quite different from those application strategies in controlled experiments, as atmospheric N deposition occurs much more frequently than what has been done in many
experiments. Moreover, natural N deposition generally remains high over winter due
to increased coal combustion under cold season. Few studies have examined whether
there would be some differences with respect to the effects of N addition on
ecosystem structure and functioning due to the variation of frequency of N inputs
between natural N deposition and tradition experimental N addition. With the aim to
fill such knowledge gaps, another large N addition experiment was established in the
IMGERS in 2008 (Zhang et al. 2017b). There are 38 experimental treatments in
total, with 10 replicate blocks for each treatment, resulting in a total of 380 plots. The
area of each plot is 8 Â 8 m. This experiment followed a randomized block design
that consisted nine N addition rates (i.e. 0, 10, 20, 30, 50, 100, 150,
200, 500 kg N ha
À1 year
À1 ) crossed with two addition frequencies
(2 times vs. 12 times addition per year) and different utilization strategies
(unmown vs. mown) (Zhang et al. 2017b). They use higher N addition rates as a
proxy for N fertilization activities in cropland and rangeland and long-term extreme
N enrichment in the grassland ecosystem. The higher frequency of N addition
(12 times per year) is used to simulate atmospheric N deposition, which occurs
continuously throughout the year. During the months with more rainfall (May to
October), the fertilizer (NH 4 NO 3 ) is weighed and mixed with purified water (1.5 L
per month) and sprinkled evenly using a sprayer to each plot to simulate wet N
deposition. In winter and early spring (November to April), the fertilizer is mixed
with sand and broadcast uniformly to the plot by hand. An untreated control (with
no N, water, sand addition, unmown) is used to detect the effect of water and sand
addition. Moreover, a mowing only control (without N, water, and sand addition) is
included to detect the effect of water and sand addition with mowing, which is
comparable with the untreated control. Mowing in the mown treatments is carried
out in late August each year, simulating typical hay harvest management in local
area. Mowing is performed with a mower at the height of 10 cm. The harvested
biomass is removed from the plots immediately after mowing.
To understand the impacts of N addition on grassland ecosystem and whether
such effects could be reversible, an experiment was established in a semiarid
temperate steppe in Duolun County, Inner Mongolia, China (Hao et al. 2018).
There are six N addition treatments (0, 30, 60, 120, 240, 480 kg N ha
À1 year
À1 ),
with each being replicated for five times. Fertilizer was applied as urea in 2005, but
as NH 4 NO 3 since 2006. Fertilizer was uniformly applied to all the plots in three
equal amounts at the beginning of June, July, and August in each year. To understand the recovery processes of grassland receiving N inputs, N addition under the
treatment with the highest addition level (480 kg N ha
À1 year
À1
) was stopped
since 2008.
In the alpine grassland in Qinghai Province, northeast part of the Tibetan Plateau
(37
18
0 N, 100
15
0 E), Dr. Yuan-He Yang’s group established a N addition experiment with a gradient of eight N levels (0, 10, 20, 40, 80, 160,
240, 320 kg N ha
À1 year
À1 ) since 2013 (Peng et al. 2017). The long-term mean
annual precipitation is 387 mm, with over 90% falling during the growing season
from May to September. The plant community is dominated by a grass Stipa
10 Impacts of Nitrogen Deposition on China’s Grassland Ecosystems
219
experiments. Moreover, natural N deposition generally remains high over winter due
to increased coal combustion under cold season. Few studies have examined whether
there would be some differences with respect to the effects of N addition on
ecosystem structure and functioning due to the variation of frequency of N inputs
between natural N deposition and tradition experimental N addition. With the aim to
fill such knowledge gaps, another large N addition experiment was established in the
IMGERS in 2008 (Zhang et al. 2017b). There are 38 experimental treatments in
total, with 10 replicate blocks for each treatment, resulting in a total of 380 plots. The
area of each plot is 8 Â 8 m. This experiment followed a randomized block design
that consisted nine N addition rates (i.e. 0, 10, 20, 30, 50, 100, 150,
200, 500 kg N ha
À1 year
À1 ) crossed with two addition frequencies
(2 times vs. 12 times addition per year) and different utilization strategies
(unmown vs. mown) (Zhang et al. 2017b). They use higher N addition rates as a
proxy for N fertilization activities in cropland and rangeland and long-term extreme
N enrichment in the grassland ecosystem. The higher frequency of N addition
(12 times per year) is used to simulate atmospheric N deposition, which occurs
continuously throughout the year. During the months with more rainfall (May to
October), the fertilizer (NH 4 NO 3 ) is weighed and mixed with purified water (1.5 L
per month) and sprinkled evenly using a sprayer to each plot to simulate wet N
deposition. In winter and early spring (November to April), the fertilizer is mixed
with sand and broadcast uniformly to the plot by hand. An untreated control (with
no N, water, sand addition, unmown) is used to detect the effect of water and sand
addition. Moreover, a mowing only control (without N, water, and sand addition) is
included to detect the effect of water and sand addition with mowing, which is
comparable with the untreated control. Mowing in the mown treatments is carried
out in late August each year, simulating typical hay harvest management in local
area. Mowing is performed with a mower at the height of 10 cm. The harvested
biomass is removed from the plots immediately after mowing.
To understand the impacts of N addition on grassland ecosystem and whether
such effects could be reversible, an experiment was established in a semiarid
temperate steppe in Duolun County, Inner Mongolia, China (Hao et al. 2018).
There are six N addition treatments (0, 30, 60, 120, 240, 480 kg N ha
À1 year
À1 ),
with each being replicated for five times. Fertilizer was applied as urea in 2005, but
as NH 4 NO 3 since 2006. Fertilizer was uniformly applied to all the plots in three
equal amounts at the beginning of June, July, and August in each year. To understand the recovery processes of grassland receiving N inputs, N addition under the
treatment with the highest addition level (480 kg N ha
À1 year
À1
) was stopped
since 2008.
In the alpine grassland in Qinghai Province, northeast part of the Tibetan Plateau
(37
18
0 N, 100
15
0 E), Dr. Yuan-He Yang’s group established a N addition experiment with a gradient of eight N levels (0, 10, 20, 40, 80, 160,
240, 320 kg N ha
À1 year
À1 ) since 2013 (Peng et al. 2017). The long-term mean
annual precipitation is 387 mm, with over 90% falling during the growing season
from May to September. The plant community is dominated by a grass Stipa
10 Impacts of Nitrogen Deposition on China’s Grassland Ecosystems
219