²].

Statistical analyses on data series were performed for particular soil kind as well as for the totality of data series, considering texture variables.

For all examined soil kinds we have worked out analyses of the significance of linear correlations between dependent variables: outflow, outflow coefficient, sediment uptake, unit sediment uptake, and independent variables: silt, sand and clay content, soil initial humidity, amount of precipitation, precipitation’s intensity, precipitation’s kinetic energy, amount of outflow and outflow coefficient. Correlations for the whole set of examined soil kinds are shown in table 2.

On the base of the correlation matrices statistical analyses were splited into two groups of dependent variables: variables related to outflow and variables related to sediment wash off, deriving aspects of hydrology and sediment movement, characterized by significantly different physical mechanisms [12]. These was considered in later three-variables correlation analyses for runoff coefficient and sediment uptake as dependent variables, silt and sand content, initial soil humidity and runoff as independent variables, as well as in analyses of non-linear estimation by exponential model [11]. The non-linear estimations was fallowed by basic statistical analyses of match quality of received estimation function: distribution of funtions’ rests, normal graphs of rests’ probabilities.

In the non-linear, exponential estimation the amount of considered variables in hydrologic part was reduced, through introduction of dependent variable: runoff coefficient, which was modeled with silt content and initial soil humidity as independent variables (figures 3-5).

In erosional part the estimation was based on sediment uptake as a dependent variable as well as silt content and runoff as independent variables (figures 6-8). The decision to use silt content as a main texture variable came from the results of the correlation analyses, showing best correlation quality for this variable in both parts. The results of estimation have shown, that exponential non-linear estimation fits much better the hydrologic part, while it does not suit erosional part.

DISCUSSION

The largest, significant correlations for all studied soils (tab. 2) was ascertained between the volume of outflow and humidity value (0.55), then between volume of outflow and of silt fraction content (0.51), the volume of rainwash and silt fraction content (0.49), the volume of superficial rainwash and outflow volume (0.42), the volume of outflow and sand fraction content (0.40), the volume of superficial rainwash and content of sand fraction (0.36), the volume of outflow and volume of rainfall (0.30), the volume of superficial rainwash and humidity (0.21.

Correlation matrices for whole data population show that, the amount of outflow is shaped by variables of initial humidity and rainfall depth and is strongly dependent upon the content of silt fraction.

The correlations of variables the ZMYW (sediment uptake) and ZMYW_JED (the unit rainwash), describing the processes of superficial wash out show on exact dependence of the volume of sediment uptake from the volume of outflow for groups of sandy and loamy soils, what indicates superiority of the processes of superficial wash over splashing. The relationship of rainwash variable from rainfall intensity variable in group of silty soils shows however the essential role of splash processes in formation of amount of soil material carried out.

CONCLUSIONS

1. The ranking of studied soils according to increasing susceptibility to superficial rainwash in conditions of simulated rains was following: loose sand, weak loamy sand, heavy loam, medium loam, light loamy sand, strong loamy sand, light clay, loess (loamy silt), ordinary silt, strong loamy silty sand.

2. In case of investigations with utilization of simulated rain the influence of the splashing processes was practically omitable. This is reflected in low correlation factors revealed in statistical investigations of variables' pairs with precipitation energy as an independent variable.

3. The investigation with application of simulated rain poses the basis to more exact studies, than the field investigations, defining the dependences and mechanisms of physical interactions of variables: outflow and rainwash as well as derived variables.

4. The statistical investigations utilizing non-linear estimations show essential differences in representation of mechanisms of physical arrangements of variable pairs: the rain - outflow as well as outflow - sediment uptake, however the estimation of hydrology shows very good matching.

5. The introduced results presents the contribution to further investigations in the direction of evaluation of quantitative indicators and models characterizing the water erosion of soils.

1. EEA, 2000. Down to Earth: soil degradation and sustainable development in Europe. Environmental Issues Series, Number 16, 32pp. European Environment Agency.

2. EEA, 2003. Assessment and reporting on soil erosion. Background and workshop report. Technical report nr. 94/2003.European Environment Agency.

3. Fischer Weltalmanach. Wyd. Fischer Taschenbuch Verlag: 520, in German.

4. Józefaciuk A., Józefaciuk Cz., Nowocien E., 1996. Metodyczna koncepcja badań podatności gleb na spłukiwanie powierzchniowe i deflację [Methodological conception for the research of sois’ susceptibility to surface wash out and deflation]. Mater. nauk. ogólnopol. symp. “Ochrona agroekosystemów zagrożonych erozją’. Wyd. JUNG, Puławy, t. I, 259-263, [in Polish].

5. Józefaciuk Cz., 1966. Zastosowanie deszczowni do badań wodnej erozji gleb [The use of a sprinkler in research on soils’ water erosion]. Wiad. IMUZ, VI, z. 3, 285-290, [in Polish].

6. Józefaciuk A., Józefaciuk Cz., 1995. Erozja agroekosystemów [Erosion of Agroecosystems]. Bibl.Monit.Środ.: 168, [in Polish].

7. Józefaciuk Cz., Józefaciuk A., Nowocien E., 1996. Modelowe badania podatności gleb na erozję - rozwiązania techniczne [The model research on soils’ susceptibility to erosion – technical solutions]. Mater. nauk. ogólnopol. symp. “Ochrona agroekosystemów zagrożonych erozją”. Wyd. JUNG, Puławy, t.I, 265-272, [in Polish].

8. Jóźwiak M., 1992. Określenie intensywności erozji wodnej powierzchniowej w warunkach symulowanego deszczu [The estimation of the intensity of surface water erosion under conditions of simulated rain]. Zesz. Nauk. AR w Krakowie, 35, II, 105-112, [in Polish].

9. Nowocień E., Podolski B., Wawer R., 2002. Badania ilościowe podatności różnych gatunków gleb na erozję wodną w warunkach symulowanego deszczu [The quantitative research on various soil kinds’ susceptibility to erosion under conditions of simulated rain]. Wrocław, Zeszyty problemowe Nauk Rolniczych, Nr 487: 174-182, [in Polish].

10. Nowocień E., Podolski B., Wawer R., Jadczyszyn J., 2002. Określenie wskaźników podatności różnych gatunków gleb na erozję wodną i wietrzną [The estimation of indicators of soils’ susceptibility to water and wind erosion]. Raport końcowy z tematu badawczego nr 2.21, s: 106, [in Polish].

11. Ratkowsky D. A., 1990. Handbook of nonlinear regression models. ss.221.

12. Schmidt J. Ed., 2000. Soil Erosion – Application of Physically Based Erosion Models. Springer-Verlag Berlin: 307.

13. Wischmeier W. H., Smith D.D., 1965. Predicting rainfall erosion losses from cropland east of the Rocky Mountains. Agriculture Handbook, USDA-ARS, ss: 573.

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed ‘Discussions’ in each series and hyperlinked to the article.