重庆大学AFM发文揭示稻田夜间甲烷峰值通量的日变化模式

题目：

Nocturnal peak methane flux diel patterns in rice paddy fields

期刊： 

Agricultural and Forest Meteorology

第一作者：

Hong Li

第一发表单位：  

重庆大学

用于田间取样策略和时间尺度外推的甲烷（CH₄）排放昼夜模式通常被认为在白天而非夜间达到峰值。然而，在水分限制和高温等水稻普遍经历的特定条件下，CH₄排放的昼夜模式仍不清楚。本研究在水稻的不同生长阶段下，利用三年连续高频的CH₄通量测量，识别了在不同水分和温度条件下的CH₄通量昼夜模式。结果显示，CH₄通量在早期水稻阶段在白天（13:30–14:30）出现显著的单峰值。然而在生殖阶段，白天的CH₄通量显著减少，导致频繁观察到（80–86%）明显的昼夜反向模式。白天的CH₄排放未出现峰值，且在水分受限和高温条件下的白天CH₄排放仅为夜间水平的41.67%。如果使用营养阶段的夜间/白天比例来计算生殖阶段的夜间CH₄通量，可能低估每日CH₄排放的28.49–32.98%。这一与通常的白天峰值模式相反的夜间峰值昼夜模式的发现表明，仅基于白天数据的测量和外推可能低估稻田中的CH₄排放。

水稻田贡献了全球30%至50%的甲烷（CH₄）排放，这表明对灌溉稻田中CH₄排放及其控制的定量理解是减缓全球变暖的关键关注点。许多研究已经估算了水稻田的CH₄通量的规模，但仍然存在很大的不确定性。CH₄排放估算中的不确定性可能源于水稻田的取样策略和时间尺度外推，这些通常基于固定的CH₄昼夜模式。然而，CH₄通量的昼夜模式在稻田中的生物和环境条件下可能受到影响，在特定条件下仍然不明确。因此，迫切需要在水稻田中识别CH₄通量的昼夜模式及其在特定条件下的控制，以获得准确的估算。

许多研究已经尝试识别水稻田中CH₄排放的昼夜变化及其控制。大多数研究报告表明，CH₄通量在稻米生长阶段通常在下午早些时候出现单一峰值，或者在某些时期不定期变化。仅有一项研究在中国浙江省的水稻田中报告了一个星期的CH₄通量显著的夜间峰值模式。大多数这些研究是在持续淹水的稻田中进行的，而在水分受限和高温条件下，CH₄通量的昼夜模式尚未有多少研究探索。然而，土壤水分含量和土壤温度被广泛认为是控制CH₄通量昼夜变化的最重要的环境因素，因为它们调节了稻田中的厌氧条件和酶介导的过程。

水稻田中，产甲烷微生物（产甲烷菌）在土壤/水中以有机物的厌氧分解为终产物产生CH₄，在释放到大气之前，先在有氧条件下被矿化。除了土壤水分含量和温度外，水稻植物还可以通过提供甲烷生成的碳底物来源以及从根区向大气输送甲烷气体来影响CH₄通量的昼夜变化，绕过了土壤氧化层。一些研究发现，单一的下午早期峰值的昼夜模式可能是由水稻光合作用的日常波动引起的。

在水稻田中，不同生长阶段的光合作用碳分配和气体传输的植物结构有所不同。因此，稻米生长阶段之间，水稻植物控制CH₄通量的方式预计会有所不同。此外，在整个生长季节中，水和温度条件也会影响稻米植物介导的CH₄传输。众所周知，在夏季的白天，水分受限的情况下，气孔会部分或完全关闭以减少水分流失，这可能会在生殖阶段减少白天的CH₄排放。同时，只有少数研究在不同的水稻生长阶段的水分受限田间条件下调查了CH₄通量的昼夜模式。

中国占全球水稻田面积的19%，提供了全球30%的稻米产量。这些水稻田中超过80%采用了节水技术，如季中排水和交替湿润干燥，显著减少了CH₄排放。然而，尽管中国稻田中的一些研究通过微气象通量测量报告了CH₄通量的昼夜变化，但在使用节水技术的水稻田中，不同生长阶段的CH₄通量昼夜变化及其控制仍不明确。本研究中，我们在经历了不同水分和温度条件的稻米生长阶段识别了CH₄通量的昼夜模式。这些CH₄通量通过在稻田中应用节水技术使用涡度相关（EC）技术连续三年高频测量。

（注：以上翻译来着ChatGPT，具体文章内容请以原文内容为准。若解读有误欢迎探讨指正。）

Fig. 1: Location and the satellite image from Google Earth taken on 1st October 2018 of the study area (a). The date of each rice growth stage (b, black vertical lines) and mid-season drainage (b, gold vertical lines) during the rice growing season (horizontal lines) in 2016 (green), 2017 (blue) and 2018 (purple). For example, in 2016 (blue line), rice growing season in the field started on DOY 161 (black vertical lines) and harvested on DOY 310, in which DOY 161–212, DOY 213–270 and 271–310 is the vegetative, reproductive, and ripening stage, respectively. The mid-season drainage started from DOY192 to DOY214 (gold vertical lines).

Fig. 2: The average with standard error (SE) of half-hourly CH4 flux, GEP and 5 cm soil temperature over the entire vegetative stage (DOY 164–203, 159–199, 158–198 in 2016, 2017, and 2018, respectively), the entire mid-season drainage (MSD, DOY 204–222, 200–216, 199–214 in 2016, 2017, and 2018, respectively) period, the entire reproductive stage (DOY 223–271, 217–269, 215–271 in 2016, 2017, and 2018, respectively) and the entire ripening stage (DOY 272–315, 270–296, 272–318 in 2016, 2017, and 2018, respectively) during 0–24 h in 2016 (green), 2017 (blue) and 2018 (purple).

Fig. 3: The cross-correlation of half-hourly GEP & CH4 flux (FCH4) and 5 cm soil temperature (Tsoil) & FCH4 for the vegetative stage, the mid-season drainage (MSD) period, the reproductive stage and the ripening stage in 2016 (green), 2017 (blue) and 2018 (purple). The time lag k value (a, b, c, d) returned by ccf (GEP, CH4 flux) estimates the correlation between GEP [t+k] and CH4 flux [t]. When GEP [t+k] are predictors of CH4 flux [t], with k negative in x-axis, it indicates that GEP leads CH4 flux with time k. For example, the variations in GEP leads CH4 flux 1–2 hours(a) while 5 cm soil temperature lagged CH4 flux several hours during the vegetative stage in each year.

Fig. 4: The quartile diagram of half-hourly CH4 flux (FCH4) over the vegetative stage (a, b), on days with distinct nocturnal peaks (NP, 24 days in 2016 and 30 days in 2017, c, d) and the five days near the NP (Near-NP, e, f) at the reproductive stage during 0–24 h in 2016 (green) and 2017 (blue).

Fig. 5: The average diel variation in CH4 flux (FCH4, a1, a2), volumetric water content (VWC, b1, b2), relative humidity (RH, c1, c2), GEP (d1, d2), air temperature (T, green, blue, e1, e2), 5 cm soil temperature (T, gray lines, e1, e2), vapor pressure deficit (VPD, f1, f2), equilibrium pressure gradient (ΔP, g1, g2) induced by thermal transpiration, stomatal conductance related to water stress (Gsw, h1, h2) on days with distinct nocturnal peaks (NP, solid lines, 24 days in 2016 and 30 days in 2017) and the five days near the NP (Near-NP, dotted lines) during the reproductive stage, and during the vegetative stage (dashed lines) in 2016 (40 days, green) and 2017 (41 days, blue).

Fig. 6:The significant cross-correlation (a-d) are plotted for Tsoil & CH4 flux during 10:00–19:00 (a, r = 0.37 and 0.46), VPD & CH4 flux (b, r = -0.71 and -0.65) and ΔP & CH4 flux (c, r = 0.63 and 0.56) during 20:00:00–9:30, Gsw & CH4 flux (d, r = 0.62 and 0.46) during 0:00–24:00 in 2016 (green) and 2017 (blue). The regression showed that CH4 flux significantly decreased with VPD (e, R2 = 0.62 and 0.50) and significantly increased with ΔP (f, R2 = 0.43 and 0.35) during 20:00–9:30 on the days with distinct nocturnal peak. For the whole day (0:00–24:00), CH4 flux significantly increased with stomatal conductance (g0, h0). In particular, before stomatal conductance increased to a high level, CH4 flux showed a strong dependence on stomatal conductance (g, R2 = 0.62 in 2016; 6 h, R2 = 0.35 in 2017).

Fig. 7:The quartile diagram of half-hourly CH4 flux (FCH4), air temperature (Ta) and volumetric water content (VWC) during 0–24 h at the vegetative and early reproductive stage in 2013. The vegetative before mid-season drainage and reproductive stage after mid-season drainage ranged from DOY174 to DOY 201 and DOY 215 to DOY 270, respectively. However, the nighttime CH4 flux data were almost all missing after DOY 243. Then, to reduce uncertainties, the CH4 flux data during the reproductive stage in this figure were only shown during the period of DOY215–242.