臺灣中部土壤特性與稻穀收量關係調查報告

A Report on Soil Characteristics Associated with Rice Yield in the Central Taiwan

林家棻(C. F. Lin) ； 陳春泉(C. C. Chen) ； 李顯琨(H. K. Li) ； 王幼惠(Y. H. Wong)

《中華農業研究》 23卷3期 (1974/09) Pp. 155-165

https://doi.org/10.29951/JARC.197409.0001

摘要

本調查目的乃企圖在中部類似氣候與集約耕作下，探讨不同土壤類别間，稻谷產量與其產量構成因子以及表土肥力之關係，藉供探索提高單位面積稻谷產量的途徑，提供土壤管理之参考。
本調查係從民國61年第一期作收穫時舉行，調查地區爲溪州、田尾、北斗與西螺等4鄉鎮，計400户，每户採取20欉稻穗調查其稻谷收量舆産量構成因子，同時採取土壤剖面及表土樣本。經整理後，得土壤樣本398個，其中屬於粘板岩新冲積土者123個，粘板岩老冲積土者275個，分屬於26土系。稻穗樣本經整理校對後，選取324個。兹將所得結果摘要於下：
1．無論粘板岩新冲積土或老冲積土稻田，稻谷產量均與結實率或千粒重呈極顯著之相關係数。又該地區第一期作收穫時，稻株全倒及半倒者佔50%以上，倒伏導致結實率與千粒重偏低，因此欲提高本地區第一期作稻谷產量，應着重稻田氮肥的管理。
2．粘板岩老冲積土的土壤肥力（包括有機物，C. E. C.，有效態氮磷鉀含量）舆稻谷產量與粘板岩新冲積土者比較，差異達到顯著平準。由此顯示此二類土壤性質有所不同，而在集約耕作，施用足量肥料之下，土壤肥力仍佔重要地位。
3．在粘板岩新冲積土或老冲積土中，其排水良者與排水不完全者間之稻谷産量、産量構成因子以及肥力因子之差異均不顯著。由此提示排水良與排水不完全，在土壤剖面形態上確有顯明差異，且老冲積土之肥力，略示受排水情形影響之趨势，故此特性可作土系分類根據，但對稻谷產量關係尚難榷定。
4．在粘板岩新冲積土中，二水系之稻谷產量與結實率確較镇平系或潮洋厝系爲高，而其土壤肥力因子並無顯明差異，但與土壤剖面質地有關。在粘板岩老冲積土中，二林系及鹿港系之稻谷產量與每穗粒數顯较管嶼厝爲高，土壤肥力因子（有機物，C. E. C.，有效態氮及磷）中亦以前兩者高於後者，且後者質地較粗。由此顯示，在粘板岩新冲積土中，二水系稻谷收量高於鎮平或潮洋厝系者，可能受土壤剖面質地及人爲管理因子的影響較土壤肥力爲大；而在粘板岩老冲積土中，二林、鹿港兩土系之稻谷收量高於管嶼借厝系者，可能受土壤肥力因子及土壤剖面質地的影響較管理因子爲大。
5．一般土壤調查只重视調查當時土壤剖面形態的差異，而土壤類别間稻谷産量的高低，则多未兼顧，舉行本調查可補其不足。似有配合土壤調查，績作類此調查之必要。

並列摘要

The aim of this survey was in an attempt to find out the relationship between soil characteristics and rice yield under similar environments and intensive farming in the cent ral Taiwan. The results will be useful in ascertaining the soil factors in rice production and the possible ways of increasing rice production in the area surveyed.
This survey was conducted in Changhua and Yunlin county in the first crop of 1972. A total of 400 field plots were selected and surveyed. In each plot, twenty hills of rice samples were harvested and the harvested panicles were also collected for laboratory studies including the estimation of per-hectare yield and the yield components. In addition, examination of soil profile and collection of surface soil sample for chemical analysis were also conducted on each site.
The results may he summarized as follows:
1. It was found that the correlation coefficients between rice yield and percentage of filled grain as well as 1000-grain weight were remarkably significant on the rice samples taken from slate recent alluvial and slate older alluvial soils. It was also found that rice crop at harvest in the area lodged more than 50 per cent of field plots surveyed. Lodging caused reduction in percentage of filled grain and 1000-grain weight. To prevent lodging by proper management of N-fertilization seems to be one of the most important factors for increasing rice production in the area.
2. It was found that rice yield per unit area, and soil ferility level from slate older alluvial soils were higher than those from slate recent alluvial soils. Their differences were significant through Fisher's t-lest. It indicated that soil fertility level played an important role in rice production, even under sufficient fertilization.
3. The drainage factor ranged from well in drainage to imperfect in drainage did not make significant difference in rice yield, yield components, and soil fertility level, though it caused remarkable difference in morphological features in the soil profile.
4. The number of grains per panicle, grain yield, and soil fertility level selected from slate older alluvial soils were higher in Erhlin and Lukang series, and lower in Kuanhsutseu series. Their differences were significant through Fisher's t-test. The percentage of filled grain and grain yield selected from slate recent alluvial soils were higher in Erhshai series than in Chenpin series or Tsoyangtseu series. However, the difference in soil fertility level was insignificant among them. It seems to indicate that the differences in grain yield in Erhlin, Lukang and Kuanhsutseu series were due probably to soil factor rather than management factor. While a reverse was in the case of Erhshai, Chenpin and Tsoyangtseu series of slate recent alluvial soils.
5. Low yield data were often obtained from soil series with coarse texture in profile (Chenpin, Tsoyangtseu, and Kuanhsutseu series).