Downloadarticlefile.Do?Attachtype=PDF&Id=8194

bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

1 Title page

2 3 Title: Categorized analysis of forest ecological values in the China’s conversion cropland to forest program

4 Wen-Ge Yuan a, b, c Jian-Wei Zheng a, c Jian-Cai Gu a, c* Gui-Qiao Lu a, c

5 a Forestry College, Agriculture University of Hebei, No. 2596, Southern street of Lekai, Baoding 071000, China 6 b Langfang Academy of Agriculture and Forestry Sciences, No. 285, Guangyang Road, Langfang 065000, China 7 c Key Lab. of Genetic Resources of Forest and Forest Protection of Hebei Province, Baoding 071000, China

8 E-mails: 9 1. Wen-Ge Yuan, [email protected] 10 2. Jian-Wei Zheng, [email protected] 11 3. Jian-Cai Gu, [email protected] 12 4. Gui-Qiao Lu, [email protected]

13 *Corresponding author: 14 Jian-Cai Gu 15 Tel(Fax) 0086-312-7528724, Mobile 13930835669 16 [email protected] 17

1 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

29 Categorized analysis of forest ecological values in the China’s conversion cropland to

30 forest program

31 Abstract

32 Background

33 The China’s Conversion Cropland to Forest Program (CCFP) is one of the large state ecological construction

34 programs. Up to now, the program has effectively improved the ecological environment and produced large

35 ecological benefit. However, there were also some problems in its implementation process, the program has been

36 sometimes less effective than the expected.

37 Methods

38 Based on the data and the methods of ‘State report on monitoring ecological effects in CCFP’ and the Chinese

39 Forest Ecosystem Research Network (CFERN) in 2013, we analyzed the categorized ‘forest ecological benefit

40 value’ (B-V) s in the three forest restoration ways in different regions in China to provide references for CCFP

41 construction.

42 Results

43 The unit area B-Vs in CCFP varied between 35 000 RMBs.hm-2.a-1 and 100 000 RMBs.hm-2.a-1. Water

44 conservation B-V and species conservation B-V were the two largest constituents, nutrient accumulation B-V was

45 the least in all the categorized B-Vs on regional and unit area scale. The rank of restoration ways on average unit

46 area total B-Vs was—‘hillside forest conservation’ > ‘returning cropland to forest’ > ‘afforestation on suitable

47 barren hills and wasteland’ in CCFP. Among the categorized B-Vs, some pairs were positively correlated with each

2 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

48 other and some were negatively correlative. The correlation coefficients and some regression equations were given

49 in the text and the attached Fig.s.

50 Conclusions

51 Water conservation B-V was the highest and nutrient accumulation B-V was the lowest whether on regional or unit

52 area scale in CCFP.

53 Forest ecological B-Vs varied in different forest restoration ways and different regions in CCFP. The ‘hillside

54 forest conservation’ restoration way and the water conservation B-V should be paid more attention in China’s

55 future forest restoration. We suggest that suitable forest restoration ways should be selective according to the

56 regional specific and ecological targets.

57 There were correlations among the categorized B-Vs, and the correlations varied with different forest restoration

58 ways in CCFP. Knowing about the correlations could clarify the targeted restoration ways according to the goal of

59 ecological benefit.

60 Keywords Categorized analysis, Forest ecological value, Restoration way, CFERN, CCFP

61 1. Background

62 Since the 20th century, a series of ecological problems have become more serious (Zhou 2008) with the increasing

63 population (Wikipedia 2004), irrational development and utilization of natural resources, such as deforestation,

64 biodiversity loss, soil erosion, desertification and so on (Pandit et al. 2007; Sims et al. 1996). The ecological

65 problems have attracted worldwide attention (Wu et al. 2009). Ecosystem service function becomes the hotspot and

66 frontier of current international research (Bailey 1998, 2014; Beier et al. 2010; Zhang et al. 2010; Higgins et al.

67 2005). As a critical component of ecosystem, forest ecosystem plays important roles in water and soil conservation

68 (So-co), carbon fixation and oxygen release (Cf-Or), nutrient accumulation (Nu-ac), atmosphere purification

3 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

69 (At-pu), biological diversity protection and so on (Wang et al. 2009; Metzger et al. 2005). To discuss the service

70 functions of forest ecosystem, a lot of researches (Jonge et al. 2012; Li et al. 2009; Zhang et al. 2004, 2001; Liu et

71 al. 2003, 1996; Zhang et al. 1988; Zhou et al. 1995) and practices have been carried out and many valuable results

72 have been applied to improve the environmental quality. Some scientists divided the forests into northern and

73 southern types to study the forest service functions (Constanza et al. 1997). Capotorti et al. (2012) discussed the

74 ecological classification of land and conservation of biodiversity in Italy. Niu et al. (2012) studied categorized

75 forest ecological values and concluded that the percentages of water and soil values were 40.51% and 9.91%,

76 respectively, in Chinese forest ecosystem.

77 In order to improve the situation of serious ecological deterioration, the largest ecological program—CCFP has

78 been implemented in China since 1999. It is of great importance in improving the state ecological environment,

79 preventing water and soil erosion, improving water conservation (Wa-co) ability, speeding up the adjustment of

80 rural industrial structure, increasing the overall agricultural capacity, promoting the harmonious development

81 between human beings and nature. Although the program has made great achievement, there were also many

82 problems in the implementation (Yang et al. 2011). For example, the program has been sometimes less effective

83 than the expected. (Cao et al. 2011). Understanding characteristics of categorized B-Vs in different forest

84 restoration ways and different regions could make the afforestation more efficient for the future implementation in

85 CCFP.

86 For the purpose of more systematical observation and study on the functions of the forest ecosystem, CFERN has

87 been established in China since the end of 1950s (Wang et al. 2010). By 2015, the number of ecological

88 observation stations had reached 110 in CFERN. Nearly 100 ecological indicators including atmosphere, soil, forest

89 and creatures were involved in the range of the observation. This would strongly promote the state ecological

90 construction. Whereas, categorized comparative analyses and correlation studies for B-Vs in different restoration

4 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

91 ways on a large scale and in long-term in CCFP were rarely reported.

92 Based on the data of ‘State report on monitoring ecological effects in CCFP’ and CFERN in 2013, we carried on

93 the categorized analysis on B-Vs in CCFP. The objective of the study is to find the differences, features and the

94 relationships of the categorized B-Vs among the three forest restoration ways—‘hillside forest conservation’

95 (H-f-c), ‘afforestation on suitable barren hills and wasteland’ (A-b-w), ‘returning cropland to forest’ (R-c-f). We

96 hope it will be able to provide references for the construction of forest restoration and be an interesting issue for us

97 to communicate with the peers.

98 2. Methods

99 2.1. Classifications of B-Vs and restoration ways

100 We divided the B-Vs into six categories: Water conservation; Soil conservation; Carbon fixation and oxygen

101 release; Nutrient accumulation; Atmosphere purification and species conservation. The forest restoration ways were

102 classified as three kinds—‘hillside forest conservation’, ‘returning cropland to forest’, ‘afforestation on suitable

103 barren hills and wasteland’.

104 2.2. Original data and study regions

105 The data of ‘State report on monitoring ecological effects in CCFP in 2013’ (China’s State Forestry Administration

106 2013) and the Chinese Forest Ecosystem Research Network (CFERN) in 2013 were cited in this study. All the

107 observation stations, which the evaluation data came from, are under the technical standards and management

108 regulations of the observation and evaluation of ecological effects in CCFP (CFERN 2013). Six key ecological

109 monitoring provinces and the relevant zones, which represent the main biotope types in CCFP, were involved in the

110 analysis. They are Hebei province (HE-B), Liaoning province (L-N), Hubei province (HU-B), Hunan province

5 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

111 (HU-N), Yunnan province (Y-N) and Gansu province (G-S). The representative zones are: Shijiazhuang city,

112 Tangshan city and Qinhuangdao city in HE-B; Shenyang city and Panjin city in L-N; Shiyan city, Xiaogan city and

113 Enshi Tujia and Miao autonomous prefecture in HU-B; Huaihua city, Hengyang city and Xiangxi Tujia and Miao

114 autonomous prefecture in HU-N; Kunming city and Dali city in Y-N; Lanzhou city and Gannan Tibetan

115 autonomous prefecture in G-S.

116 2.3. Discounted ecological value

117 Parameters of ecological value were discounted to 2013.They are calculated as follows:

118 t = (Dr + Lr) / 2 (1)

119 t: deposit and loan equilibrium interest rate

120 Dr: average deposit rate

121 Lr: average loan rate

122 d = (1 + tn+1) (1 + tn+2) ··· (1 + tm) (2)

123 d: discount rate

124 t: deposit and loan equilibrium interest rate

125 n: the year of obtained parameters

126 m: the year of evaluation

127 2.4. Data processing

128 Observation Methodology for Long-term Forest Ecosystem Research (LY/T1952–2011) (China's State Forestry

129 Administration 2011), Indicators System for Long-term Observation of Forest Ecosystem (LY/T1606–2003)

130 (China’s State Forestry Administration 2003), Specifications for Assessment of Forest Ecosystem Services in

6 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

131 China (LY/T1721–2008) (China’s State Forestry Administration 2008) and the calculation formulas in ‘State report

132 on monitoring ecological effects in CCFP in 2013’ (China’s State Forestry Administration 2013) were exploited in

133 the B-V calculated process. The gained B-Vs and SPSS 19.0 were used in the analysis.

134 3. Nature conditions of study regions

135 The locations of the six key ecological monitoring provinces are shown in Fig. 1.

136 HE-B is located in 113°27′ E–119°50′ E, 36°05′ N–42°40′ N with total area of 188 500 km2. The terrain is

137 downward from northwest to southeast. The northwest part is mainly mountainous and hilly land, the central and

138 southern parts are plain. The average annual precipitation is 484.5 mm and the average annual temperature is

139 -0.5°C–14.2°C (China Meteorological Administration 1981–2010). Soil types are predominantly red and yellow

140 earths. Forests consist of coniferous, broad-leaved, mixed forest and shrub.

141 L-N is located in 118°53′ E–125°46′ E, 38°43′ N–43°26′ N with total area of 148 000 km2. The terrain is

142 downward from the north to the south. The east and west parts are mainly mountain and hilly lands, the central part

143 is plain. The average annual precipitation is 660.0 mm and the average annual temperature is 8.3°C (China

144 Meteorological Administration 1981–2010). Soil types are mainly dark-brown earths and solonetzs. Forest types

145 are mainly coniferous, broad-leaved, mixed forest and shrub.

146 HU-N is located in 108°48′ E–114°15′ E, 30°08′ N–24°38′ N, mainly consists of low mountain and hills with

147 total area of 211 800 km2. The terrain is high in the south and low, flat in the center and north. The average annual

148 precipitation is 1 200.0 mm–1 700.0 mm and the average annual temperature is 15°C–18°C (China Meteorological

149 Administration 1981–2010). Soil types are mainly red or yellow earths. Forest types are mainly coniferous,

150 broad-leaved, mixed and bamboo forest.

7 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

151 HU-B is located in 108°21′ E–116°07′ E, 29°05′ N–33°20′ N, consists of northern mountain land, hills and plain

152 in the center and south with total area of 185 900 km2. The average annual precipitation is 800.0 mm–1 600.0 mm

153 and the average annual temperature is 15°C–17°C (China Meteorological Administration 1981–2010). Soil types

154 are mainly red or yellow earths. Forest types are mainly coniferous, broad-leaved, mixed and bamboo forest.

155 Y-N is located in 97°52′ E–106°18′ E, 21°13′ N–29°25′ N, consists of mountain, hills , basin and plateau with

156 total area of 390 000 km2. The average annual precipitation is 1 100.0 mm–1 600.0 mm and the average annual

157 temperature is 5°C–24°C (China Meteorological Administration 1981–2010). Soil types are mainly red or yellow

158 earths with part of Grey-cinnamon soils. Forest types are mainly coniferous, broad-leaved, mixed and bamboo

159 forest.

160 G-S is located in 92°13′ E–108°46′ E, 32°31′ N–42°57′ N, consists of staggered mountain, valley and plain with

161 total area of 425 900 km2. The average annual precipitation is 386.0 mm and the average annual temperature is

162 0°C–15°C (China Meteorological Administration 1981–2010). Soil type is mainly Aeolian sandy soil. Forest types

163 are mainly coniferous, broad-leaved, mixed and shrub.

164 4. Results

165 4.1. Area distribution of different forest restoration ways

166 From 1999 to 2013, the total forest restoration area in CCFP reached 29 819 100 hm2. ‘Afforestation on suitable

167 barren hills and wasteland’ accounted for 17 455 000 hm2, ‘returning cropland to forest’ accounted for 9 264 133

168 hm2 and ‘hillside forest conservation’ accounted for 3 100 000 hm2. HE-B and G-S formed larger forest restoration

169 areas of 1 866 700 hm2 and 1 896 900 hm2, respectively (Fig.2).

170 4.2. B-Vs of ‘hillside forest conservation’

8 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

171 4.2.1. B-V features of ‘hillside forest conservation’

172 ‘Hillside forest conservation’ is a forest restoration way to avoid human destruction and facilitate natural

173 reforestation through regular hillside-closing measures in suitable mountain regions. Totally 3 100 000 hm2 of

174 ‘hillside forest conservation’ in CCFP had been formed until 2013.

175 For this way of conservation, it was shown that the water conservation B-V was higher than the other categorized

176 B-Vs whether on regional or on unit area scale in the ecological monitoring provinces. The water conservation B-V

177 accounted for approximately 46.6% (HE-B had the highest percentage of 57.6%, L-N had the lowest percentage of

178 31.7%) of the total B-V on regional scale and approximately 46.5% (annual average, 28 287.42 RMBs·hm-2.a-1 / 60

179 814.01 RMBs·hm-2.a-1) on unit area scale. The water conservation B-V was obviously higher in HE-B with unit

180 area B-V of 55 076.01 RMBs·hm-2.a-1. The species conservation B-V ranked the second except in G-S where

181 produced the highest soil conservation B-V with unit area B-V of 8 205.44 RMBs·hm-2.a-1, and it was higher in the

182 southern provinces (e.g. HU-N, HU-B and Y-N) whether in regions or in unit areas. For example, HU-N produced

183 the highest species conservation B-V with unit area B-V of 25 150.16 RMBs·hm-2.a-1, and the northwest

184 province—G-S produced only 3713.04 RMBs·hm-2.a-1. The nutrient accumulation B-V was the lowest in all the

185 categorized B-Vs with an average of 928.85 RMBs·hm-2.a-1. The lowest ‘carbon fixation and oxygen release’ B-V

186 occurred in HU-N with unit area B-V of 670.94 RMBs·hm-2.a-1.

187 There were also different performances between regional and unit area scale. The rank of regional total B-V was

188 HE-B > Y-N > HU-N > HU-B > L-N > G-S, whereas, the rank of unit area total B-V was HE-B > HU-B > Y-N >

189 HU-N > L-N > G-S. L-N had formed the highest regional annual soil conservation B-V with 1 322.00 million

190 RMBs, but its unit area soil conservation B-V ranked the second with 7 457.01 RMBs·hm-2.a-1. HU-N ranked the

191 second regional water conservation B-V with 4322.00 million RMBs, meanwhile, its unit area water conservation

192 B-V ranked the third with 29216.48 RMBs·hm-2.a-1 (Fig.3).

9 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

193 4.2.2. Correlation analysis for ‘hillside forest conservation’ B-Vs

194 To find the relationships among the ‘hillside forest conservation’ B-Vs, fourteen data lines of categorized unit area

195 B-Vs that come from the six monitoring provinces in 2013 were calculated. The result showed that the unit area

196 water conservation B-V and the unit area atmosphere purification B-V had significantly positive correlations with

197 their total B-Vs (r=0.906, p<0.01; r=0.914, p<0.01), so were the water conservation B-V with atmosphere

198 purification B-V (r=0.722, p<0.01) and water conservation B-V with nutrient accumulation B-V (r=0.633, p<0.01).

199 Meanwhile, the unit area nutrient accumulation B-V had significantly negative correlation with the unit area

200 species conservation B-V (r=-0.532, p<0.05). We adopted 77 data lines of categorized regional B-Vs to analyze

201 their correlations. The result showed that regional total B-V had significantly positive correlations with all the

202 relevant categorized regional B-Vs (Table 1). Several regressions among categorized B-Vs of ‘hillside forest

203 conservation’ were shown in Fig. 4.

204 Table 1 Correlation coefficients among the annual B-Vs of hillside forest conservation

Wa-co So-co Cf-Or Nu-ac At-pu Sp-co Total Wa-co So-co Cf-Or Nu-ac At-pu Sp-co Total

Of unit area B-Vs (p<0.01**; p<0.05*) Of regional B-Vs (for all correlations p<0.01)

Wa-co 1 1

So-co -0.242 1 0.492 1

Cf-Or 0.518* 0.445 1 0.841 0.623 1

Nu-ac 0.633** 0.249 0.822 1 0.819 0.518 0.957 1

At-pu 0.722** -0.022 0.45 0.304 1 0.936 0.539 0.923* 0.900 1

Sp-co 0.185 -0.403 -0.403 -0.532* 0.535* 1 0.641 0.420 0.601 0.505 0.669 1

10 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

Total 0.906** -0.117 0.495* 0.439 0.914** 0.474* 1 0.972 0.607 0.914 0.868 0.965 0.749 1

205 4.3. B-Vs of ‘returning cropland to forest’

206 4.3.1. B-V Features of ‘returning cropland to forest’

207 According to the state regulation (China’s State Forestry Administration 2002), the ‘returning cropland to forest’

208 was performed in the regions with serious soil and water loss, desertification and stony desertification, salinization;

209 The low yield regions with crucial ecological function; The regions at river source or river side and the croplands

210 with ecological importance which were seriously damaged by the wind and sand.

211 By 2013, China had finished 9 264 133 hm2 of ‘returning cropland to forest’. It was shown that water

212 conservation was the main part in the relevant total B-V, which accounted for approximately 49.0% (HE-B had the

213 highest percentage of 58.3%, L-N had the lowest percentage of 30.0%) of the total B-V on regional scale and

214 approximately 47.3% (annual average, 25 238.93 RMBs·hm-2.a-1 / 53 386.17 RMBs·hm-2.a-1) on unit area scale.

215 The nutrient accumulation B-V was still the lowest with the unit area B-V of 781.06 RMBs·hm-2.a-1. Obviously,

216 HU-N had the highest species conservation B-V and the lowest ‘carbon fixation and oxygen release’ B-V whether

217 on regional scale or on unit area scale. G-S produced the highest soil conservation B-V with unit area B-V of 9

218 305.57 RMBs·hm-2.a-1.

219 The rank of regional total B-V was HU-N > HE-B > G-S > Y-N > HU-B > L-N, and the rank of unit area total

220 B-V was HU-N > Y-N > HE-B > HU-B > G-S > L-N. HE-B produced the highest regional annual water

221 conservation B-V with 19 228.00 million RMBs, but its unit area water conservation B-V ranked the second with

222 31 811.40 RMBs·hm-2.a-1. In addition, HE-B produced the highest regional annual B-V of ‘carbon fixation and

223 oxygen release’ with 5 410.00 million RMBs, and its unit area B-V of ‘carbon fixation and oxygen release’ ranked

224 the third with 8 950.47 RMBs·hm-2.a-1 (Fig. 5).

11 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

225 4.3.2. Correlation analysis for B-Vs of ‘returning cropland to forest’

226 Thirty-one data lines of categorized regional B-Vs and fourteen data lines of categorized unit area B-Vs in the

227 ‘returning cropland to forest’ way were analyzed. The result showed that some of the correlations among the

228 categorized B-Vs were positive and others were negative whether on regional scale or on unit area scale. However,

229 the consistent result showed that the B-V pairs of water conservation with its total, ‘carbon fixation and oxygen

230 release’ B-V with its relevant nutrient accumulation B-V, atmosphere purification B-V with its relevant species

231 conservation B-V, atmosphere purification B-V with its total, and species conservation B-V with its total had

232 significantly positive correlations. Conversely, the B-V pair of nutrient accumulation B-V with its relevant species

233 conservation B-V was negatively correlative (Table 2). Several regressions of the B-Vs were shown in Fig. 6.

234 Table 2 Correlation coefficients among the annual B-Vs of returning cropland to forest

Wa-co So-co Cf-Or Nu-ac At-pu Sp-co Total Wa-co So-co Cf-Or Nu-ac At-pu Sp-co Total

Of regional B-Vs Of unit area B-Vs

Wa-co 1 1

So-co 0.592** 1 -0.593* 1

Cf-Or 0.33* 0.645** 1 -0.265 0.232 1

Nu-ac 0.101 0.402* 0.752** 1 -0.088 0.111 0.797** 1

At-pu 0.924** 0.463** 0.139 -0.123 1 0.376 -0.474* -0.851** -0.840** 1

Sp-co 0.821** 0.187 -0.115 -0.235 0.933** 1 0.227 -0.257 -0.842** -0.690** 0.887** 1

Total 0.987** 0.632** 0.351* 0.11 0.952** 0.845** 1 0.840** -0.409 -0.478* -0.317 0.630** 0.637** 1

235 Note: p<0.01**; p<0.05*

12 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

236 4.4. B-Vs of ‘afforestation on suitable barren hills and wasteland’

237 4.4.1. B-V Features of ‘afforestation on suitable barren hills and wasteland’

238 By 2013, the total area of ‘afforestation on suitable barren hills and wasteland’ had reached 17 455 000 hm2. In the

239 way of ‘afforestation on suitable barren hills and wasteland’, the categorized B-Vs had the similar ranking trends

240 between regional scale and unit area scale among the monitoring provinces. Water conservation B-V was still the

241 largest in all the B-Vs, which accounted for 45.7% (HE-B had the highest percentage of 58.5%, L-N had the lowest

242 percentage of 24.7%) of the total B-V on regional scale and approximately 44.9% (annual average, 22 950.56

243 RMBs·hm-2.a-1 / 51 082.92 RMBs·hm-2.a-1) on unit area scale. Nutrient accumulation B-V was the lowest with the

244 unit area B-V of 4 877.82 RMBs·hm-2.a-1. Atmosphere purification B-Vs varied little among the monitoring

245 provinces. G-S produced the highest soil conservation B-V with regional annual 12 414.00 million RMBs, and unit

246 area of 11 599.37 RMBs·hm-2.a-1. HU-N possessed the highest species conservation B-V and the lowest ‘carbon

247 fixation and oxygen release’ B-V (Fig. 7).

248 The rank of regional total B-V was G-S > HE-B > HU-N> Y-N > L-N > HU-B, meanwhile, the rank of unit area

249 total B-V was Y-N > HU-N > HU-B > G-S > HE-B > L-N. HU-N produced the highest unit area water

250 conservation B-V of 31 006.82 RMBs·hm-2.a-1, but its regional annual ranked the second with 23492.00 million

251 RMBs. HU-B formed the highest unit area of ‘carbon fixation and oxygen release’ B-V with 10 826.46

252 RMBs·hm-2.a-1, and its regional annual ranked the second with 6 771.00 million RMBs.

253 4.4.2. Correlation analysis for B-Vs of ‘afforestation on suitable barren hills and wasteland’

254 We analyzed 14 data lines of unit area B-Vs and 31 data lines of regional B-Vs of ‘afforestation on suitable barren

255 hills and wasteland’. For regional B-Vs, The results suggested that every categorized regional B-V had significantly

13 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

256 positive correlations with other relevant B-Vs except the four pairs: species conservation with soil conservation,

257 species conservation with ‘carbon fixation and oxygen release’, species conservation with nutrient accumulation

258 and atmosphere purification with nutrient accumulation. For unit area B-Vs, the following pairs had significantly

259 positive correlations at p<0.01 levels: water conservation with its total, ‘carbon fixation and oxygen release’ with

260 its relevant nutrient accumulation, atmosphere purification with its relevant species conservation, and species

261 conservation with its total. On the contrary, some pairs had significantly negative correlations at p<0.01 levels,

262 including ‘carbon fixation and oxygen release’ with its relevant species conservation, ‘carbon fixation and oxygen

263 release’ with its relevant atmosphere purification, nutrient accumulation with its relevant species conservation, and

264 nutrient accumulation with its relevant atmosphere purification (Table 3). Several regressions were given in Fig. 8.

265 Table 3 Correlation coefficients among the annual B-Vs of afforestation on suitable barren hills and wasteland

Wa-co So-co Cf-Or Nu-ac At-pu Sp-co Total Wa-co So-co Cf-Or Nu-ac At-pu Sp-co Total

Of regional B-Vs Of unit area B-Vs

Wa-co 1 1

So-co 0.548** 1 -0.615* 1

Cf-Or 0.455** 0.652** 1 -0.493* 0.26 1

Nu-ac 0.307* 0.364* 0.787** 1 -0.31 0.106 0.783** 1

At-pu 0.808** 0.752** 0.493** 0.182 1 0.335 -0.384 -0.745** -0.898** 1

Sp-co 0.737** 0.225 0.073 0.007 0.679** 1 0.337 -0.163 -0.788** -0.718** 0.703** 1

Total 0.941** 0.751** 0.619** 0.403* 0.91** 0.72** 1 0.835** -0.300 -0.540* -0.460 0.411 0.671** 1

266 Note: p<0.01**; p<0.05*

267 4.5. Comparisons of annual categorized unit area B-Vs in different regions and forest restoration ways

14 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

268 Categorized unit area B-Vs were calculated in different forest restoration ways (Fig. 9). For the water conservation

269 B-Vs, there were no obviously differences among the three restoration ways in the monitoring provinces expect

270 HE-B, in which the water conservation B-V of ‘hillside forest conservation’ was obviously higher than the B-Vs of

271 other two restoration ways. The soil conservation B-Vs of G-S, L-N and Y-N were higher than other provinces in

272 all the three restoration ways. For the ‘carbon fixation and oxygen release’ B-V, the ‘hillside forest conservation’

273 was higher than the other two ways in HE-B, and there were no obviously differences among the three ways in

274 other provinces. The ‘carbon fixation and oxygen release’ B-Vs of the three ways in HU-N and G-S were lower

275 than other provinces especially in HU-N. HE-B produced higher nutrient accumulation B-V in the ‘hillside forest

276 conservation’ way. L-N and G-S had higher nutrient accumulation B-Vs in the way of ‘afforestation on suitable

277 barren hills and wasteland’. HU-N had the least nutrient accumulation B-V, but its species conservation and the

278 atmosphere purification B-Vs were obviously higher in all the three restoration ways. For species conservation B-V,

279 HU-B, HE-B and Y-N produced higher B-Vs in ‘hillside forest conservation’ way. HE-B produced higher ‘hillside

280 forest conservation’ B-Vs. HU-N produced higher atmosphere purification B-V in ‘returning cropland to forest’

281 way. Except soil conservation B-V, ‘hillside forest conservation’ had higher categorized B-Vs especially in water

282 conservation B-V, atmosphere purification B-V and species conservation B-V. ‘Afforestation on suitable barren

283 hills and wasteland’ produced more soil conservation B-V (Table 4).

284 Table 4 Summary of annual B-Vs in different forest restoration ways

Wa-co So-co Cf-Or Nu-ac At-pu Sp-co Total Wa-co So-co Cf-Or Nu-ac At-pu Sp-co Total

For regional area of the six monitoring provinces/Million RMBs For unit area of the six monitoring provinces/RMBs.hm-2

H-f-c 25761 4890 6694 903 4323 12720 55291 28287.42 5210.17 7200.61 928.85 4761.38 14425.58 60814.01

R-c-f 71625 14586 18180 2109 11442 28382 146324 25238.93 5384.72 7117.79 781.06 4127.75 10735.91 53386.17

15 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

A-b-w 113562 29133 34491 4168 18703 48567 248624 22950.56 5737.75 7169.74 812.97 3888.72 10523.19 51082.92

285 5. Discussions

286 In this study, we took the CCFP as the object to analyze the features and the correlations of categorized forest

287 ecological benefit values. The results are also of common interest to general forest management. Although there

288 were some studies on the similar topic (Niu et al. 2012; Wang et al. 2011), this study will help to understand more

289 about the details of categorized forest ecological benefit values in the aspects of forest restoration ways, regional

290 specifics. In addition, the results could be of relevance to the environment protection and be used for reference for

291 the future construction of CCFP.

292 5.1. Regional differences of B-Vs

293 Regional B-Vs are related to the local natural conditions including landform, soil, climate and so on (China’s State

294 Forestry Administration 2014; Sun et al. 2007, 2006; Jackson et al. 2005). We have the consistent results in this

295 study.

296 The unit area B-Vs of species conservation and water conservation in southern regions (e.g. HU-N, Y-N) were

297 higher than the northern regions (Fig. 3; Fig. 5; Fig. 7). The abundant water and heat resources in southern regions

298 make the forest and other spices easy to make use of the resources for growth and reproduction, and hence the

299 diversity and water-holding ability of these species increased. Only one exception is in HE-B where produced the

300 highest unit area water conservation B-V especially in the ‘hillside forest conservation’ way. The more plateau

301 sandy land and more shrubs in the northern parts in HE-B, and the more raining in its hot season, make the shrubs

302 easy to develop their stronger function on water conservation in the sandy land that suffers from serious water loss.

303 The larger area of fast-growing forest seems to be the main reason for the highest unit area B-V of nutrient

16 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

304 accumulation in HE-B. More hillsides with serious soil erosion make the soil loss more obvious in G-S and Y-N,

305 which might be the reason for having high soil conservation B-Vs, as shown in our study (Fig. 3; Fig. 5; Fig. 7).

306 According to Yin (2010), the ‘carbon fixation and oxygen release’ B-V of mixed forest is significantly higher

307 than that of coniferous forest (e.g. Pinus massoniana Lamb.), and we have the consistent result that HU-N had the

308 lowest ‘carbon fixation and oxygen release’ B-V but produced the highest atmosphere purification B-V with its

309 more coniferous forest. Probably, the low growth rate of coniferous forests resulted in the lowest ‘carbon fixation

310 and oxygen release’ B-V and nutrient accumulation B-V. In addition, the result also indirectly reflected that the

311 coniferous forests might have stronger ability of atmosphere purification. This speculation is contrary to the

312 previous study of Nie et al. (2015), in which they concluded that the purification capacity of different types of

313 urban forest on atmosphere could be ranked as broad-leaved mixed forest > planted bush > conifer forest. Whereas,

314 Shi et al. (2016) presented the same result as we do. They concluded that the ability of atmosphere purification of

315 coniferous forests is higher than broad-leaved forests.

316 5.2. B-V ranks of forest restoration ways

317 The rank of classified regional B-Vs in the six monitoring provinces was—‘afforestation on suitable barren hills

318 and wasteland’ > ‘returning cropland to forest’ > ‘hillside forest conservation’. Whereas, the rank of average unit

319 area total B-Vs was opposite—‘hillside forest conservation’ > ‘returning cropland to forest’ > ‘afforestation on

320 suitable barren hills and wasteland’ in CCFP (Table 4). It appeared that the regional B-V in the way of

321 ‘afforestation on suitable barren hills and wasteland’ was larger than the other two ways. In fact, this was partly

322 because of its larger area (Fig. 2), and the rank of unit area total B-Vs reflected the reality of the ecosystem service.

323 Shi et al. (2016) concluded that the ecosystem services of ‘returning farmland to forests’ and ‘closing hillsides to

324 facilitate afforestation’ were better than those of ‘afforestation on barren hills and wasteland’. This result gave a

17 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

325 further example to illustrate our conclusion. The features of unit area total B-V can be reasonably considered in

326 choosing forest restoration way. Whereas, B-V is not the only goal of CCFP, regional specifics and actual needs

327 should be taken into account to achieve joint-win of ecological, social and economic benefits in the process of

328 forest restoration.

329 5.3. Relationships among the B-Vs

330 Forests usually cannot simultaneously produce multiple, positive ecosystem services because of the trade-offs

331 among different or competing functions. Maximizing one service may cause substantial declines of other services

332 (Bennett et al. 2009). However, our result showed that forests could produce multiple, consistent positive

333 ecosystem services for the China’s ‘hillside forest conservation’ way on regional scale (Table 1). It is essential to

334 understand the dynamic relationships among all forest ecosystem services (Wang et al. 2011). In addition, Wang et

335 al. (2011) analyzed the relationships between forest cover and runoff on different area scales and concluded that the

336 correlations varied between large scale and meso scale. However, few studies have reported on the correlations

337 among the categorized B-Vs, which were calculated in this study. Our results showed that some of the correlations

338 were consistent and others were not between regional scale and unit area scale. This result could be helpful for

339 considering pros and cons between forest single managing goal and total ecological benefits.

340 5.4. Unit area total B-Vs in CCFP

341 Wang et al. (2011) released the calculated result of B-V in China general forest ecosystem (40 000

342 RMBs.hm-2.a-1–50 000 RMBs.hm-2.a-1). According to the result of this study, the range of forest annual unit area

343 total B-Vs was 35 000 RMBs.hm-2.a-1–100 000 RMBs.hm-2.a-1 in CCFP. The average annual unit area total B-Vs of

344 different forest restoration ways was 51 082 RMBs.hm-2.a-1–60 814 RMBs.hm-2.a-1 (Table 4). This result indicated

18 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

345 that the unit area total B-Vs in CCFP was larger than that of China general forest ecosystem.

346 Furthermore, Niu et al. (2012) concluded that the percentages of water values were 40.51% of the total value in

347 Chinese forest ecosystem. Shi et al. (2016) released the percentages of water values were 28.90%, and Qin (2009)

348 released the percentages of water values was 54.09% in CCFP. Our study presented more detailed percentages of

349 water values than the previous in deferent restoration ways. The results were 46.6% on regional scale and 46.5% on

350 unit area scale in ‘hillside forest conservation’ way, 49.0% on regional scale and 47.3% on unit area scale in

351 ‘returning cropland to forest’ way, 45.7% on regional scale and 44.9% on unit area scale in ‘afforestation on barren

352 hills and wasteland’ way in CCFP.

353 We deduced that CCFP was carried out in the regions with serious ecological degradation, in which their B-Vs

354 were more obvious than the general regions. Up to present, the unified calculation methods of B-V were not formed

355 yet. Different calculation methods might also result in the inconsistencies.

356 5.5. Performances of categorized B-Vs

357 Water is the most sensitive and limiting ecological factor in forest eco-system (Wang et al. 2011). Our result

358 showed that water conservation B-V was the main part in total B-V in CCFP, and hence, it makes the water more

359 sensitive to the total B-V.

360 Nutrient accumulation B-V was the least in total B-V and varied in different regions in CCFP. The higher

361 nutrient accumulation B-Vs occurred in the North and Northeast China (Fig. 3, Fig. 5 and Fig. 7). This result might

362 be related to the differences of climate, soil and afforestation species.

363 Wang et al. (2011) concluded the performance of the overall B-Vs in China’s general forest by the following

364 sequence: water conservation B-V > species conservation B-V > carbon fixation and oxygen release B-V > soil

365 conservation B-V > atmosphere purification B-V > nutrient accumulation B-V. We had the same result in CCFP

19 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

366 not only in regional but also in unit area (Table 4).

367 5.6. Limitations

368 We did not present all the regressions of the B-Vs due to space limitations in the article. There are also many

369 ecological service values for the forest should be taken into the analysis. Such as noise reduction, landscape value

370 and so on. We also did not involve them for lack of the detailed materials of these aspects. Duo to the limitations of

371 the level of cognition and study, some points of the discussion are speculative. We are willing to communicate with

372 our peers to improve our research.

373 6. Conclusions

374 In the six ecological monitoring provinces in 2013, Water conservation B-V was the highest and nutrient

375 accumulation B-V was the lowest whether on regional or unit area scale in CCFP. The rank of categorized B-Vs

376 was—(water conservation B-V) > (species conservation B-V) > (carbon fixation and oxygen release B-V) > (soil

377 conservation B-V) > (atmosphere purification B-V) > (nutrient accumulation B-V) in CCFP.

378 In CCFP, forest ecological B-Vs varied in different forest restoration ways and different regions. The rank of

379 average unit area total B-Vs was—‘hillside forest conservation’ > ‘returning cropland to forest’ > ‘afforestation on

380 suitable barren hills and wasteland’. Unit area B-Vs of species conservation and water conservation in southern

381 regions were higher than that of northern and northwestern regions in CCFP. The hot and rainy regions produced

382 higher species conservation B-Vs, and the regions with more coniferous forest had higher atmosphere purification

383 B-Vs and lower ‘carbon fixation and oxygen release’ B-Vs. The regions with more hilly area or more sandy land

384 had higher soil conservation B-Vs. ‘Hillside forest conservation’ was a better way for the regions aiming at water

385 conservation, atmosphere purification and species conservation. For the regions aiming at soil conservation,

20 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

386 ‘afforestation on suitable barren hills and wasteland’ was more suitable. The ‘hillside forest conservation’

387 restoration way and the water conservation B-V should be paid more attention in China’s future forest restoration.

388 We suggest that suitable forest restoration ways should be selective according to the regional specific.

389 There were correlations among the categorized B-Vs, and the correlations varied with different forest restoration

390 ways in CCFP. Water conservation B-V had significantly positive correlation with the relevant total B-V and

391 positive correlation with the relevant atmosphere purification B-V on both regional scale and unit area scale.

392 Species conservation B-V of unit area was negatively correlated with the relevant nutrient accumulation B-V

393 except in the way of ‘afforestation on suitable barren hills and wasteland’. Regional species conservation B-V had

394 significantly negative correlation with the relevant nutrient accumulation B-V except the ‘hillside forest

395 conservation’ way. Knowing about the correlations among the categorized B-Vs could clarify the targeted

396 restoration ways according to the goal of ecological benefit.

397 Abbreviations

398 CCFP: The China’s Conversion Cropland to Forest Program

399 CFERN: The Chinese Forest Ecosystem Research Network

400 B-V: Forest ecological benefit value

401 Ecological monitoring provinces

402 HE-B: Hebei province

403 L-N: Liaoning province

404 HU-B: Hubei province

405 HU-N: Hunan province

406 Y-N: Yunnan province

21 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

407 G-S: Gansu province

408 In Figures and Tables

409 H-f-c: Hillside forest conservation

410 R-c-f: Returning cropland to forest

411 A-b-w: Afforestation on suitable barren hills and wasteland

412 Wa-co: water conservation

413 Sp-co: Species conservation

414 At-pu: Atmosphere purification

415 So-co: Soil conservation

416 Cf-Or: Carbon fixation and oxygen release

417 Nu-ac: Nutrient accumulation

418 Declarations

419 Acknowledgements

420 The Hebei Provincial Science & Technology Supporting Program (No.15227652D) and CFERN & BEJING TECHNO SOLUSIONS

421 Award Funds on excellent academic achievements provided the Project support. The work was also guided by ‘Observation

422 Methodology for Long-term Forest Ecosystem Research’ of National Standards of the People’s Republic of China (GB/T 33027–2016).

423 We appreciate that Dr. Bing Wang, a researcher from Chinese Academy of Forestry provided some of the technical data. The author is

424 also indebted to Dr. Yuwu Li from Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences for his advice on the

425 article. In addition, Key Lab. of Genetic Resources of Forest and Forest Protection of Hebei Province should be credited for part of its

426 Supporting. We thank the journal reviewers for their detailed and the constructive comments on the manuscript.

427 Funding

22 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

428 — Hebei Provincial Science & Technology Supporting Program (No.15227652D).

429 — CFERN & BEJING TECHNO SOLUSIONS Award Funds on excellent academic achievements.

430 Availability of data and materials

431 We declare that the materials described in the manuscript, including all relevant raw data, are freely available, without breaching

432 participant confidentiality.

433 Legal statement

434 All research work reported in this study was performed in accordance with all relevant legislation and guidelines.

435 Authors' contributions

436 All authors conceived the study performed research and analyzed data. Wen-Ge Yuan wrote the paper.

437 Ethics approval and consent to participate

438 This manuscript does not report on or involve the use of any animal or human data or tissue.

439 Consent for publication

440 This manuscript does not involve animal or human study. We understand that the text and any pictures published in the article will be

441 freely available on the internet and may be seen by the public.

442 Competing interests

443 The authors declare that they have no competing interests. Both at the time of conducting this research as well as at present, none

444 declared by others.

445 Authors' information

446 Wen-Ge Yuan a, b, c Jian-Wei Zheng a, c Jian-Cai Gu a, c* Gui-Qiao Lu a, c

23 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

447 a Forestry College, Agriculture University of Hebei, No. 2596, Southern street of Lekai, Baoding 071000, China

448 b Langfang Academy of Agriculture and Forestry Sciences, No. 285, Guangyang Road, Langfang 065000, China

449 c Key Lab. of Genetic Resources of Forest and Forest Protection of Hebei Province, Baoding 071000, China

450 References

451 Bailey RG (2014) Ecoregions of the Continents: The Polar Ecoregions. In: Bailey RG (ed) Ecoregions. Springer, New York, pp 47–54.

452 Available via DIALOG. https://link.springer.com/chapter/10.1007%2F978-1-4939-0524-9_5. Accessed 19 Nov 2017.

453 Bailey RG (1998) Ecoregions: Ecosystem geography of the oceans and continents. In: Bailey RG (ed) Ecoregions. Springer, New

454 York. Available via DIALOG. https://link.springer.com/book/10.1007%2F978-1-4612-2200-2#toc. Accessed 3 Sept 2017.

455 Beier P, Brost B (2010) Use of land facets to plan for climate change: conserving the arenas, not the actors. Conservation Biology

456 24(3):701–710.

457 Bennett EM, Peterson GD, Gordon LJ (2009) Understanding relationships among multiple ecosystem services. Ecology Letters

458 12(12):1394–1404.

459 Cao SX, Sun G, Zhang, ZQ, et al. (2011) Greening China Naturally. AMBIO 40 (7):828–831.

460 CFERN (2013) The technical standards and management regulations for the observation and evaluation of ecological effects in CCFP

461 (in Chinese with English subtitles). In: standards and regulations. CFERN. Available via DIALOG.

462 https://www.cfern.org/portal/article/index/id/10705/page/1.html. Accessed 11 Jun 2017.

463 China’s State Forestry Administration (2002) Regulations of returning cropland to forest (in Chinese). In: Afforestation. Information

464 release. Available via DIALOG. http://www.forestry.gov.cn/main/3960/content-621602.html. Accessed 19 Aug 2017.

465 China’s State Forestry Administration (2003) Indicators System for Long-term Observation of forest Ecosystem (LY/T1606–2003). In:

466 Forestry. Agriculture and forestry. Available via DIALOG. http://sitewww.51zbz.net/biaozhun/31665.html. Accessed 20 Oct

467 2017.

468 China’s State Forestry Administration (2008) Specifications for Assessment of Forest Ecosystem Services in China (LY/T1721–2008).

24 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

469 In: Forestry. Agriculture and forestry. Available via DIALOG. http://aww.51zbz.net/biaozhun/31900.html. Accessed 20 Oct

470 2017.

471 China’s State Forestry Administration (2011) Observation Methodology for Long-term Forest Ecosystem Research (LY/T1952–2011).

472 http://www.doc88.com/p-3505558517824.html. Accessed 3 Nov 2017.

473 China’s State Forestry Administration (2013) State report on monitoring ecological effects in conversion of cropland to forest program.

474 Chinese Forestry Publishing House, Beijing, China (in Chinese).

475 China’s State Forestry Administration (2014) State report on monitoring ecological effects in conversion of cropland to forest program.

476 Chinese Forestry Publishing House, Beijing, China (in Chinese).

477 China Meteorological Administration (1981–2010) Data set of ground normals in China. In: Ground materials. China Meteorological

478 Data Service Center. Available via DIALOG. http://data.cma.cn/data/detail/dataCode/A.0029.0003.html. Accessed 16 Nov 2017.

479 Constanza R, D’Arge R, Groot RD, et al. (1997) The value of the world’s ecosystem services and natural capital. Nature 387:253–260.

480 Capotorti G, Giuda D, Siervo V, et al. (2012) Ecological classification of land and conservation of biodiversity at the national level:

481 The case of Italy. Biological Conservation 147(1): 174–183.

482 Higgins JV, Bryer MT, Khoury ML, et al. (2005) A freshwater classification approach for biodiversity conservation planning.

483 Conservation Biology 19(2):432–445.

484 Jackson RB, Jobbágy EG, Avissar R, et al. (2005) Trading water for carbon with biological sequestration. Science

485 310(5756):1944–1947.

486 Jonge VND, Pinto R, Turner RK (2012) Integrating ecological, economic and social aspects to generate useful management

487 information under the EU Directives’ ‘ecosystem approach’. Ocean & Coastal Management 68(68):169–188.

488 Pandit MK, Sodhi NS, Lian PK, et al. (2007) Unreported yet massive deforestation driving loss of endemic biodiversity in Indian

489 Himalaya. Biodiversity & Conservation 16(1):153–163.

490 Liu SR, Wen YG, Wang B, et al. (1996) The Low of Eco-hydrological functions of forest ecosystems in China. Chinese Forestry

25 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

491 Publishing House, Beijing, China (in Chinese).

492 Liu SR, Sun PS, Wen YG (2003) Comparative analysis of hydrological functions of major forest ecosystems in China. Acta

493 Phytoecologica Sinica 27(1):16–22 (in Chinese with English abstract).

494 Metzger MJ, Bunce RGH, Jongman RHG, et al. (2005) A climatic stratification of the environment of Europe. Global Ecology &

495 Biogeography 14(6):549–563.

496 Nie L, Deng ZH, Chen QB, et al. (2015) SO2 and NOx purification ability of forest in Kunming city. Journal of West China Forestry

497 Science 44(4): 116–120. http://www.cnki.com.cn/Article/CJFDTotal-YNLK201504022.htm. Accessed 23 Nov 2017.

498 Niu X, Wang B, Liu S, et al. (2012) Economical assessment of forest ecosystem services in China. Ecological Complexity

499 11(11):1–11.

500 Shi HX, Wang B, Niu X (2016) Ecosystem services of Grain for Green Project in the provinces of the upper and middle reaches of

501 Yangtze and Yellow River. Chinese Journal of Ecology 35(11):2903–2911.

502 http://www.cje.net.cn/CN/article/downloadArticleFile.do?attachType=PDF&id=8194. Accessed 15 Sep 2017.

503 Qin CX (2009) Study on comprehensive benefit evaluation of the cropland conversion to forest in Chengde county. Master thesis.

504 Agricultural University of Hebei (in Chinese with English abstract).

505 http://cdmd.cnki.com.cn/Article/CDMD-11920-2009131688.htm. Accessed 10 Aug 2017.

506 Sims RA, Corns IGW, Klinka K (1996) Introduction-global to local: Ecological land Classification. Environmental Monitoring and

507 Assessment 39(1–3):1–10.

508 Sun G, Zhou G, Zhang Z, et al. (2006) Potential water yield reduction due to reforestation across China. Journal of Hydrology

509 328(3–4):548–558.

510 Sun G, Zhang ZQ, Zhou GY, et al. (2007) Forest watershed hydrological simulation models: concepts, use and application problems in

511 China. Journal of Beijing Forestry University 29(3):178–184(in Chinese with English abstract).

512 The item group of ‘Assessment of service function of forest ecosystem in China’ (2010) Assessment of service function of forest

26 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

513 ecosystem in China. Chinese Forestry Publishing House, Beijing, China (in Chinese).

514 Wang B, Ding FJ (2010) Construction of standard system for long-term located observation of forest ecosystems. Journal of Beijing

515 Forestry University 32(6):141–145(in Chinese with English abstract).

516 Wang B, Ma XQ, Guo H, et al. (2009) Evaluation of the Chinese fir forest ecosystem services value. Scientia Silvae Sinicae

517 45(4):124–130 (in Chinese with English abstract).

518 Wang B, Ren XX, Hu W (2011) Assessment of Forest Ecosystem Services Value in China. Scientia Silvae Sinicae 47(2): 145–153 (in

519 Chinese with English abstract).

520 Wang S, Fu BJ, He CS, et al. (2011) A comparative analysis of forest cover and catchment water yield relationships in northern China.

521 Forest Ecology & Management 262(7):1189–1198.

522 Wikipedia (2004) Convention on biological diversity. Paper presented at the Seventh Ordinary Meeting of the Conference of the

523 Parties to the Convention on Biological Diversity (COP–7). Kuala Lumpur, Malaysia, 9–20 February 2004.

524 https://en.wikipedia.org/wiki/Convention on Biological Diversity. Accessed 13 Sep 2017.

525 Wu YX, Kang WX, Guo QH, et al. (2009) Functional value of absorption and purgation to atmospheric pollutants of urban forest

526 in Guangzhou. Scientia Silvae Sinicae 45(5):42–48 (in Chinese with English abstract).

527 Yang CJ, Dai QS (2011) Problems and countermeasures in consolidating the results of returning farmland to forest. Central South

528 Forest Inventory and Planning 30(4):11–14 (in Chinese with English abstract).

529 Yin GQ (2010) Study on Eco-environmental Effects under Different ‘Grain for Green’ Patterns in Huitong, Hunan Province. PhD

530 thesis. Central South University of Forestry and Technology (in Chinese with English abstract).

531 http://xueshu.baidu.com/s?wd=paperuri:(b77b0b0f48e993fb462b90c232c4ab0a)&filter. Accessed 15 Nov 2017.

532 Zhang B, Li WH, Xie GD (2010) Research on ecosystem services in China: progress and perspectives. Ecological economics

533 69(7):1389–1395.

534 Zhang ZZ, Yu XX (1988) The status and major progress on Chinese forest hydrological research. Journal of Beijing Forestry

27 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

535 University 10(2):79–87 (in Chinese).

536 Zhang ZQ, Wang LX, Yu XX, et al. (2001) Forest vegetation impacts on the runoff generation: a review. Journal Natural Resources

537 16(1):79–84 (in Chinese). http://www.cqvip.com/read/read.aspx?id=4942955 Accessed 23 Nov 2017.

538 Zhang ZQ, Wang LX, Wang SP (2004) Forest hydrology research in China. Science of Soil and Water Conservation 2(2):68–73 (in

539 English with Chinese abstract).

540 Zhou X, Wang C, Li Q, et al. (1995) Studies on forest water cycling. Journal of Northeast Forestry University 6(3):89–95(in Chinese

541 with English abstract).

542 Zhou KF (2008) Research on remote sensing model for ecological water consumption and its application in Arid Areas. PhD thesis.

543 Xinjiang Institute of Ecology and Geography (in Chinese with English abstract). http://www.doc88.com/p-1962207728451.html.

544 Accessed 24 Jan 2018.

545 Figure legends

546 Fig. 1 The six key monitoring provinces, China

547 Fig. 2 Forest areas of different vegetation restoration ways in key ecological monitoring provinces in CCFP

548 Fig. 3 Annual B-Vs of hillside forest conservation in different provinces

549 a: For unit area; b: For regional area.

550 Fig. 4 The relationships among annual B-Vs of hillside forest conservation

551 a, b: For unit areas; c, d: For regional areas

552 Fig. 5 Annual B-VS in the returning cropland to forest way in different provinces

553 a: For unit areas; b: For regional areas.

554 Fig. 6 The relationships among annual B-Vs of returning cropland to forest

555 a: For regional areas; b: For unit areas

556 Fig. 7 Annual B-Vs of afforestation on suitable barren hills and wasteland in different provinces

28 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

557 a: For regional area; b: For unit area.

558 Fig. 8 The relationships among B-Vs of afforestation on suitable barren hills and wasteland

559 a: For regional area; b: For unit area.

560 Fig. 9 Annual categorized unit area B-Vs of different forest restoration ways

561

562

563

564

565

566

567

29 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

20 2 m h

0 18 0 0 0 0

1 16 / a e r

A 14

12 R-c-f

10 A-b-w H-f-c 8 Total 6

0 Provinces HE-B L-N HU-B HU-N Y-N G-S bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

6 2 - m

H a . s B 5 M R 0 0 0

0 Wa-co 1

4 /

V So-co B Cf-Or 3 Nu-ac At-pu 2 Sp-co

0 HE-B L-N HU-B HU-N Y-N G-S Provinces

l 10000 a t o T

- 9000

s b B

M 8000 R 0 0 0

0 7000 Wa-co 0 0 1 / 6000 So-co V

B Cf-Or 5000 Nu-ac 4000 At-pu 3000 Sp-co

2000

1000

120000

y = 1.4225x + 19264 100000 a R² = 0.8205 s B

80000 M

R Nu-ac At-pu 60000 Total B-V

40000

20000 y = 0.1073x + 1372.7 y = 0.0431x - 619.85 R² = 0.5207 R² = 0.4005

0 Wa-co/RMBs 0 10000 20000 30000 40000 50000 60000 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

120000

100000 s B b M y = 10.135x + 54806 R R² = 0.1929 80000

60000 Sp-co Total B-V

40000

20000 y = -5.7112x + 21023 R² = 0.2835

0 Nu-ac/RMBs 0 500 1000 1500 2000 2500 3000 3500

2500

C(million) 2000 s

B y = 1.9733x + 75.406

M R² = 0.8831 1500 R At-pu Total B-V 1000

500 y = 0.1317x + 9.4018 R² = 0.7284

0 Wa-co/RMBs 0 200 400 600 800 1000 1200 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

2500

2000 d(million)

s y = 5.0163x + 254.01 B R² = 0.6352 M R 1500 At-pu Total B-V 1000

500 y = 0.33x + 21.635 R² = 0.5093 Cf-Or/RMBs 0 0 50 100 150 200 250 300 350 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

3.5 2 - m H . s

B 3 a M R 0 0

0 2.5 0 1

Wa-co / V

B So-co 2 Cf-Or

1.5 Nu-ac At-pu 1 Sp-co

0.5

l 25000 a t o T - s B

M 20000 b R 0 0 0

0 Wa-co 0 0 1 / 15000 So-co V B Cf-Or Nu-ac 10000 At-pu Sp-co 5000

10000

y = 2.0342x + 122.99 a 8000 R² = 0.9736 s B 6000 M At-pu R 0

0 Sp-co 0 0

0 4000 Total B-V 0 1 y = 0.5626x - 112.00 2000 R² = 0.6741 y = 0.1990x - 29.714 R² = 0.8542 0 0 1000 2000 3000 4000 5000

Wa-co/1000000RMBs bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

80000

70000 y = 9073.1x0.1916 R² = 0.3587 b 60000

50000 Cf-Or s

B 40000 At-pu M R Total B-V 30000 y = 1.4E+10x-1.6024 20000 R² = 0.3842 y = 14.722x0.6038 10000 R² = 0.7547

0 0 5000 10000 15000 20000 25000 30000

Sp-co/RMBs bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

l 30000 a t o T - s B 25000 a M R 0 0 0 0

0 20000 Wa-co 0 1 / So-co V B 15000 Cf-Or Nu-ac

10000 At-pu Sp-co

5000

Provinces 0 HE-B L-N HU-B HU-N Y-N G-S

3.5 2 - m H . s

B 3 M

R b 0 0

0 2.5 0

1 Wa-co / V B 2 So-co Cf-Or 1.5 Nu-ac At-pu 1 Sp-co

0.5

10000

s y = 2.2023x + 388.73 B

M R² = 0.8854

8000 R 0 0

0 a 0 0 0 1 6000 So-co Total B-V 4000

2000 y = 0.3140x -33.199 R² = 0.3006

0 Wa-co/1000000RMBs 0 1000 2000 3000 4000 5000

12000 b

10000 Cf-Or At-pu 8000 Nu-ac s

B y = -0.514x + 12902

M R² = 0.6212 R 6000 y = 0.115x + 2665.1 R² = 0.4943

4000

2000 y = -0.0801x + 1828.9 R² = 0.5151 0 Sp-co/RMBs 0 5000 10000 15000 20000 25000 bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

6 2 - m H . s B 5 M R

0 a 0 0 0 1 4 / o c - a H-f-c W 3 R-c-f A-b-w 2

Provinces 0 HE-B L-N HU-B HU-N Y-N G-S bioRxiv preprint doi: https://doi.org/10.1101/358960; this version posted June 29, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.

3.5 2 - m H . s

B 3 M R 0 0 0 2.5 0 b 1 / o c -

o 2 S A-b-w R-c-f 1.5 H-f-c

0.5

3.5 2 - m H . s

B 3 M R 0 0

0 2.5 c 0 1 / r O - f 2 C A-b-w R-c-f 1.5 H-f-c

0.5

0.35 2 - m H . s

B 0.3 M R 0 0

0 0.25 0

1 d

/ c a - u 0.2 H-f-c N R-c-f 0.15 A-b-w

0.1

0.05

3 2 - m H . s B 2.5 M R 0 0 0

0 e 1 2 / o c - p

S H-f-c 1.5 R-c-f A-b-w 1

0.5

1.8 2 - m H .

s 1.6 B M R

0 1.4 0 0 0 1 1.2 f / u p - t

A 1 A-b-w R-c-f 0.8 H-f-c 0.6

0.4

0.2

0 Provinces HE-B L-N HU-B HU-N Y-N G-S