松岡 延浩

The area that was studies and analyzed is in Yulin Prefecture, Shanxi Province of China, located in the transitional morphology from the Maowusu Sandland to the Ocher Plateau. It is half the size of the Maowusu Sandland: 570,000 ha. This study uses Landsat data and climate data to analyze how the land cover has changed as a result of human reforestation on the land that had recently turned to desert. A significant decrease of desert land area occurred in the Yulin region between 2000 and 2005. During the 19 years between 1986 and 2005, approximately 24% of the desert land (about 5000 km2) became grassland, concomitant with the rise of NDVI. These changes were caused not by the influences of precipitation and temperature, but by artificial greening. However, the decreased size of the lakes was attributed to both climate change and human activity.

Severe desertification on the Loess Plateau of China since the 17th century as a result of improper land use has caused critical soil erosion and water shortages in the lower reaches of the Yellow River. To prevent further soil erosion, it is very important to cover the bare land surface with natural vegetation. Evapotranspiration (ET) over the grassland (Stipa bungeana) in Shenmu County of the Loess Plateau during the growing season is estimated to be about 1 mm day(-1), and the ratio of ET to reference Evapotranspiration (ET/ET0) is below 0.3. Evaporation from the bare soil surface simulated by a three-layer soil model is less than ET; that is, ET is 1.5 times greater than evaporation from the soil surface during the growing season. This study examines the relationships among the surface resistance r(s) in the Penman-Monteith equation. solar radiation, vapor-pressure deficit, temperature, wind speed, and soil water content. Surface resistance (r(s)) is strongly affected by the vapor-pressure deficit and soil water content. (c) 2005 Elsevier SAS. All rights reserved.

Spatial and temporal changes of heat and water balances are crucial to planting to combat desertification in the Loess Plateau of China. A three-layer soil model and meteorological data from 43 observatories were used to estimate the beat and water balances of the bare soil surface. Results indicate that the annual mean of sensible heat flux ranged from 13 to 36 W m(-2) and the latent heat flux from 14 W m(-2) (179 mm yr(-1)) to 40 W m(-2) (521 mm yr(-1)). Water retention and restriction of evaporation from the soil surface are affected by the yellow loessial soil including medium loam, light loam and sandy loam, but not heavy loam. The distribution of fractional extractable water and the existing vegetation have been compared to reveal the potential vegetation distribution leading to the aridity distribution in the plateau. (c) 2005 Elsevier Ltd. All rights reserved.

Field experiments were conducted on August 10 and 11 during the summer of 2004 in a maize field to investigate the correlation between stomatal conductance and leaf-specific hydraulic conductance. During these two days, stomatal conductance, transpiration rate, volumetric soil water content and leaf water potential were measured independently. Leaf-specific hydraulic conductance was calculated by the transpiration rate and the water potential difference between soil and leaf. The measured data were checked according to their diurnal variations and their respective standard deviations among samples. Results show that a strong positive correlation exists between leaf stomatal conductance and leaf-specific hydraulic conductance (r2=0.66 and 0.74, respectively on each day). Leaf-specific hydraulic conductance was strongly and positively correlated to transpiration rate on these two days (r2=0.87 and 0.84,respectively). The probable reasons explaining the relationship between hydraulic conductance and stomatal conductance are discussed. It is concluded that their respective relationship to transpiration rate cause the coordination between stomatal conductance and leaf-specific hydraulic conductance under the condition of no soil water stress. These results are expected to: (1) improve models involving plant gas exchange by incorporating the hydraulic constraints into these models currently, most models do not take explicit account of plant hydraulics and, (2) provide a new insight on the mechanism of hydraulic constraints on stomatal conductance and on related plant gas exchange, and hence enforce our understanding on SPAC water relations. © 2005, The Society of Agricultural Meteorology of Japan. All rights reserved.

We have researched environmental monitoring to support anti-desertification activities in the Loess Plateau. In this thesis, the spatial distribution, the variance (stability) and the trend of annual precipitation were studied. In addition, the stability of precipitation in the summer rainy season was estimated from the viewpoint of seasonal features of precipitation. The sampling distribution of the mean of the annual precipitation was resampled by the Bootstrap method at each observation point, and the mean of the annual precipitation was estimated by interval estimation. The time series trend of the annual precipitation was analyzed by using the interval estimation mean of the annual precipitation. The stability of precipitation in the summer rainy season was analyzed from the relationship between the precipitation ratio and the precipitation stability index for the period of five days. It seems that the mean of annual precipitation was the lowest in the Tengeri Desert and the Wulanbu Desert. But the region on the Quilian mountain range at the west side of those deserts near Menyuan had much precipitation. The rainfall increased in low-latitude regions. The variation of annual precipitation was the largest in the northern Maowusu Desert, the northeastern Wulanbu Desert, and around Wugong. In contrast, the variations of annual precipitation were small around Menyuan. We could not see a trend for annual precipitation in the whole of the Loess Plateau from 1980 to 2000. However, it seems that there were significant trends at a few observation points. We proposed the rainy season stability index {RSS(j)} based on the relationship between the precipitation ratio PRP(t,j) and the precipitation stability index PSI(t,y ). The analysis of the feature of precipitation by using RSS(j) clarified that the rainy season is not equally stable between the east side and the west side of the Loess Plateau, even if the annual precipitation is at the same level. © 2004, The Society of Agricultural Meteorology of Japan. All rights reserved.

A three-layer soil model was used for clarifying the heat balance and soil water content in the Loess Plateau, China, and applied to actual bare soil fields (Tottori Sand Dune and Shenmu District, Shaanxi Province, China). The difference between the observed and calculated evaporation per day was approximately 0.18mm (5 Wm-2)to 0.21 mm (6 Wm-2). Because of its moderate mixture of sand, silt and clay, yellow loessial soil in Shenmu affects water retentivity and restriction of evaporation from the soil surface. The seasonal change of heat balance and soil water content was examined using meteorological data in Yulin near Shenmu. The annual means of sensible heat and latent heat flux were 23Wm-2 and 19Wm-2 (239mm per year), respectively. The soil water content of the second and third layers remained comparatively high in winter. © 2004, The Society of Agricultural Meteorology of Japan. All rights reserved.

The present study deals with the relationships between water status parameters of plant leaves and reflectances (R-lambda) at characteristic wavelengths, between 522 and 2450 nm, as well as reflectance ratios, R-lambda/R-1430, R-lambda/R-1650, R-lambda/R-1850, R-lambda/R-1920, and R-lambda/R-1950, based on the air-drying experimental results of soybean (Glycine max Merr.), maize (Zea mays L.), tuliptree (Liriodendron tulipifera L.) and viburnum (Viburnum awabuki K. Koch.) plants. The water status parameters include leaf water content per unit leaf area (LWC), specific leaf water content (SWC), leaf moisture percentage of fresh weight (LMP), relative leaf water content (RWC) and relative leaf moisture percentage on fresh weight basis (RMP). Effective spectral reflectances and reflectance ratios for estimating the LWC, SWC, LMP, RWC and RMP were identified. With these spectral indices, approaches to estimating LWC, RWC and RMP were discussed. Eventually, an attempt on universal formulas was made for estimating the leaf moisture conditions of both herbaceous and woody plants as mentioned above. Moreover, applicability of these formulas was checked with the field experimental results of soybean and maize grown under water and nutrient stresses.

In this study, an environmental variable model for estimating stomatal conductance of maize (Zea mays L.) leaves over a long term is developed, based on research results concerning responses of stomatal conductance to environmental variables. This model is actually a combination of optimized formulae for potential stomatal conductance during daytime, PSC, and for the relative degree of stomatal opening during daytime, RDO, with a basic form expressed as a multiplication of inter-day and intra-day factors. We call this environmental variable model as the combination model. Compared with other models presented in the past, this combination model is convenient in practical use because it can estimate not only maximum and average values but also instantaneous values of stomatal conductance during daytime. Also, the theoretical description of this model has been clarified by introducing into it the concept of environmental stress. As compared with Jarvis-type multiplication environmental variable models, we think this combination model is a refinement and development of the Jarvis-type models. (C) 1998 Elsevier Science B.V. All rights reserved.

When a crop is grown in rows, the row direction influences the energy balance and moisture budget of the soil-canopy system of the crop. Attempts were made to compare energy balance and moisture budget of E-W row soybeans with those of N-S rows at partial cover during the summer of 1996. The energy balance and moisture budget were considered for the soil and canopy separately. Daily fluctuations and diurnal patterns of energy and water balance of field, soil and canopy were strongly influenced by row orientations. Average latent heat flux density (LE) observed in N-S row soybean field was 6.6% larger as compared to E-W row soybean field. Average ratios of -LE to net radiation (Rn) were found 0.85 and 0.87 for E-W and N-S rows, respectively. Soil heat flux (G) accounted for 14 to 20% of Rn for E-W rows and 6 to 9% for N-S rows. Soil surface net radiation (Rns) of E-W rows was about 25% larger than that of N-S rows. Latent heat flux from the soil (LES) was about 13% larger in E-W rows as compared to N-S rows. The -LES/Rns ratios often exceeded 1.0 in both the cases. The sensible heat flux from the soil showed a positive trend in both the cases indicating soil surface was absorbing convective heat. The -G/Rns ratios were found 0.41 and 0.26 for E-W and N-S rows, respectively. Average net radiation of E-W row canopy (Rnc) was 20% smaller as compared to N-S row canopy. Latent heat flux from the E-W row canopy (LEc) was 19% smaller as compared to N-S row canopy. The -LEc/Rnc ratios were found 0.80 and 0.79 for E-W and N-S rows, respectively. Sensible heat flux from the canopy showed negative trends in both the cases. Results suggest that only field energy balance measurements would not provide sufficient information on how energy balances of the soil and canopy are partitioned. © 1997, The Society of Agricultural Meteorology of Japan. All rights reserved.

To simulate the movement of water and to understand its working mechanism in the soilplant-atmosphere continuum (SPAC), studies on the distribution characteristics of resistance to water flow in the SPAC are important. The present paper discusses the spatial distribution and variability of resistance to water flow in the SPAC based on field experiments on maize and soybean at Shenyang, China, and Matsudo, Japan. Under the experimental conditions employed, the total conductive resistance to water flow from root xylem to leaf xylem (R_RL) in maize and soybean plants was 85.6〜93.5% of the total resistance to water flow from soil to leaf xylem (R_SL), and the total absorptive resistance for the root system (R_SR) was 6.5〜14.4% of R_SL. In addition, soil resistance, soil-root contact resistance, root absorptive resistance, and xylem resistance for root and stalk were 2.7〜5.0%, 0.9〜1.1%, 2.6〜8.4% and 10.1〜73.6%, and 20.0〜75.5% of R_SL, respectively. Meanwhile, the rates of bulk boundary resistance (R_AV) and bulk stomatal resistance (R_ST) for the canopy relative to the total transpiration resistance (R_LA) were 51.5% and 48.5%, respectively. Aerodynamic resistance to momentum transfer, excess resistance to water vapor exchange, and leaf boundary resistance were 15.6%, 9.2%, 23.6% of R_LA, respectively. Therefore, the total resistance in stomata and boundary of the leaf was about 75% of R_LA. The leaf stomatal resistance shows different patterns of diurnal variation in various environmental regimes, and a significant seasonal variation in the growth period. The stomatal resistance increased along with drying of the soil, and its value for the adaxial surface was larger than that for the abaxial surface. The ratio between the two surfaces varied under different environmental conditions.

We measured CO₂ concentration in a deciduous forest canopy and compared it with wind speed, net radiation and soil respiration. On a clear day, minimum CO₂ concentration was observed at the height where the canopy was the densest. Maximum concentration was observed just above the ground before sunrise on a calm day. The concentration tended to be higher near the ground than near the top of the canopy on all days and this concentration gradient was steeper at night, resulting from the high respiration rate of the soil and from the small transport of CO₂ out of the canopy due to low wind. This suggests that the respiration rate of the soil should be considered when measuring assimilation and respiration rate of plants by the aerodynamic method.

From September 27 to 28, 1991, Typhoon 9119 struck Kyushu Island, Japan, and wind damage to trees occurred around the northern part of this area. In the forests, we used two scenes of Landsat TM data, obtained on September 21, 1990 (before the damage) and March 5, 1993 (after the damage). The ground truth datum of the damaged forests is “Damaged forest map”, which illustrates the fraction of damaged trees (100%, 80%, 60% and others). Specifying the location of the forest, we adopted the maximum likelihood method for the TM data. To assess the intensity of the damage, we calculated a Normalized Difference Vegetation Index (NDVI). Before the damage, there were smaller differences in the averaged reflectance of each band and NDVI than after the damage among the damaged forest classes (100%, 80%, 60% and others). After the damage, there were differences in the average reflectance of band 5 (1.55 to 1.75 um) and band 7 (2.08 to 2.35 um) among the damaged forest classes. Therefore band 5 and band 7 were useful to detect damaged forests. Since the reflectance of the band 5 and band 7 were increased after the damage and the reflectance of both band 5 and band 7 are related to the water contents of the plants we presume that the damaged forests suffered water stress. © 1997, The Society of Agricultural Meteorology of Japan. All rights reserved.

When crops are grown in rows, differential shading on the soil surface between rows due to row-sun geometry may result in spatial variations in the micrometeorology of row crops. A field experiment was conducted during the autumn of 1995 in the farm of the Faculty of Horticulture, Chiba University (the latitude 35°C'N and the longitude 139°C54'E) to examine spatial variations of some selected micrometeorologicalfactors in the soil beneath the potato-row canopy at partial cover. Results of the study indicate that the magnitudes and patterns of micrometeorological factors at different locations observed between N-S rows were fully different from the micrometeorology observed between E-W rows. The patterns of micrometeorological factors at each position were associated with the patterns of soil irradiance, which were the interactions of row-sun geometry. Soil irradiance, net radiation and soil heat flux between N-S rows were three to five time larger than those in E-W rows. Daily fluctuations of soil irradiance, net radiation, soil heat flux, soil and air temperatures at east, center and west locationsbetween N-S rows were three to five times higher than those measured directly beneath the canopy. On the other hand, daily fluctuations of soil irradiance, net radiation, soil heat flux, soiland air temperature were higher directly beneath the canopy of E-W rows as compared to the magnitude and patterns observed at other locations. Temperature differences between the sunlit and shaded portions of the surface of N-S rows exceeded 10-15°C. Air temperatures measured at 5cm above each position, were also dependent onthe position with respect to the N-S and E-W plant rows. © 1996, The Society of Agricultural Meteorology of Japan. All rights reserved.

Air temperature, sunshine duration and precipitation are used to estimate the soil moisture in the shallow root zone. The model for the estimation of the soil moisture is based on the water balance of the root zone. The daily amount of evapotranspiration was estimated using the Priestley and Taylor model, modified by the ratio of total soil moisture and soil moisture at the field capacity in a given soil layer. The upward capillary flow was evaluated from the unsaturated hydraulic conductivity of soil and the gradient of water potential between the center of the root zone and at a depth 100 cm from the center. This model was e Abash,zursi edd isttor icet stimina teH othkek aisdooi,l Jmapoains.ture in the 40cm deep root zone from surface in the Abashiri district in Hokkaido,Japan Estimated soil moisture was in agreement with the measured soil moisture by tensiometer in the range of PF 1.8 to PF 2.8. The amount of upward capillary flow was higher than the expected values. Therefore, we concluded that the soil moisture could not be estimated precisely without the estimation of the upward capillary flow. © 1993, The Society of Agricultural Meteorology of Japan. All rights reserved.

Shortage of water resource in Japan requires efficient uses of water. For systemetical water use, we designed mesh data of soil moisture content for Shamo area in Hokkaido island, where is a district with little rain, therefore irrigation is indispensable for farming. This study is divided into two parts; the one is planning of meteorological mesh data and the other is evaluation of mesh data for soil moisture contents. First, using AMeDAS meteorological data and numerical terrestrial data, we constructed multiple regression equations. Their dependent variables are the meteorological data such as temperature, duration of sunshine and precipitation, and their independent variables are geographical factors such as longitude, latitude and altitude. The estimation of temperature and duration of sunshine for AMeDAS observation points have small errors, but that of precipitation has large error as much as 45mm at its maximum. Next, we made mesh data of soil moisture contents. To estimate soil moisture contents, method by Nakayama (1991) was adopted. The mean prediction error was 6.4mm and its standard deviation was 14.3mm for shamo area. Therefore this mesh data was appropriate for the planning the irrigation.

Air temperature, sunshine duration and precipitation are used to estimate the soil moisture in the shallow root zone. The model for the estimation of the soil moisture is based on the water balance of the root zone. The daily amount of evapotranspiration was estimated using the Priestley and Taylor model, modified by the ratio of total soil moisture and soil moisture at the field capacity in a given soil layer. The upward capillary flow was evaluated from the unsaturated hydraulic conductivity of soil and the gradient of water potential between the center of the root zone and at a depth 100 cm from the center. This model was e Abash,zursi edd isttor icet stimina teH othkek aisdooi,l Jmapoains.ture in the 40cm deep root zone from surface in the Abashiri district in Hokkaido,Japan Estimated soil moisture was in agreement with the measured soil moisture by tensiometer in the range of PF 1.8 to PF 2.8. The amount of upward capillary flow was higher than the expected values. Therefore, we concluded that the soil moisture could not be estimated precisely without the estimation of the upward capillary flow. © 1993, The Society of Agricultural Meteorology of Japan. All rights reserved.