NITROGEN RESORPTION AND NITROGEN USE EFFICIENCY

VI. NITROGEN RESORPTION AND NITROGEN USE EFFICIENCY

IN CACAO AGROFORESTRY SYSTEMS MANAGED DIFFERENTLY INTRODUCTION Agricultural and forestlands in the tropics have been subjected to unprecendented development pressures for the past several decades Collier et al. 1994; Janzen 1998. Millions of hectares of primary forest have been degraded by logging Putz et al. 2000 and millions more converted into intensive agricultural areas Lenne and Wood 1999. At the same time, traditional agroforestry, perennial and long-fallow shifting cultivation systems i.e., forest farming have been displaced by technified monocultures Thrupp 1998. The role and importance of agricultural lands, particularly traditional forest farming systems, have been neglected in global biodiversity conservation efforts Thrupp 1998; Fox et al. 2000. In closed-canopy forests both external micro-climatologically factors and plant metabolic processes vary from the top of the canopy to the forest floor. As a consequence, one essentially always finds pronounced vertical gradients in the physiology of foliage in the forest canopies Ishii et al. 2000; Lewis et al. 2000. The foliage also responds to factors that vary through the canopy, such as foliar nitrogen and translocation or resorption of nitrogen Ishii et al. 2000; McDowell et al. 2000. An agroforestry system such as shade-grown cacao Theobroma cacao is one alternative land use to slash-and-burn agriculture and is recommended for sustainable development in the humid tropics Johns 1999. Because cacao is shade- tolerant and can grow under a canopy of trees maintained, regenerated, or replanted from the traditional tropical forest Beer et al. 1998, its inclusion into a forest agro- ecosystem can maintain biodiversity and ecological services provided by the native forest Rice and Greenberg 2000; Glor et al. 2001. Light and nitrogen are two of the most important resources for plant life. Light is the energy source with which plants fix CO 2 through photosynthesis. More than 50 of leaf nitrogen is directly related to photosynthesis and strong correlations have been demonstrated between both Hirose and Bazzaz 1998. Nitrogen is a major constituent of chlorophyll and is involved in carboxylating enzymes of photosynthesis, especially Rubisco Waring and Schlesinger 1985. An observed difference in N concentration of foliage tissue generally reflects changes in enzyme concentration McGuire et al. 1995. Foliar nitrogen depends on many variables, including soil nitrogen mineralization and nitrification rates, soil CN ratio, species, temperature and irradiance Yin 1994, MaGill et al. 1996, Ollinger et al. 2002. Species traits, nitrogen availability, and climate are three control factors that likely determine canopy foliar nitrogen concentration Pan et al. 2004. Nutrient resorption is a process in which nutrients are withdrawn from leaves prior abscission and redeployed in developing tissues. The resorption of nutrients prior to abscission is one of the key processes by which plants conserve them. Resorption may occur throughout a leaf’s life, particularly as leaves become progressively shaded Ackerly and Bazzaz 1995, Hikosaka 1996. This process reduces the likelihood of nutrient loss in litter dropped on the forest floor Bormann et al. 1977 and subsequently, the withdrawn nutrients are redeployed in new tissue, such as leaves and reproductive structures or stored for later use Wright and Westoby 2003. On average, plants withdraw about 5-80 of leaf nitrogen via resorption. The proportion of nutrients withdrawn from leaves the resorption efficiency varies widely between species. For example, less than 5 up to 80 of leaf nitrogen, and 0 up to 95 of leaf phosphate, may be resorbed Aerts 1996, Aerts and Chapin 2000; depending on species and environmental conditions Lambers et al. 1998. In infertile habitats, nitrogen in senescent leaves is reduced to lower levels than in fertile habitats Wright and Westoby 2003. The efficiency of plant nutrient use is simply the inverse of nutrient concentration in senescent leaves. Consequently, plants from nutrient-poor habitats are typically more efficient in their N use than plants from nutrient-rich habitats Vitousek 1982, Tateno and Kawaguchi 2002. In ecosystems scale, the efficiency of plant nutrient use or nutrient efficiency use NUE in grams in organic matter produced per unit of nutrient taken up is simply the inverse of nutrient concentration in plant tissue, and that consequently plants from nutrient-poor habitats appear to be more efficient than plants from nutrient-rich habitats Vitousek 1982, as evidence by less nutrients return to soil in litterfall Vitousek 1982 and 1984, Cuevas and Medina 1986, Silver 1994. The nutrient use efficiency for a forest is therefore the inverse of nutrient concentration in the aboveground litterfall, root turnover, and the organic matter increment of the vegetation. NUE as an index that helps explain species distributions across landscapes that vary in soil fertility and other resources Rundel 1982, Vitousek 1982, Schlesinger et al. 1989. Nevertheless, in some experimental studies lower NUE were found in poor soils than in rich soils Chapin and Kedrowsky 1983; Lüttge et al. 1991, although higher NUEs have been more often found in nutrient poor environments than in rich ones Birk and Vitousek 1986; Bridgham et al. 1995. Sometimes NUE was similar at different nutrient supply levels Folk and Grossnicke 2000. The NUE is difficult to calculate due to the great number of variables that determine it. Consequently, the results often depend on the method employed in its determination. A lot of different equation and methods have been employed to determine NUE Vitousek 1982; Aerts 1989; Kitayama et al. 2000. Higher NUE values have been found in poorer environments in studies which have compared different conditions for a single species and have calculated the NUE as litterfall production divided by the nutrient lost in the litterfall across s gradient of nutrient availability Vitousek 1982; Bridgham et al. 1995. In this study, fully-developed cacao leaves and senescent cacao leaves, soil samples, and litterfall were collected from three types of cacao agroforestry systems over one year. The objective of this study was to quantify N NUE at ecosystem scale and at the cacao tree, nitrogen resorption in cacao leaves and asses their implication for cacao production. MATERIALS AND METHODS Experimental set-up The experiment began in March 2005 and ended in March 2006. Ten litter traps 1 m x 1 m, 50 cm high above the ground were installed at 20 subplots on each plot. Litter was collected at monthly intervals for one year March 2005 to February 2006. The litter was taken to the laboratory and dried to constant dry weight at 80ºC. The total litterfall was obtained by weighing the samples. Nitrogen content in litterfall was determined with Kjeldahl technique. Leaves were collected three-monthly interval over one year March 2005 to March 2006, from three types of cacao agroforestry systems. At each collecting time, three to five leaves samples fully-developed leaf and senescent leaf were collected from five cacao trees inside each plot. Cacao senescent leaf was easily identified as they exhibited yellowish color, different from green living leaves, and could be dropped by shaking the branch. Fully-developed leaf and senescent leaf from each sampling period were thoroughly washed and air dried and then kept in an oven at 60ºC for 48 hours. The dried leaves were ground, stored in paper bags for chemical analyses. Leaf nitrogen concentration was determined with Kjeldhal technique. Soil material was also collected at three- monthly intervals over one year using a soil core down to 15-20 cm depth. At each collecting time, twenty-five of soil samples were collected randomly in each plot. For determination of the total soil N content, soil samples were stored in sealed plastic bag and stored at 4ºC until analysis. The total soil N content was determined with Kjeldahl technique. Nitrogen use efficiency N NUE C in cacao plants in the agroforestry systems were calculated according to Vitousek 1982 and Tateno and Kawaguchi 2002: N NUE C = 1 nitrogen concentration in senescent leaf Nitrogen use efficiency ecosystem scale - N NUE Es in cacao agroforestry systems and in the natural forest as a control was calculated according to Vitousek 1982: N NUE Es = Litterfall production g m -2 y -1 N content in litterfall g m -2 y -1 The natural forest NF was used as the undisturbed ecosystem to compare to N NUE Es with cacao agroforestry systems. The proportional nitrogen resorption was calculated according to Sharma and Pande 1986: Resorption = N in fully-developed leaf – N in senescent leaf X 100 N in fully-developed leaf Statistical analyses Data on annual litterfall, foliar nitrogen content, N resorption, soil nitrogen content, N NUE C and that on N NUE Es in cacao agroforestry systems were analyzed using general linear models GLMs. All post hoc tests were carried out using Tukey-test. All analyses were done using SPSS 13.0 software. The relationship between parameters was analyzed using regression and Pearson’s correlation analysis. 64 RESULTS The cacao agroforestry systems investigated in this study were in the local people’s plantation at the margin of LLNP. These plantations were never fertilized. Means of annual litterfall, cacao foliar nitrogen content in fully-developed and senescent leaf, nitrogen resorption, soil nitrogen content, and that of N NUE C in study sites were presented in Table 6.1. Means of annual litterfall p = 0.02, cacao foliar nitrogen content in fully-developed leaf and senescent leaf p = 0.004, nitrogen resorption p = 0.03, and that of N NUE C p = 0.03 differed significantly between the three agroforestry systems, but those of soil nitrogen content p = 0.07 did not. CF2 had a mean of annual litterfall of 497.7 g m -2 y -1 which was not significantly different from that of CF1 but significantly lower than that of CP. Foliar nitrogen content of cacao fully-developed leaves under CF2 1.2 was not different from that of CF1 1.3 but lower than that of CP 1.5. CF2 had the highest value of N NUE C 97.3. Table 6.1. Mean of annual litterfall, cacao foliar N, N resorption, soil N content, and N NUE C in the cacao agroforestry systems. Study site Slope Annual litterfall g m -2 y -1 N in fully- developed leaves N in senescent leaves N resorption Soil N content mg kg -1 N NUE C CF1 70 693.4±21.9 ab 1.3±0.05 ab 1.1±0.04 ab 17.6±2.5 a 2209.1±153.2 a 91.1± 7.4 a CF2 35 497.7±27.7 a 1.2±0.05 a 1.0±0.04 a 12.7±2.0 a 1840.2±133.5 a 97.3± 5.3 a CP 20 822.9±37.4 b 1.5±0.07 b 1.3±0.06 b 14.4±1.8 a 1921.5±136.7 a 80.7±12.0 a Different superscripts in the same column indicate significant differences among mean p 0.05, Tukey-test. Data are means ± SE. Relationships between cacao foliar nitrogen in fully-developed leaf content and canopy cover and that of cacao foliar nitrogen content and temperature revealed different pattern Figure 6.1. The relationship between foliar nitrogen content and canopy cover showed a negative correlation Pearson’s r = - 0.36, while that of foliar nitrogen content and temperature showed a positive correlation Pearson’s r = 0.34, indicating that cacao foliar nitrogen content decreased with increasing canopy cover, increased with decreasing temperature. Increasing canopy 65 cover would decrease temperature in the forest. Canopy cover was not related to cacao foliar nitrogen content r 2 = 0.13. Figure 6.1. Cacao foliar nitrogen in relation to a temperature C and b canopy cover in cacao agroforestry systems. The relationships between soil nitrogen content and foliar nitrogen content in cacao agroforestry systems CF1, CF2, and CP although showed an increasing trend were not significant, however correlation between soil nitrogen content and cacao foliar nitrogen content was significant in cacao agroforestry systems CF1 and CF2 Pearson’s r = 0.58 and 0.5, p 0.01, Figure 6.2. Figure 6.2. Cacao foliar N in relation to N soil content mg kg -1 in the cacao agroforestry systems CF1, CF2, and CP. 23.00 22.75 22.50 22.25 22.00 21.75 21.50 Temperature o C 2.25 2.00 1.75 1.50 1.25 1.00 0.75 r 2 = 0.12 r = 0.34, p = 0.005, p0.01 C ac ao fo li ar N co n te n t CF1 CF2 CP 75 70 65 60 55 50 45 Canopy cover 2.25 2.00 1.75 1.50 1.25 1.00 0.75 r 2 = 0.13 r = - 0.36, p = 0.008, p0.01 C ac ao fol ia r N co nt ent CP CF2 CF1 1400 1200 1000 800 600 400 200 Soil N content mg kg -1 2.25 2.00 1.75 1.50 1.25 1.00 0.75 C ac ao fol ia r N Fit line for CP Fit line for CF2 Fit line for CF1 CP CF2 CF1 r 2 = 0.34 r 2 = 0.05 r 2 = 0.26 Relationship between nitrogen resorption and soil nitrogen content was shown in Figure 6.3. There was no relationship between soil nitrogen content and nitrogen resorption in all study sites. However, the correlation between soil nitrogen content and nitrogen resorption was significant only in cacao agroforestry CF2 Pearson’s r = 0.65, p 0.05. Figure 6.3. Nitrogen resorption in relation to soil nitrogen content mg kg -1 in the cacao agroforestry systems CF1, CF2, and CP. The efficiency of a plant utilizing soil nutrients for producing biomass may indicate its capability to optimize growth at a given soil nutrient content. The relationship between cacao N NUE C in the cacao agroforestry systems and soil nitrogen content is shown in Figure 6.4. There is a trend of decreasing N NUEc with increasing soil nitrogen content in cacao agroforestry systems, although the relationship is not significant. For N NUE Es in the ecosystem scale, we found that the natural forest NF as the undisturbed ecosystem was not different from CF1 nor CF2, but significantly higher p 0.01 than CP. The N NUE Es in CP 63.8 was lower than the other ecosystems CF1 = 108.0, CF2 = 109.6, and NF = 94.9 Table 6.2. 1400 1200 1000 800 600 400 200 Soil nitrogen content mg kg -1 50 40 30 20 10 N re sor pt ion Fit line for CP Fit line for CF2 Fit line for CF1 CP CF2 CF1 r 2 = 0.007 r 2 = 0.43 r 2 = 0.03 Figure 6.4. Nitrogen use efficiency of cacao N NUE C in relation to soil nitrogen content g kg -1 in cacao agroforestry systems CF1, CF2, and CP. Table 6.2. Nitrogen use efficiency N NUE Es in natural forest NF and cacao agroforestry systems CF1, CF2, and CP. Land use types Annual litterfall g m -2 y -1 N content in litterfall N NUE Es NF 1367.4±32.5 c 1.15±0.2 ab 94.9±12.7 b CF1 693.4±21.9 ab 0.92±0.05 a 108.0± 6.7 b CF2 497.7±27.7 a 0.89±0.08 a 109.6± 9.4 b CP 822.9±37.4 b 1.62±0.15 b 63.8± 7.6 a Different superscripts in the same column indicate significant differences among mean p 0.05, Tukey-test. Data are means ± SE. 1200 1000 800 600 400 200 Soil N content 120 110 100 90 80 70 60 N N U E c Fit line for CP Fit line for CF2 Fit line for CF1 CP CF2 CF1 r 2 = 0.17 r 2 = 0.07 r 2 = 0.1 DISCUSSIONS The canopy cover in the cacao agroforestry system CP 49.8 was lower than that in CF2 69.4, and that of CF1 72.1. The lower canopy cover is equivalent to a higher light intensity and also higher temperature. Wood and Lass 1985 reported that a higher light intensity is a very important factor for cacao productivity. If nitrogen is not limited, a higher light intensity linearly increases the carbon gain of the cacao plants leading to a high biomass production. This is indicated by a higher p = 0.02 litterfall under CP 822.9 g m -2 y -1 than that under CF2 497.7 g m -2 y -1 , and similar to that under CF1 693.4 g m -2 y -1 . This high biomass production coincided with an elevated supply rate of nitrogen as reflected by the fact that foliar nitrogen content of cacao under CP 1.5 was higher p = 0.004 than that under CF2 1.2 and similar to that under CF1 1.3. The relationship between canopy cover and foliar nitrogen content was significant Pearson’s r = - 0.36 and showed that a greater canopy cover reduces foliar nitrogen content of cacao tree Figure 1. This finding is similar to that reported by Tateno and Kawaguchi 2002 and Leal and Thomas 2003 where high light intensity requires a high supply of nitrogen reflected in the foliar nitrogen content of mature leaf as a manifestation of internal nitrogen demand. The plant internal nitrogen demand may be satisfied by nitrogen taken up from the soil andor by nitrogen resorbed from senescent leaf. When soil nitrogen content is high, plants generally prefer to absorb nitrogen from soil, while a high degree of nitrogen resorption is the rule when soil nitrogen content is low Baddeley et al. 1994; Bowman 1994; Rapp et al. 1999; Tateno and Kawaguchi 2002; Wright and Westoby 2003; Singh et al. 2005. Cacao trees cultivated under CP agroforestry system satisfied its internal nitrogen demand by both nitrogen soil uptake and nitrogen resorption. Gliricidia sepium, the nitrogen-fixing shade tree planted in the CP agroforestry system contributes a good amount of litter, and by decomposition, the soil nitrogen content for plants including cacao trees may improve. Therefore, the cacao trees in this CP agroforestry system benefited from a slight improvement of soil nitrogen content. However, even though the mean of soil nitrogen content under CP agroforestry was somewhat higher than that in CF1 or CF2, the differences were not significant. On the other hand there was a significant relationship between cacao foliar nitrogen content and soil nitrogen content showing an increasing foliar nitrogen content with an increasing of soil nitrogen content, mainly in cacao tree growing under CF1 and CF2 Figure 6.2. This type of relationship was similar to that reported by Singh et al. 2005. Cacao tree growing under CP agroforestry systems showed a nitrogen resorption of only 14.4 which was less than that of CF1. Cacao agroforestry under CF1 was located in a foothill with about 80 inclination. In this steeply sloping condition surface runoff was considerable with the consequence that nutrients were washed out into the river below thereby reducing soil nutrient content in the cacao plantation. This situation apparently forced cacao plants under CF1 agroforestry system to rely more on a higher resorption of nitrogen 17.6. The cultivation of cacao trees under the CF2 regimes resulted in less litterfall, i.e. 497.7 g m -2 y -1 . The cacao trees especially under CF2 agroforestry system showed a considerable higher weed infestation than under CF1 or CP agroforestry systems. It seemed that the cacao trees under CF2 system suffered a strong competition from weeds. We assumed that the weeds accumulated considerable amounts of nitrogen and other nutrients, thus preventing cacao trees from accumulating a higher amount of biomass or litterfall. With a low amount of litterfall, these cacao trees apparently required a smaller supply rate of nitrogen, what would explain a relatively small resorption rate 12.7 which seemed to be sufficient to satisfy the internal demand of nitrogen. The nitrogen resorption of cacao trees under CF2 agroforestry system did not differ significantly from that under CP. It may be taken as an indication that, with respect to nitrogen root uptake, the conditions in both systems were similar, i.e. sufficient regarding soil nitrogen content. However, the internal nitrogen demand differed as a consequence of different litterfall production. The relationship between nitrogen resorption and soil nitrogen content in this study indicated a peculiar relationship, where nitrogen resorption increased with increasing soil nitrogen content. The CF1 plot was located under the canopy of the remaining forest, so the density of shrubs and grasses was small, while the plot of the CP agroforestry system was located in a well maintained cacao plantation using G. sepium as the shading trees, weed infested by weeds during the measurement at various degree with the heaviest infestation occurring under CF2. Although the soil nitrogen content was quite high, it appears that competition with weeds reduced the nitrogen available to the cacao trees which may have promoted a higher foliar nitrogen resorption rate. The nitrogen use efficiency of cacao trees growing under CP agroforestry was lower than that under CF2. This finding supports Tateno and Kawaguchi 2002 that plants with high foliar nitrogen content tends to have lower N NUE. Comparing the means of N NUE Es at the ecosystem scale among the agroforestry systems and with the natural forest, the results showed that CF1 and CF2 agroforestry systems did not differ from that of natural forest, while N NUE Es of CP agroforestry was significantly lower than in the other systems. This finding agreed with the findings of Smith et al. 1998 Tateno and Kawaguchi 2002 that N NUE Es of legume trees as a nitrogen-fixing species is lower than N NUE Es in the natural forest. The values of nitrogen use efficiency in this work are lower than those reported by Ankilla et al. 2002 in Costa Rica low-land forest, but higher than those reported by Smith et al. 1998 in Curuá-Una, Brazil Amazon. Agroforestry systems of cacao established by planting cacao seedling under a forest canopy following the cutting of selected forest tress or planting cacao seedling under planted trees were similar in their N NUE Es to nearby natural forest in Sulawesi. CONCLUSIONS Nitrogen use efficiency at cacao tree in different agroforestry systems were similar, whereas, nitrogen use efficiency at ecosystem under CF1 and CF2 were similar to NF and higher than those under CP. Nitrogen resorption at cacao tree in different agroforestry system were similar. Reducing the cover by shade trees in cacao plantations increases foliar N concentration and, thus, may be one reason for a higher productivity of cacao in unshaded systems. However, more light also promotes weed infestation and thus competition for nitrogen.

VII. GENERAL DISCUSSIONS