SSSAJ Grow Your Career with SSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Poudel, D.D.
Right arrow Articles by West, L.T.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Poudel, D.D.
Right arrow Articles by West, L.T.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Poudel, D.D.
Right arrow Articles by West, L.T.
Soil Science Society of America Journal 63:1366-1376 (1999)
© 1999 Soil Science Society of America

DIVISION S-6-SOIL & WATER MANAGEMENT & CONSERVATION

Erosion and Productivity of Vegetable Systems on Sloping Volcanic Ash–Derived Philippine Soils

D.D. Poudela, D.J. Midmoreb and L.T. Westc

a Dep. of Agronomy and Range Sci., Univ. of California, Davis, CA 95616 USA
b School of Biological and Environmental Sciences, Central Queensland Univ., Rockhampton, QLD 4702, Australia
c Dep. of Crop and Soil Sciences, Miller Plant Sciences Bldg. Rm. 3111, Univ. of Georgia, Athens, GA 30602 USA

ddpoudel{at}ucdavis.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 Materials and methods
 Results and discussion
 Conclusions
 REFERENCES
 
Soil erosion is a major constraint to the sustainability of sloping-land vegetable systems. Little information is available on the effectiveness of soil conservation measures under sloping intensified vegetable systems on volcanic ash–derived soils. We hypothesized that contouring, strip cropping, and high-value contour hedgerows — asparagus (Asparagus officinalis L.), pineapple [Ananas comosus (L.) Merr.], pigeonpea [Cajanus cajan (L.) Huth], and lemongrass [Cymbopogon flexuosus (Nees ex Steudel) J. F. Watson] — reduce soil loss compared with the farmer's traditional practice of up-and-down cultivation on sloping lands. A field experiment tested these soil conservation technologies from 1995 to 1998 in a completely randomized block design on a 42% natural slope on a clayey, halloysitic, isothermic, Typic Kandiudox. The greatest annual soil loss (65.3 t ha-1) was in the up-and-down system and comparative values were 37.8 t ha-1 for contouring, 43.7 t ha-1 for strip cropping, and 45.4 t ha-1 for high-value contour hedgerows. Three rain events alone caused 47% of the total soil loss. All erosion–runoff plots showed large differences in soil properties and crop yields between the upper and the lower slope. Crop yields downslope were greater by 40% for tomato (Lycopersicon esculentum Miller), 36% for corn (Zea mays L.), and 78% for cabbage (Brassica oleracea var. capitata L.) than for upslope. In the contour hedgerow treatment, rapid terrace development changed soil properties, and crop yields for the bottom portions of bioterraces were greater by 121% for corn and 50% for tomato than yields of top portions.

Abbreviations: CBD, 0.3 M Na-citrate–0.1 M Na-bicarbonate solution and Na-dithionate • CEC, cation-exchange capacity • ECEC, effective cation-exchange capacity • ER, enrichment ratio • IBSRAM, International Board for Soil Research and Management • NIA-PRIP, National Irrigation Association-Pulangi River Irrigation Project • NOMIARC-DA, Northern Mindanao Integrated Agricultural Research Center-Department of Agriculture • NRC, National Research Council • SALT, Sloping Agricultural Land Technology • SANREM-CRSP, Sustainable Agriculture and Natural Resources Management Collaborative Research Support Program • USAID, United States Agency for International Development


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
Materials and methods
 Results and discussion
 Conclusions
 REFERENCES
 
PRODUCTIVITY DECLINE induced by soil erosion is one of the major problems constraining the sustainability of agricultural crop production in the steeplands of Southeast Asia (Hashim et al., 1995; Presbitero et al., 1995; Midmore and Poudel, 1996). Soil erosion reduces productivity in several ways: (i) loss of plant-available soil water holding capacity and of plant nutrients through degradation of soil structure and compaction and (ii) by nonuniform removal of soil within a field (National Soil Erosion-Soil Productivity Research Planning Committee, 1981). Off-site impacts of soil erosion such as siltation of reservoirs, the breaking of waterways and channels, pollution of water bodies, and effects on aquatic habitats (Midmore et al., 1996; El-Swaify, 1997) are of equal concern. Consequences of soil erosion on productivity and environment are exacerbated on steeplands where the potential for soil erosion and runoff losses is largely due to their unique topography (El-Swaify, 1997). Enrichment ratio (ER), which is defined as the ratio of the nutrient content of the sediment (eroded soil) to that of the source soil, has been used by several researchers in assessing the short- to medium-term effect of erosion in soil properties (Sombatpanit et al., 1995; Gachene et al., 1997; Wan and El-Swaify, 1998).

In attempts to stem and manage soil erosion on steeplands, a number of technologies have been researched. Alley cropping, a special form of an agroforestry system in which food crops are grown in alleys formed by hedgerows of trees or shrubs (Kang et al., 1981, 1986), has been effective in minimizing soil erosion on steeplands (Tacio, 1993; Comia et al., 1994; Paningbatan, 1994). According to Paningbatan (1994), the annual soil loss under alley cropping treatments were <10 t ha-1, while those under local farmers' practice of up-and-down cultivation were as high as 144, 44, and 41 t ha-1 for Laguna, Batangas, and Rizal provinces, respectively, in the Philippines. Comia et al. (1994) also found alley cropping to be very effective in minimizing soil erosion on the footslopes of Mount Makiling in Laguna, the Philippines. They compared conventional tillage and alley cropping systems (tilled and unmulched, tilled and mulched, and untilled and mulched). The alley cropping systems consisted of 1-m-wide wild tantan [Desmanthus virgatus (L.) Willd.] contoured hedgerows between 5-m-wide alleys where corn and bean (Phaseolus aureus L.) were grown sequentially. An annual soil loss <3 t ha-1 was recorded under a mulched alley cropping system, while annual soil loss under conventional tillage was as high as 141.3 t ha-1. The magnitude of soil loss was matched by proportional nutrient loss. Tacio (1993) reported a notable effect of Sloping Agricultural Land Technology (SALT), in which 3- to 5-m bands of crops planted between double-contoured rows of N-fixing shrubs and trees reduced annual soil loss from 194.3 t ha-1 in the control treatment to 3.4 t ha-1 in the hedgerow treatment in southern Mindanao, the Philippines.

Alley cropping on steeplands leads to the formation of natural bioterraces with a remarkable fertility gradient across the alley (Garrity et al., 1995; Garrity, 1996; Agus et al., 1997). This soil "scouring" is due to the combined effect of water erosion and tillage practices, and results in the movement of fertile topsoil from the upper part of the alley to the lower part. Garrity et al. (1995) reported a doubling of soil organic C and available P content in the lower vs. the upper alley zone. Remediatory management practices are necessary to avoid or alleviate the effect of soil scouring (Garrity, 1996).

Strip cropping, in which row crops and protection-effective crops are planted in alternating strips on the contour (Morgan, 1986), is another conservation measure aimed at minimizing soil erosion on sloping lands. In strip cropping, erosion is commonly limited to row crops and soil loss from them is trapped by erosion protection-effective crops. There is a paucity of research on strip cropping on Southeast Asian sloping lands. International Board for Soil Research and Management (IBSRAM, 1995) compared strip cropping of upland rice (Oryza sativa L.) and soybean [Glycine max (L.) Merr.] planted in 6-m strips with the farmers' practice of planting rice up-and-down the slope in Laos. Based on the soil loss records during the first 5 mo, strip cropping reduced soil erosion by 58% compared with farmers' practices. Strip cropping was also effective in lowering N, P, and K losses from fields.

Contouring is perhaps the most appreciated technique among mechanical methods for erosion control on sloping lands. In this technique, plowing, planting, and cultivation is done along the contour. Several researchers reported the effectiveness of contouring over up-and-down cultivation system (Knoblauch, 1942; Smith, 1946; Sombatpanit et al., 1995). According to Smith (1946), contouring reduced soil loss, on average, by one-half that of planting up-and-down the slope. Knoblauch (1942) reported 86% less soil loss in contoured vs. up-and-down plantings, while Sombatpanit et al. (1995) reported 66% less average soil loss on contoured vs. up-and-down cultivation.

Considerable interest on the part of various research and extension agencies, especially in Southeast Asia, exists in the planting of leguminous contour hedgerow trees on sloping lands as barriers for soil erosion and for the production of green manure for field crops. However, there is not strong evidence of farmers' adoption of these technologies (Fujisaka, 1989; Garrity, 1993; Comia et al., 1994). Some of the major constraints to the adoption of contour hedgerows by farmers include a reduction in arable land area, regular maintenance requirements without which lateral spread of hedgerows shades crops, and the dubious economic reward (Poudel, 1995; Poudel et al., 1998).

Since 1992, in an attempt to promote sustainable food production, particularly in areas most exposed to environmental degradation, the United States Agency for International Development (USAID) has funded a global program called Sustainable Agriculture and Natural Resources Management Collaborative Research Support Program (SANREM-CRSP) (National Research Council [NRC], 1991). This program is based on the active participation of all stakeholders within a catchment domain.

As an integral part of the SANREM-CRSP, a field research site was set up in 1995 in the Manupali watershed in northern Mindanao, the Philippines. Objectives of this study were to (i) quantify soil, water, and nutrient losses under different vegetable production systems, and (ii) to assess the impacts of soil erosion on soil productivity. We present results on these aspects of erosion and soil productivity.


   Materials and methods
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
Results and discussion
 Conclusions
 REFERENCES
 
Description of the Study Area
The field experiment was conducted from July 1995 to February 1998 in Victory in the Manupali watershed [124°47' to 125°08' E and 7°57' to 8°08' N (Kanemasu et al., 1997)] at 1210 m above sea level in northern Mindanao, the Philippines. The soil parent materials were thick deposits of siliceous volcanic ejecta either deposited in place (volcanic cone) or transported from upslope as colluvial or alluvial materials. Soil profile samples were taken from the face of a pit in the research site. Selected properties of the soils are presented in Table 1 .


View this table:
[in this window]
[in a new window]
  		Table 1 Selected physical, chemical and mineralogical properties of soils in the research site.{dagger}

 
Experimental Design and Installation
Twelve treatments, comprising four technologies and three crop combinations, were assigned to runoff-erosion plots in each of two blocks in a randomized complete block design. Each plot was 19 m long and 8 m wide. Natural slope of these plots ranged from 35 to 44%, with an average of 42%. The four technologies installed on the plots were: (i) up-and-down planting, representing the farmers' current practice; (ii) planting on the contour; (iii) planting vegetables up and down the slope but with strips of a bean (Phaseolus vulgaris L.) crop across the slope [two 4.75-m strips of vegetables and two 4.75-m strips of bean per plot (strips alternated between vegetable and bean crops over seasons)], and (iv) planting vegetables up and down the slope, but with four hedgerows of high-value crops planted on the contour from top to bottom at 4.0-m intervals: asparagus, pineapple, pigeonpea, and lemongrass replacing tea [Camellia sinensis var. assamica (J. Masters) Kitam.] after the first season. For each technology, a sequence of vegetable crops was planted, with approximately three crops planted per year on each plot. The three vegetable crops at the start of the experiment were cabbage, tomato, and potato (Solanum tuberosum L.), but the latter was replaced by corn after the first season, because potato succumbed to bacterial wilt. A cropping sequence of tomato–corn–cabbage was adopted to avoid any repetitive plantings of solanaceae or cruciferae sequences. A rain gauge was installed at the center of the research site. Total daily rainfall was recorded at 0700 h.

A basic soil characterization was done for each of the erosion–runoff plots before the start of the experiment. Composite random samples (0–15 cm) from each plot were analyzed for soil texture, organic C, pH, Ca, Mg, P, K, S, and total N. Two-thirds of the erosion–runoff plots were clay loam and clay textural class, while the rest had either silt loam, silty clay loam, or silty clay soil texture. However, the particle-size distribution analysis is believed to have underestimated clay content due to difficulties in clay dispersion. Final soil characterization for each of the erosion–runoff plots was done at the end of the experiment for these soil parameters.

Cropping and Management, Crop Cover, and Crop Yields
Tomato was planted in a single row system with 1.5- and 0.4-m plant spacings between and within rows. Cabbage was planted in a paired-row system, with 0.8 m between bed centers and a 0.35-m plant spacing within and between rows on the bed. Corn plant spacing was 0.75 m between and 0.25 m within rows, as was plant spacing for the beans. During the first cropping season, potato seedlings were transplanted in a paired-row system, with 0.75 m between bed centers and a plant spacing of 0.3 m between and 0.4 m within rows on the bed. Asparagus and the other high-value hedgerow crops were planted in a triangular formation within an alternating double-row system. Between- and within-row spacings in the double row were 0.5 and 0.3 m, 0.4 and 0.35 m, 0.5 and 0.35 m, 0.5 and 0.35 m, and 0.5 and 0.4 m for asparagus, pineapple, pigeon pea, lemongrass, and tea, respectively.

The first cultivation, with a single-share moldboard plow, was by draft animal, and all other cultivation and tillage practices were done manually with grub hoes. In tomato, earthing up to 0.2 m was done {approx}1 mo after transplanting. Fertilization was done according to the local practices (Poudel et al., 1998). Fertilization rates were 253 kg N, 59 kg P, and 172 kg K ha-1 in tomato; 230 kg N, 64 kg P, and 173 kg K ha-1 in cabbage; 222 kg N, 65 kg P, and 64 kg K ha-1 in corn; 127 kg N, 24 kg P, and 46 kg K ha-1 in bean; and 147 kg N, 65 kg P, and 110 kg K ha-1 in potato. These amounts were split into three equal applications: basal, first side dressing 30 d after planting, and the second side dressing 30 d after the first side dressing in all crops except cabbage, in which the basal application was followed by one side dressing. Pests and diseases were controlled with chemicals, while weeds were controlled by hand. After crop harvest, all aboveground crop residues were removed from each plot.

A 0.7 by 0.3 m wooden frame, partitioned into 100 equal grids with plastic string was positioned above each sampling point to visually measure crop cover percentage. Crop cover was measured at the top, the middle, and the bottom of each erosion–runoff plot. In each slope position, crop cover was measured within and between crop rows. The first crop cover measurement was done 2 wk after transplanting vegetable crops, or the third week after sowing corn. Subsequent measurements were made after every 2 wk until corn was 105 d old.

Crop yield data were separately collected for the upper and the lower halves of all the erosion–runoff plots. Yields of cabbage and tomatoes were expressed as fresh weights, and corn as dried shelled weight.

Eroded Soil, Runoff Collection, and Assessment of Soil Scouring
Each erosion–runoff plot was demarcated with galvanized iron sheets 22.5 cm above and 20 cm below the ground surface. Each erosion plot had a leveled soil collecting buffer at its base covered with galvanized iron sheets. These soil collecting buffers were covered by tents to avoid mixing of eroded soil and direct rainfall falling in the buffer zone. Runoff water was funneled into the first barrel of a runoff water collection system. A pair of runoff collection barrels were set at the bottom of each erosion-runoff plot. The first barrel had ten equally sized holes and one of them was connected to the second barrel with a hose, so that we could measure the amount of runoff water that was overflown from the first barrel. Both barrels were leveled every time the volume of runoff water was measured. At the end of the experiment, all collection units were calibrated to account for slight differences in weir dimensions.

After every rain event, eroded soil was collected, air-dried, and weighed. All soil samples representing every rain event were mixed at the end of the cropping season to prepare one composite sample for each plot. These samples were analyzed for organic C, pH, Ca, Mg, P, K, S, and total N. The volume of runoff water was measured for every rain event and analyzed for NO-3. Because the first runoff water collection barrel for all 12 plots was completely filled with sediment during the 27 Oct. 1997 rain event (75.3 mm), runoff measurements for this rain event were not correct. Therefore, runoff data for this rain event were not included in the analyses.

Gradients in crop establishment, growth, and yield were observed after the third cropping season in most plots. Therefore, random surface (0–15 cm), composite soil samples from two slope positions (upslope and downslope) of eight plots in corn were collected at the end of the fifth cropping season. These samples were analyzed for selected physical and chemical properties. Infiltration rate was also measured with a single ring infiltrometer and potable water at the two slope positions.

A rapid development of natural terraces was observed in all erosion–runoff plots with high-value contour hedgerows. Differences in crop growth were also observed between the upper and the lower parts of each alley. Therefore, composite soil samples (0–15 cm), each representing the upper and the lower parts of the three middle terraces of two erosion–runoff plots with contour hedgerows, were collected at the end of the fifth cropping season, and the samples were analyzed for selected physical and chemical properties. Scouring analysis was limited to the three middle terraces, each delimited by two contour hedgerows, numbered as Terraces 2, 3, and 4 starting from top to bottom. Crop yield data on these newly developed terraces in all contour hedgerow plots were separately collected from the upper and the lower zones during the last two cropping seasons.

Laboratory Methods
For plot characterization, particle-size distribution was determined by the pipette method (Gee and Bauder, 1982), and soil pH was measured in 1:2 H2O and 1:1 KCl. Soil pH for profile samples was additionally measured in 1:50 M NaF solution (Fieldes and Perrott, 1966). Total C was determined by dry combustion (Tabatabai and Bremner, 1991). Organic C was determined by a modified Walkley-Black method (Nelson and Sommers, 1982), while total N was determined by modified Kjeldahl method (Black, 1965). Available S was determined turbidimetrically with a spectrophotometer (Beaton et al., 1968). Exchangeable Al and total acidity were determined by 1 M KCl extraction (Thomas, 1982). Exchangeable K, Ca, Mg, and Na were extracted with NH4OAc at pH 7, and the cations in the leachate were measured by atomic adsorption spectrophotometry. Available P was determined with Bray-2 extraction (Murphy and Riley, 1962). The cation-exchange capacity (CEC) was determined by saturation with NH4OAc at pH 7 and subsequent replacement of NH+4 by KCl extraction. Effective cation-exchange capacity (ECEC) was calculated as the sum of extractable cations plus KCl extractable Al. Bulk density values were determined by field core sampling. Pretreatment and fractionation of clays for x-ray diffraction analyses was done according to Jackson (1979) and Whittig and Allardice (1986). Carbonate was removed by treating samples with NaOAc at pH 5.0 and organic matter was removed by using 30% H2O2 (w/v). The samples were extracted with 0.3 M Na-citrate–0.1 M Na-bicarbonate solution and Na-dithionate (CBD) following Mehra and Jackson (1960). The CBD-extractable Fe, Al, and Si were analyzed by atomic adsorption spectrophotometry. A filter-membrane technique (Drever, 1973) was used to prepare oriented clays for x-ray diffraction analyses. Saturation treatments were: Mg-saturated, air dry; Mg-saturated, glycol solvated; K-saturated, air dry; and K-saturated heated to 100, 300, and 550°C. The slides were scanned from 3 to 32° 2{theta} on a diffractometer equipped with a theta-compensating slit and curved-crystal monochromator. Specific surface area was determined by N2 gas adsorption (Brunauer et al., 1938) on a Micrometrics Flowsorb II BET instrument (Micrometrics Instrument Corp., Norcross, GA). Kaolinite and gibbsite in clay fractions were quantified with differential scanning calorimetry by comparison of endotherm area to standards (Tan et al., 1986). Runoff water samples from all rain events were analyzed for NO-3 using Merck's RQflex reflectometer (Merck, Darmstadt, Germany).

Statistical Analyses
The effect of erosion control practices and crops on soil, runoff and NO-3 losses were analyzed by analysis of variance (SAS Institute, 1994). Treatment means were compared by Fisher's protected least significant difference test (LSD), and Student's t test. Correlation analyses were done to establish relationships between selected soil properties such as pH, organic C, Ca, Mg, K, and cumulative soil loss. Similar analyses were performed to understand the relationship between the soil fertility gradient among slope positions and crop yield. A significant level of P <= 0.05 is used throughout unless otherwise indicated.


   Results and discussion
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
Conclusions
 REFERENCES
 
Rainfall Distribution
The total annual rainfall of 3381 mm in the first cropping year (1995-1996) decreased to 2492 mm in the second cropping year (1996-1997). There was a total of 1102 mm of rainfall recorded in the first half of the third cropping year (1997-1998). Mean monthly rainfall for the duration of our research was slightly greater than the 20-yr monthly mean for the Manupali watershed (Fig. 1) . Defining days with rainfall events >1 mm as "rainy", there were a total of 235, 173, and 79 rainy days in the first year, second year, and the first half of the third year, respectively. The frequency distribution analysis of daily rainfall events for this 2.5-yr period resulted in: 56% of the total events with 10 mm of rainfall or less, 22% between 10 to 20 mm, 18% between 20 to 50 mm, and 4% between 50 to 100 mm. The nineteen rain events with 50 to 100 mm d-1 showed a bimodal distribution pattern: two-thirds occurred in August to October, and one-third in March to May. These rainfall events contributed 18% of the total rainfall, while events of 10 mm or less contributed only 17% of the total rainfall. On average, 37.5% of the total annual rainfall was recorded in August, September, and October.



View larger version (26K):
[in this window]
[in a new window]
  		Fig. 1 Monthly rainfall (mm) during the experimental period and long-term mean monthly rainfall in research site

 
Soil, Runoff, and Nitrate Loss
Significant differences in soil loss were recorded among erosion control practices and among seasons (Tables 2 and 3) . Average soil loss throughout the experiment was 50% greater in the up-and-down cultivation than in the contouring, strip cropping, or hedgerows (Table 2). Results of a similar magnitude were obtained by Presbitero et al. (1995) with double-row white lead-tree [Leucaena leucocephala (Lam.) de Wit] hedgerows in steep plots in Leyte, the Philippines. In our study, there was a treatment x season interaction effect on soil loss significant at P <= 0.10, which is indicative of the increasing effectiveness of contour hedgerows as they become better established. A decline in the proportionate soil loss from the first to the seventh cropping season was evident under the hedgerow system (Table 3). Contour hedgerows lost considerably more soil during the first cropping season (47.8 t ha-1 ) than up-and-down (34.7 t ha-1), contouring (22.1 t ha-1), and strip cropping (17.8 t ha-1). This weak performance by contour hedgerows in the first cropping season was largely attributed to the digging of holes several times to transplant and retransplant tea, an effect confounded by the largest monthly rainfall amounts recorded during the experiment. Similar unsatisfactory results were reported by Hashim et al. (1995). They found greater sediment concentration and peak runoff rates during the times when holes were dug, and a large amount of entrainable material was generated during banana (Musa x paradisiaca L. var. paradisiaca) planting on the east coast of peninsular Malaysia. Entrainment, which is defined as the erosion of original soil by runoff, and reentrainment of deposited sediment, are considered important processes in soil removal in sloping lands (Rose, 1988). These processes are expected to be even more dominant on larger plots and longer slope lengths because of greater streampower (Hashim et al., 1995).


View this table:
[in this window]
[in a new window]
  		Table 2 Effects of erosion control practices on cumulative soil loss, runoff, and NO-3 loss through runoff during a cropping season (average of seven cropping seasons)

 

View this table:
[in this window]
[in a new window]
  		Table 3 Proportionate soil, runoff, and NO-3 loss between different technologies during different cropping seasons.{dagger}

 
Differences in soil loss among crops were also significant. Fifty percent more soil was eroded from tomato than from corn and cabbage plantings (Table 4) . These differences in soil loss are believed to be largely due to the variation in crop cover, although the effect of differences in the root system cannot be ruled out (Tan and Fulton, 1985). They reported the total root length of corn to be twice that of tomato, with a much greater spatial density. In our study, the average crop cover was in the order: tomato < cabbage < corn (Fig. 2) . The poor crop cover in tomato was largely due to greater distances between rows. However, such a distance is necessary to provide enough space between the rows for support stakes.


View this table:
[in this window]
[in a new window]
  		Table 4 Effects of crops on cumulative soil loss, runoff and NO-3 loss during a cropping season (average of seven cropping seasons).{dagger}

 


View larger version (48K):
[in this window]
[in a new window]
  		Fig. 2 Average crop cover for tomato, cabbage, and corn. Vertical bars represent standard errors

 
A large proportion of soil loss occurred during a few erosive rain events. Of the total 487 (1 mm or greater) rainfall events during the research period, only 32 events (i.e., 6.6%) resulted in soil loss from at least one erosion–runoff plot. Indeed, 47% of the total soil loss during the seven cropping seasons was attributed to three rainfall events. Midmore et al. (1996) highlighted a single rainfall event as responsible for catastrophic erosion in a 10-yr period in the highlands of Malaysia, and Paningbatan (1994) reported 24.7% of a total annual soil loss from a single rain event in Laguna province in the Philippines. In our study, a 60.8-mm rainfall event on 8 May 1996 similarly contributed 20% of the total soil loss. During this rain event, all erosion–runoff plots were bare and land was being prepared for transplanting. The second major rainfall event that contributed 18.9% of the total soil loss occurred 27 Oct. 1997. This was a 75.3-mm rainfall event, and the crops were barely 2 wk old during this event. The third major rainfall event occurred in 9 Mar. 1997. This heaviest 1-d (95.3 mm) rain event contributed only 7.8% of the total soil loss. Since crops were more than 1-mo old during this event, crop cover reduced raindrop impact and helped to minimize soil loss.

Contour hedgerows, and to a lesser extent contouring and strip cropping, contained soil loss from these three rain events better than up-and-down planting. This relatively lesser contribution of high-value contour hedgerows (19% of total soil loss was via contour hedgerows) to the total soil loss during these three rain events indicates their effectiveness for soil conservation even during periods of high erosivity. Because of the characteristic large amount of soil loss in a few highly erosive rain events, sufficient consideration should be given to the erosivity of these rain events while developing a soil erosion management plan for such environments.

The traditional farmers' practice of up-and-down cultivation resulted in significantly greater runoff than did contouring and hedgerow planting (Table 2). Data in Table 2 can be expressed as 71.1, 41.1, 57.4, and 48.0 mm of runoff yr-1 for farmers' practice of up-and-down cultivation, contouring, strip cropping, and contour hedgerows, respectively. Contouring and contour hedgerow treatments were able to reduce runoff by 42 and 33%, respectively, compared with up-and-down cultivation. Strip cropping reduced runoff by 20%, compared with the up-and-down system (Table 2), but the difference was not significant. Presbitero et al. (1995) reported annual runoff in their steep slope experiments (50% slope) in Leyte, the Philippines of 55.3, 83.7, 39.8, and 15.6 mm under bare plot, farmers' practice of up-and-down cultivation, alley cropping, and contouring, respectively, which are comparable to our data. As for soil loss, the wider spacing and poor crop cover of tomato resulted in the greatest runoff (Table 4). This suggests that maintaining land cover during the greater part of the year will probably increase soil and water conservation in this environment.

For NO-3 loss in runoff water, differences were not evident between control measures or between crops (Table 2 and 4). Despite the application of chemical fertilizers such as urea for N, NO-3 loss via runoff water was found to decline continuously under all the conservation practices from the first to the last cropping seasons. Compared with the first cropping season, NO-3 loss in the seventh cropping season decreased by as much as 92% in the up-and-down system, a value similar to that in contouring (97%), strip cropping (90%), and the hedgerow system (89%) (Table 3). This may be indicative of a high N uptake by vegetable crops, a decline in N content in surface soil as cultivation continued, or a combination of both.

Sediment Enrichment Ratio
The average ER was 1.2 for organic C, 4.7 for extractable P, 1.7 for Ca, and 1.7 for S. In contrast, the ER was 0.8 for K, 0.8 for Mg, and 0.85 for N. With the exception of K and Mg, the ER results were similar to those reported by Gachene (1986) and Lal (1976). For some Kenyan soils under laboratory conditions, Gachene (1986) reported average ER values of 1.4, 1.5, 1.6, and 1.7 for P, K, Mg, and Ca, respectively. Lal (1976) reported average ER values of 2.4, 1.6, and 5.8 for organic C, N, and P, respectively, for some Nigerian soils. In sandy soils with 5% clay in Thailand, Sombatpanit et al. (1995) reported an ER of >5 for N, 0.9 for available P, and 2.5 for exchangeable K. These workers suggested ER not only depends on the surface accumulation of nutrients but also on soil clay content.

The high enrichment ratio for extractable P indicated its selective removal by erosion. Because these soils have shown large capacities for phosphorus fixation (Poudel, 1998), and P is one of the most yield-limiting nutrients in these soils (Sanyal et al., 1993; Shoji et al., 1993), unwarranted removal of P from the fields requires immediate attention. The phosphate anion is chemisorbed on surfaces by a binuclear bridging mechanism and is nonreversible in its adsorption–desorption behavior (McBride, 1994). Because specific absorption retards the movement of the specifically sorbed ion, specifically sorbed phosphate applied from fertilizer sources tends to accumulate only in the Ap horizons (Nanzyo et al., 1993), a large portion of which is generally lost through soil erosion (Gachene et al., 1997). Therefore, controlling erosion will also control loss of P.

Soil Quality Following Erosion
The values of several fertility parameters such as soil pH, total N, organic C, exchangeable K, and exchangeable Mg showed a notable decline by the end of the experiment (Table 5) , suggesting a degradation in soil quality with cropping seasons. Taking all treatments together, the original soil pH 5.28 (± 0.05) decreased to an average pH of 3.8 (± 0.04) by the end of the seventh cropping season. Exchangeable K and Mg declined from their original values of 0.97 (± 0.06) cmolc kg-1 and 1.43 (± 0.08) cmolc kg-1 to 0.26 (± 0.01) and 0.66 (± 0.04) cmolc kg-1, respectively. Organic C decreased from its initial value of 5.73% (± 0.31) to an average 4.45% (± 0.17), as total N decreased from 0.43% (± 0.07) to an average 0.25% (± 0.02). In contrast, the average level of extractable P at the end of the experiment was almost three times that of the original average extractable P level of 3.67 (± 0.28) mg kg-1 in erosion–runoff plots. Although this decline in pH, Ca, Mg, N, and K can be remedied by fertilizer applications and soil amendments, loss of organic C will be difficult to remedy in the short term.


View this table:
[in this window]
[in a new window]
  		Table 5 Selected chemical properties of original soil surface (0–15 cm), eroded sediment, and final soil surface (0–15 cm) on plots with imposed technologies. Empty spaces indicate that there are no data available. (n = 6)

 
The decline in organic C is one of the major factors affecting the sustainability of agricultural systems (Lefroy et al., 1993; Blair et al., 1995), for decline in organic C affects not only soil chemical and physical properties but also microbial activities in soils. In our study, significant correlations between organic C content and exchangeable Ca (r = 0.59, n = 24), Mg (r = 0.57, n = 24), and K (r = 0.49, n = 24) were observed at the end of the experiment. In addition, soil pH was correlated with Ca (r = 0.59, n = 24) and Mg (r = 0.48, n = 24). These results suggest that a decline in organic C or soil pH results in lower values of these cations.

Despite the fact that erosion was responsible for the loss of a large amount of soil and nutrients from the plots in our experiment, there was no significant relationship between total soil loss for each plot and soil pH, organic C content, exchangeable Ca, Mg, K, N, and P contents at the end of the experiment. This suggests that factors other than soil loss or runoff were also responsible for the observed changes in soil characteristics, at least not in the short-term of these treatments. Therefore, in addition to the adoption of effective soil conservation practices for longer-term fertility decline, appropriate agronomic measures such as liming, residue incorporation, fertilizer application, and replenishment of micronutrients are important to improve and maintain soil productivity on these soils.

Although slight differences were noticed between treatments in the selected fertility parameters at the end of the experiment (Table 5), no effect was statistically significant at P = 0.05. Increasing acidity and nutrient depletion due to soil erosion and inappropriate soil and nutrient management practices are undoubtedly some of the major constraints to the sustainable agricultural production in sloping volcanic ash-derived soils.

Soil Fertility Gradient
With the exception of soil pH and K, large differences were evident by the end of the fifth cropping season for soil fertility parameters between two slope positions (Table 6) . These differences are attributable to soil erosion within the treatments. Soil samples representing upslope positions showed on average 40, 57, 33, 32, and 26% lesser values for organic C, P, total N, Ca, and Mg, respectively, compared with soil samples representing downslope positions. The movement of nutrient-rich topsoil (Table 1) from upslope position and redeposition to the downslope positions was responsible for this observed effect. Similarly, IBSRAM (1995) reported 28 and 42% less soil organic C content and saturated hydraulic conductivity, respectively, in upslope vs. downslope positions in 4 yr in Tanay, in the Philippines. Upslope positions in our study were characterized by exposure of subsoil and 116% more exchangeable Al and 52% more total acidity than downslope positions. Most of the acidity apparently came from exchangeable Al, for it was closely correlated with total acidity (r = 0.96, P <= 0.001). The significant negative correlation between exchangeable Al and organic C (r = -0.90, P <= 0.001), believed to be due to the complexation of Al by organic matter, suggests that with loss of organic C, Al toxicity is likely to rise in these soils. Bulk densities determined from nine surface core samples taken at random using a steel ring core sampler during the third cropping season ranged between 0.79 and 1.07 g cm-3, with slightly greater values for upslope than downslope positions. Infiltration rates were several times lower for upslope than downslope positions (Table 6). Due to these differential infiltration rates down the slope, runoff upslope is likely to occur sooner than downslope, resulting in soil movement from upslope and deposition downslope during runoff events.


View this table:
[in this window]
[in a new window]
  		Table 6 Chemical and physical properties of soils at two slope positions at the end of the fifth cropping season. (n = 8)

 
The traditional farmers' practice of up-and-down slope cultivation showed values lower by 24% for available P, 21% for K, 40% for infiltration rate, 12% for total N, and values higher by 27% for soil acidity, and 19% for Al in the upslope position compared with other technologies. This suggests that the up-and-down cultivation system leads to soil degradation, especially in the upper slope, earlier than for the three conservation measures. The low infiltration rate under the farmers' conventional practice suggests relatively earlier runoff there compared with the other treatments. However, differences between the farmers' practice and the soil conservation measures in the downslope positions were not evident.

The impact of the fertility gradient down the slope was manifested by a large difference in crop yields between the upper and the lower slope positions. The lower half showed, on average, yields greater by 36% for corn, 40% for tomato, and 78% for cabbage compared to the upper half. With the exception of strip cropping, yield differences between the two slope positions for tomato and cabbage were statistically significant for all the technologies being tested (Table 7) . The statistically nonsignificant differences on crop yields between the two slope positions in strip cropping was attributed to the confounded effect of an interchange of beans and vegetables strips in the subsequent cropping seasons. In the fifth cropping season, upslope corn yield was found to be positively correlated with Mg (r = 0.74, P <= 0.05, n = 7) and organic C (r = 0.89, P <= 0.01, n = 7) there. This suggests that a reduction in Mg or organic C or an increase in Al toxicity is likely to reduce corn yields in these soils. This strengthens the suggestion that corn yields will respond to applications of dolomitic rather than calcitic limestone in these soils. Our data show no such correlations for the downslope position.


View this table:
[in this window]
[in a new window]
  		Table 7 Average crop yield for the seven cropping seasons measured at the upper and the lower portions of erosion plots. (n = 14)

 
Large differences in crop yields between the two slope positions, even for the three conservation technologies (Table 7), suggest that management strategies are necessary along the slope transect for uniform crop productivity on sloping land. This finding also indicates that land degradation on steepland occurs on the upper slope first and proceeds down the slope gradient, and that the overall impact of soil erosion on productivity could be large even if soil is not removed from the fields.

Minimizing Soil Movement with High-Value Hedgerows
Rapid development of natural terraces between two contour hedgerows was observed, which apparently helped minimize soil movement down the slope. The contour hedgerows treatment showed relatively less variability in crop yields down the slope across the seven cropping seasons (Table 7). Despite contour hedgerows occupying 10% of the total area and no harvest of asparagus and pigeonpea, an economic analysis of the treatments for the duration of the study showed that economic returns from contour hedgerows were similar to the other conservation measures (data not shown). Without income from asparagus and pigeonpea, gross output (total produce x average price) from contour hedgerows was 11% less than that of contouring, the most profitable treatment. It is believed that with a better management of contour hedgerows (e.g., management specific to the hedgerow species) and realization of income from asparagus and pigeonpea, economic benefits from contour hedgerows will exceed those of contouring. Moreover, enhanced effectiveness of contour hedgerows in containing soil loss across the seven cropping seasons (Table 3) adds to the economic benefit of high-value hedgerows. Planting high-value contour hedgerows that minimize soil movement and erosion and improve economic returns, can help to increase farm income and soil productivity, factors important for long-term sustainability of sloping land commercial vegetable production systems (Poudel et al., 1998).

Managing Natural Terraces in Hedgerows
Natural terraces that formed between the two contour hedgerows showed a large difference in soil quality parameters between their top and the bottom parts (Table 8) . Soil organic C, total N, and P values in the bottom parts were greater by 87, 67, and 161%, respectively, than those of the top parts. In contrast, K was significantly greater in the top part of the terrace, while no significant differences in soil pH, Ca, Mg, Al, and total acidity were evident (Table 8). These differences in soil quality between the top and the bottom parts of the terraces were manifested by large differences in crop yields (Fig. 3) . On the average, the bottom part of the terrace showed 122% greater corn yield than the top part. Although statistically not significant at P = 0.05, tomato yield on the bottom part of the terrace was 50% more than on the top part. The differences in soil qualities and crop yields between the two parts of each terrace suggest that an effective soil productivity management program should be implemented separately on the top and the bottom parts of each terrace if overall productivity and sustainability of vegetable production with high-value contour hedgerows is to be improved.


View this table:
[in this window]
[in a new window]
  		Table 8 Selected physical properties of soils sampled at newly formed terraces in contour hedgerows. (n = 6)

 


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 3 Average crop yields for corn and tomato measured at the top and the bottom half portions of newly formed bioterraces. Vertical bars represent standard errors

 

   Conclusions
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
 Conclusions
REFERENCES
 
Only a few rainfall events were responsible for most of the soil loss measured in this study. These erosive rain events occurred in the months of August to October and March to May. Avoiding land preparation during these periods when highly erosive rain events are common is essential to minimize soil and nutrient losses. Planting of high-value hedgerow species such as asparagus, pineapple, pigeonpea, and lemongrass along the contour; planting on the contour; or planting vegetables up-and-down the slope but with strips of a P. vulgaris crops across the slope reduced soil loss by 31, 42, and 33%, respectively, and runoff by 33, 42, and 20%, respectively, when compared with the farmers' traditional practice of up-and-down planting. Although contour hedgerows, contouring, and strip cropping reduced erosion and runoff to a similar extent, contour hedgerows will in the long-term be superior to strip cropping and contouring since the hedgerows remain in the field and provide effective and continuous barriers for soil movement, especially during periods of land preparation. Because tomato was the crop most conducive to erosion, with {approx}40% more soil loss than from corn or cabbage, additional attention must be paid to the adoption of effective erosion control measures in this crop to minimize soil loss on sloping areas.

Eroded soils were enriched in plant nutrients, especially P and Ca, compared with the original soil. Therefore, effective erosion management strategies are called for to make more efficient use of these plant nutrients. Eroded soil rich in P may lead to downstream pollution if effective erosion control measures are not adopted. Hence, P management appears to be one of the most important factors in relation to the long-term soil productivity and environmental quality of this landscape. The enrichment ratio for organic C was found to be greater than unity across all the cropping seasons, and the average value of soil organic C content for erosion plots declined by 23% from the original soil test value by the end of the seventh cropping season. This continuous decline in organic C, which is apparently a cumulative effect of decomposition and erosion losses, is another major constraint to the sustainability in these cropping systems, and once again efficient soil erosion control measures are called for. Notable deterioration of soil pH and availability of plant nutrients such as N, K, Ca, and Mg were observed by the end of this experiment.

Soil erosion caused appreciable differences in soil quality and productivity down the slope. Soil on the lower slopes was generally more fertile than on the upper slope, and corresponding yield differences were quite large. Crop yields on lower slope positions were greater by 40, 36, and 78%, respectively, for tomato, corn, and cabbage than their counterpart upper-slope yields.

High-value contour hedgerows appear to be the most promising practice to minimize erosion and soil movement down the slope, through formation of natural bioterraces between hedgerows, and through contribution to farm income. However, natural bioterraces between hedgerows showed a marked spatial difference in both soil quality and crop yields from their top to bottom parts. Soils in the bottom part of the terrace showed as much as 87% more organic C content and 161% more available P than soils from the top part of the terrace. Correspondingly, corn and tomato yields were as much as 122 and 50% greater, respectively, on the bottom than on the top part of the terrace. This large difference in soil quality and crop yield across the terrace requires different soil and nutrient management practices for uniform crop yields across newly formed bioterraces under the high-value contour hedgerows system.National Research Council 1991


   ACKNOWLEDGMENTS
 
We acknowledge the SANREM-CRSP funded by the USAID under Grant No. LAG-4198-A-00-2017-00 for supporting this research study. We also acknowledge the help and cooperation of NOMIARC - DA, Mindanao, the Philippines during the field experiment.

Received for publication July 24, 1998.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
 Conclusions
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Poudel, D.D.
Right arrow Articles by West, L.T.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Poudel, D.D.
Right arrow Articles by West, L.T.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Poudel, D.D.
Right arrow Articles by West, L.T.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome